Advanced Well Testing handbook by G. Pedaci
mob: +39 347 622 47 63
1
INDEX WELL TESTING 1.0 INTRODUCTION
3
2.0 WELL TESTING PRINCIPLES
X
3.0 FIELD DATA ANC QUALITY CONTROL
X
4.0 DIFFERENT TYPES OF WELL TEST
X
5.0 GENERAL FLOW DIFFUSITY EQUATION IN POROUS SYSTEMS
X
6.0 FLOW CONDITIONS
X
7.0 GAS TESTS
X
8.0 WELL DELIVERABILITY IN BOTH OIL & GAS WELLS
X
9.0 DRAWDOWN AND BUILD-UP TESTS
X
10.0INTRODUCTION TO TYPE CURVE AND PRESSURE DERIVATIVE APPROACH
X
11.0EARLY TIME MODELS
X
12.0MIDDLE TIME MODELS
X
13.0LATE TIME SCHEDULE
X
14.0WELL TEST EQUIPMENT
X
15.0DOWN HOLE GAUGE
X
16.0WELL TEST DESIGN AND COSTS
X
17.0FLUID SAMPLING
X
18.0TEST IN AGGRESSIVE ENVIRONMENTS
X
19.0BIBLIOGRAPHY
X
2
WELL TESTING 1.0 INTRODUCTION
3
2.0 WELL TESTING PRINCIPLES 2.1 Main Targets To Be Achieved
As soon as the drilling of a well is finished, if the well results to be hydrocarbons bearing, in order to evaluate the production capacity, most notably pressure and rate, of the well it is necessary to perform a well test in the mineralized mineralized layers. The primary targets of a well test are: Determine the nature of the produced hydrocarbons and its rates.
The new reservoir has been penetrated with a well . The primary interest of all persons concerned, for commercial reasons,
is to evaluate the kind of fluid
produced ( oil or gas ?) and at which rate. Fluid sampling is
important important in order to perform PVT
(Pressure-Volume-
Temperature) analysis. analysis. The aim is at collect collect in oil and gas samples, whether whether down-hole or at the surface, in the exact ratio of gas /oil. The PVT parameters obtained in the laboratory are used directly in the well test analysis, in the recovery calculations calculations and in the design of surface facilities. Initial reservoir pressure (Pi) and temperature (T)
The determination of the initial reservoir pressure (Pi) is of paramount importance. It is standard practice in a well to run the RFT prior to setting the final production casing. This RFT provides the exact pressure of the reservoir at its depth. This method is preferable to the pressure build-up method which requires an extrapolation, quite subjective. The RFT pressure of the reservoir has to be reported to the datum by using the pressure gradient of the reservoir fluid. The temperature is very important for the gas field, since determines the formation volume factor (Bgi) of the gas originally in place.
4
2.2 Reservoir Fluid Flow Analysis
Fluid samples will allow a measurement of basic fluid properties such as composition, formation volume volume factor (Bo and Bg), gas-oil ratio (GOR) & viscosity (µo). Fluid samples will also allow measurement of fluid contaminants such as H 2S, CO2, asphaltenes, wax, mercury, etc. Representative measurements can often only be made at the well site due to degradation of samples over time. Fluid samples from different different zones will determine determine fluid variations variations with depth and in particular can be used to define fluid contacts. 2.3 Reservoir Parameters Determination Evaluation of the formation characteristics: Effective permeability (K ) and flow capacity (Kh)
The capacity Kh Kh (permeability (permeability by formation thickness) thickness) of the formation can be calculated by the build-up interpretation. interpretation. By knowing the h from the log analysis it is possible then to evaluate the permeability permeability K. The value of this K is the average effective permeability to oil or gas in the presence of the irreducible water saturation. The well test permeability should be compared with the average, absolute permeability permeability determined from the cores analysis. analysis. The effective permeability is always smaller then the absolute permeability. permeability. Damage around the wellbore by means of the skin effect (S)
Evaluation of the skin factor has great importance in either appraisal and development wells. Well become damaged either for the human activities while drilling (mud, cement, perforations, etc) or for the movements of formation solids (sand, paraffin, chemical reaction, etc). Whatever the reason for the damage, the first step in preparing a remedial job is the calculation of the magnitude of the skin factor, S. S is a dimensionless number representing the degree of formation damage caused by the mud invasion during drilling. The mud particles obstruct the
5
porosity of the formation so declining is permeability. The formation will produce less because of the mud damage. 2.4 Productitvity Index Definition
Definition of the Productivity Index (PI) of the oil well actual and ideal and the Gas Flow Equation in a gas well is carried out by means of well testing. The PI is the ratio of the oil production rate per unit of pressure drawdown. The Gas flow equation gives the gas rate for the squared pressure drawdown. 2.5 Well Efficiency Control
Control the efficiency of a well completion operation and a stimulation operation. If well tests are conducted both prior to and after those operations the value of the skin factor, S, will sanction the efficiency of those operations. If S diminishes after an operation it means that
operation has been conducted positively
otherwise (S increases) negatively. 2.6 Reservoir Geometry
For the reservoir geometry the evaluation of the presence of faults, contacts, magnitude of the reservoir, radius of drainage, by means of a well testing is important. A long and expensive flow test is required to locate boundaries (faults). If the static pressure after a certain flowing time stabilizes to a value minor to the pi value it is an indication of small reservoir. If the static pressure indicates a trend towards the pi the reservoir is consistent in magnitude. The depletion of the reservoir can be detected if the reservoir is small (the initial reservoir static pressure declines after a prolongated production) or big (the initial reservoir static pressure does not decline after a prolongated production) 2.7 The Interpretation of Well Testing
The interpretation of the well testing can give also the following:
correlation among the producing intervals of different wells;
definition of the driving mechanism of the reservoir;
6
planning the future programme of drilling for the field development;
estimate the surface facilities for the full field development.
The importance of well testing never must be underestimated. The accuracy of a calculation of the hydrocarbon reserves depends a lot on the results of the well tests. A well testing performed in good way will give the possibility to :
compute the Original Hydrocarbons in Place with good precision;
understand the future well behaviour for rates and pressure;
plan the field development in better way;
define the well completion ( tubing size, gravel packing, etc).
In order to get all the expected results, as outlined above, all the data achieved with a well testing must be taken under close examination. For a well testing one must consider:
all the rates of oil, gas and water;
all the bottom hole pressures either during the flowing period (draw-down pressure) or during the shut-in period (build-up).
temperature at the bottom hole;
all pressures and temperatures at well head and at the separator.
The most common and practical method of testing wells is the pressure build-up test for which a well is produced at a constant rate q (stb/d) for a flowing time t (hours), after which is closed-in for a pressure build-up. During the flowing period, the pressure recorded at the bottom hole is denominated p wf (psia – well flowing pressure) and during the subsequent buildup pws (psia- well static pressure) which is measured in the shut- in time Δt (hours). See figure 2.1.
7
pw (psia)
pwf, flowing presure Pws, well static pressure
Δt
t
time (hours)
q q (stb/d) t t
Δt
time (hours)
Fig. 2.1: Rate, q, and pressure profile during a well testing
8
3.0 FIELD DATA AND QUALITY CONTROL
Well testing represents a major source of data to engineers and geologists investigating the potential economic viability of hydrocarbon accumulations. However, well tests are expensive and should only be performed if the information required: a) improves the value of the project by more than the cost of the test and b)
cannot be adequately acquired more cheaply by an alternative method.
The planning stage is important in determining the value of the well test, defining clear objectives for the test, selecting the test type, specifying the equipment required and the procedures to be followed, and indicating what actions should be taken on the rig site if the observed response differs from that anticipated. 3.1 Parameters Definition Time
It must be recorded during a well test all the time of each event : when an event starts and when finishes in terms of date and hours and minute. Duration
An event must be declared in its duration in time. Event
An event must be reported for what it is. For example:
pre-flow, short period of well flowing before the real test in order to control the connection of all the equipments;
initial reservoir pressure, p i, to evaluate the original reservoir pressure before any flowing;
clean-up, event to clean the well from mud or completion fluid, in order to allow the formation fluid to reach the surface as clean as possible;
first flowing, the well is put on production with a certain choke in order to flow with a certain rate, q 1, more or less stabilized;
9
first build-up, the well after the rate of fluid stabilized in closed in order to
allow the reservoir pressure to come back at the original value of p i;
second flowing, the well is put on production with a second choke in order to flow with another rate, q2, more or less stabilized;
second build-up, the well after the rate q 2 of fluid stabilized in closed in order to allow the reservoir pressure to come back at the original value of p i;
acid job, in the well has been injected acid in order to enter into the formation and allow a clearing or a dissolution of the mud particles which obstruct the flow of the reservoir fluid;
etc.
Choke
The choke diameter has to be reported, usually in inches. An inch is divided in 64 parts. So a choke of 32/64” indicates a choke of ½ inch(1/2”).
The choke is put on the surface after the well head and determine the flow rate of the well. The choke has the particularity to stabilize the flow provided that the pressure upstream the choke is bigger than twice the pressure downstream the choke. Rate
The rate of the fluid produced, oil or gas, is of paramount importance since this indicates the capacity of the reservoir to produce. In case of oil the rate must be measured in stock tank condition (60°F and 1 psia) and usually in barrel per day. The unit of the oil rate is so stb/d. Being the time of the flowing usually inferior to the 24 hours of one day, the volume of oil produced in a certain time must be recalculated on daily basis. In case of gas the rate unit usually is the thousands of standard cubic feet in a day, M scf/d. Being the standard condition: 60°F and 1 psia. Gas Oil Ratio (GOR)
10
During the well test the ratio of the volume of gas over the volume of oil in a certain period must be measured. Both volumes must be measured at standard condition and stock tank condition. The unit of the GOR is usually standard cubic feet over stock tank barrel, scf/stb. Well Head Pressure (WHP)
The well head pressure during the test must be taken regularly, specially at the start-up of each event during the same and at the end of the event. This pressure is upstream the choke. The well head pressure should be taken specially when it is stabile. The unit of the WHP is generally the psi, as measured at the manometer, so psi gauge (psig). Bottom Hole Pressure (BHP)
The bottom hole pressure is taken with down hole equipment such Amerada or SRO. The electronic SRO allows the transmission of the pressure value at the surface. The BHP is important to indicate the p i initial reservoir pressure, and the value during the flowing periods. The most important is the recording of the bottom pressure during the build-up. The interpretation of the build-up trend will indicate a lot information on the kind of reservoir. The unit of the BHP is generally the psi absolute , as measured at the down hole pressure gauge plus 14.7 psi (psia). The BHP has to be declared always at which depth has been taken. This is relevant in case the reservoir pressure has to be reported to a datum. Usually the datum is taken in the middle of the reservoir thickness or next to the relevant contact oil/water or gas/water. The datum is a depth from the sea level. Bottom Hole Temperature (BHT)
11
This parameter is very important for a gas field since the Gas Originally in Place (GOIP) depends from the Bgi ( initial gas formation volume factor), which is function of the pressure and temperature. However the temperature is useful for the PVT analysis of both gas and oil. The unit of the BHT is usually the Fahrenheit (°F). Others
If during the well test a pressure gradient profile in the well is taken by means of wireline operations must be reported. This is useful to better determine the nature of the fluid produced in the wellbore. If the pressure gradient indicates a density of about 0.8 Kg/lt the fluid in the wellbore is oil; if it is 1.0 Kg/lt is water and if it is less than 0.07 kg(lt it is gas.
Density of the Oil (API gravity).
Gas composition (hydrocarbons; CO 2; H2S, N2; etc) and associated parameters (z, molecular weight, specific gravity, etc).
Separator of gas/oil/water parameters : pressure and temperature.
Stimulation job: acidification, fracturation etc
3.2 Test Objectives Setting
Example of data available: three wells, B1, B2, and B3 have been drilled into sand formation, two have tested oil and the third B3 logged only water bearing reservoir. The two oil bearing wells have been completed awaiting a tieback to a nearby production facility.
12
Fig. 3.1: Top
sands map, indicating discovery and possible northern block
accumulation A 3-D seismic section is available across the area, and shows that there is potential for an additional accumulation in a northern block, which is the target of appraisal well B-4. The proposed well B-4 includes the objectives to core the well once there are hydrocarbon shows in the mud returns. Coring will continue until the hydrocarbon bearing interval is fully cored. A full open hole logging suite will be run, including RFT pressure and fluid sampling. The drilling proposal also requires an outline production test proposal to ensure that the necessary equipment can be made available in time if the well is found to be hydrocarbon bearing. Before trying to set the well test objectives, it is necessary to be aware of the overall objectives of the field. These are to:
assess the presence and nature of hydrocarbons in the northern block
determine whether the sands (if present and hydrocarbon bearing) are commercially productive ;
13
corroborate the geological and geophysical model of the northern block to assist with future development of the northern block ;
determine whether the northern block is in pressure communication with the main block.
Te outcome of the production test is likely to influence whether further appraisal or development wells are necessary or whether the northern block prospect has to be included in the main field development plan. The result of the test may have very significant impact on the overall value of the project, and decision making theory should be applied to determine the value of the information gained from the test. 3.3 Quality Control Procedures
Before interpreting a test, a fundamental step is the quality control of the raw data ( Q.C.). This operation is complex and important at the same time. In fact, possible anomalies are sometimes well masked and not identifiable; moreover, the choice of parameters which are not representative of the real system leads to conclusions unrelated with the physical reality of the reservoir phenomena. It is fundamental that the control and validation of all the data recorded is carried out on site. This quality control allows for a rapid modification of the operations in order to remedy to possible failures in the surface equipments and in the electronic gauges measurements. Should the Q.C. be carried out at a later time, just before the interpretation, and data found to lack representativeness, the necessity to repeat the test would involve much higher additional costs; moreover, there is the risk that the well performances are no longer the same as those at the time of the original test. Field data must be taken with accuracy, otherwise the validity of the test will be very limited. Great accuracy must be given to the gas rate and oil rate at standard condition and stock tank. For instance if in the oil there is a lot emulsion the rate of the oil can be wrong. If in the gas stream there is a lot of inerts, such as carbon dioxide and nitrogen the gas rate can be wrong.
14
Of paramount importance is the quality of the down-hole pressure gauge for the build-up measurements. A quartz pressure gauge should have a sensitivity of 0.001 psi and an accuracy of +-0.1 % of each reading in psi. Validate Gauges
The quality control on bottom hole parameters has a remarkable impact on the test interpretation. In fact the definition of the most suitable reservoir model starts from the analysis of the log-log plot (diagnostic -plot) which describes the behaviour of the bottom hole pressure and of its derivative. The acquisition of bottom hole data, as far as pressure and temperature are concerned, is made by using high precision electronic gauges located just above the producing formation. As already mentioned, they can be of the two types: Memory or SRO Gauges, the latter allowing for real time readings. It is fundamental that the gauges, independently from the type, are accurately calibrated in laboratory. For this reason, the Service Company must provide the certification and the specifications found in the last calibration. Bottom hole pressure gauges
Quartz pressure gauge are often necessary to assure very precise pressure reading. Specifications of the pressure reading shall be as follows:
Sensitivity = 0.001 psi
Accuracy = + - 1.0 psi or + - 0.01 % of the reading
Surface pressure gauges
The simplest and cheapest location for a pressure gauge is at surface on the wellhead. Such a location can however give problems. Firstly, downhole shut-in cannot be used to avoid wellbore storage. Secondly, the wellbore skin effect is difficult to calculate as the pressure drawdown at the gauge is not only due to the reservoir but also to rate dependent friction losses in the wellbore.
15
Thirdly, phase segregation of fluids in the wellbore may cause massive pressure fluctuations. Wellhead gauges are best used in injection wells where the wellbore fluid is monophasic and incompressible. Even in this case temperature effects can cause problems in interpretation as the cool injection water is heated up throughout the wellbore. Wellbore storage
If the reservoir description near the well bore is important (eg nearby faulting) then early time build-up pressure data is important. After a rate change early time reservoir pressure response can be masked by the compressibility of the fluids within the wellbore. This phenomenon is known as wellbore storage . A means of avoiding this problem, at least for pressure build-ups, is to include a valve and pressure gauge in the test string near the perforations whereby the well is shut in down hole. Consequently there is little volume of wellbore fluid, below the valve, which can influence the reservoir pressure response. This system is probably not warranted when fluids remain monophasic within the wellbore, as liquids have fairly low compressibilities. However, if the reservoir fluid falls below bubble point and gas is present in the wellbore, then wellbore storage is likely to mask a large proportion of the reservoir pressure response. Rate measurement
Production rates are typically measured at surface through a test separator. A standard offshore 3-phase separator will operate at up to 1500 psig and handle 80 MMscf/d gas and 10,000 b/d oil. Gas is metered using an orifice plate, while oil is measured with a positive displacement meter, turbine meter or a vortex meter. It is essential to record the pressure and temperatures at which the measurements are made so that a conversion can be made to express the volumes of fluid at standard conditions and at reservoir conditions where the pressure is measured. The accuracy of such a measurement is approximately +/-10%. Any water content must also be reported to allow the conversion to be made. Oil meters are calibrated offshore at periods during the test using a gauge tank.
16
Pressure analysis techniques require downhole flow rates to calculate reservoir properties. Surface rates therefore need to be converted using an appropriate formation volume factor. Some error may be generated here as the formation volume factor is not always known for the test separator conditions on the rig, giving a possible additional error of around +/-10%. A downhole flow rate can be calculated directly by running a PLT spinner. The accuracy of such a rate is between +/2% and +/-10% depending on the spinner type and logging company. Time measurement
The pressure-time data is recorded by the clock run with the pressure gauge. Traditionally a mechanical clock was used, but this is now invariably an electronic clock. The sequence of events at surface is recorded by the test crew in absolute time, and forms part of the report provided. Fluid sampling
The objective of reservoir fluid sampling is to collect representative samples of the reservoir fluids at the time of sampling. In general terms oil, gas and even water samples are required to properly characterise the formation fluids. Sampling is generally performed in the initial exploration and/or appraisal phase when the fluid is still characterized by its original composition. This is a crucial step for reliably predicting the future reservoir behaviour. Two methods are used for sampling reservoir fluids. They are referred to as “subsurface sampling“ and “surface sampling”. In this
second method, sampling can be made at the separator (most likely) as well as at the wellhead. When sampling exploration wells, subsurface sampling is always associated with surface sampling. As a general procedure, sampling operations can be planned either during the main flow phase or at the end of the test after the final build-up. All the surface/downhole sampling must be properly validated at the wellsite before sending the fluid samples to the labs. In the case of samples inconsistency the operation must be repeated.
17
The choice of the sampling method is influnced by several factors, such as :
lable gas-oil separators equipment.
The key factor to collect a representative reservoir fluid sample is the preliminary conditioning of the well. This consists of producing the well, for a certain time, at a rate which removes all the altered (non representative) fluid from the wellbore. The recommended procedure to reach such a situation, consists of producing the well in a series of “step by step” flow rate reduction. A stabilized gas -oil ratio
(GOR) should be achieved and measured at each step. The well is considered to be sufficiently conditioned when further rate reductions have no effect on the GOR which remains constant over time. Monophasic flow conditions are then basically achieved and sampling can be successfully performed. Special attention must be dedicated when sampling oil reservoirs (light - volatile oil) if the saturation pressure (or dew point pressure for gas condensate) is closed to the initial static pressure. During the sampling phase the following parameters should be stabilized and properly monitored:
Fluid flow rates (Qoil, Qgas, Qwater ),
Bottom Sediment & Water (BSW),
Gas Oil Ratio (GOR),
Wellhead pressure and temperature,
Separator pressure and temperature.
In addition, the main physical fluid properties, such as oil and average ga s gravity as well as the presence of CO 2/H2S, should be carefully evaluated.
18
As a general procedure, all the surface/downhole samples collected during the production test must be properly validated at the wellsite before they are sent to the labs. In the case of samples inconsistency, the operation must be repeated. 3.4 Well Testing Workflow
Workflow field data :The following figures 2 and 3 give an idea of all data to be recorded during a well test workflow:
19
Date
Time
Duration
Evants
Name
Choke
06/12/98
10h40 10h45
00h05 01h45
Q1 BU1
32/64
12h30 13h50
00h20 01h40
15h30
01h30
10h30 17h30
01h00 04h20
07/12/98
21h50 12h50
15h00 10h40
08/12/98
23h30 13h15
13h45 01h00
14h15
01h45
16h00
02h00
18h00
18h00
12h00
04h00
16h00
14h00
10/12/98 12/12/98
06h00 18h00
60h00 11h00
13/12/98
05h00
04h00
09h00
04h00
Preflow Int. pross. Wash Cleanup Cleanup Ship Cleanup SRO Surf sampl. BHS Main acid Cleanup Flush lines Main flow Main flow Main flow Main BU PLTshut PLTopen PLTopen
09/12/98
Inj-1 Q2
Rate b/d
GOR Set/bbl
WHP Psi.g
BHP Psi.a 8519.6 ft
BHP °F 8519.6 ft
51
3058 3596.3
257 257
569 531
3623.6 3526
253.8 266
265
3504
265
194
1020
3532
266.5
421
82.7 1147
3595 3573
263.5 267.2
438
1 -7800
2165
3860
245
3700
816
3537
269.8
0 -1224 Adjust. 48/64
3600
BU2 Q3
32/64
0 2630
BU3 Q4
20/64
0 1020
552
660
Grd PLT
CUMUL
Q5 Inj-2
08/64
Q6
40/64
BU4
0
Q7
40/04
3700
600
953
3537
269.8
2676
Q8
32/64
2700
600
1096
3551
269.8
395
Q9
40/64
3700
600
964
3534
270
2078
3601 3601
261.8
BU5
0 40/64 adj 28/64
4300
831
3542
269.9
0.73
645
2120
1107
3564
269
0.75
317 7104
Tab. 3.1 : Well testing Workflow
20
Phase
Time
WHP p sig
WHP °F
P.sep psig
T.sep °C
Cleanup 32/64
20h30
1017
81.4
159
77.2
Q all Pd(60”F ) 2628
21h00
1017
83
130.9
77.9
2632
05/12/98 7h25 21h5 Pte4h25
21h30 21h40
1020 1020
03.2 83
130.9 130.8
78.4 77.5
2638 2626
Girf. sampl 20/64
18h00
1138
72
174
98.2
1061
20h00
1144
73
176
108
07/12/98 2h00 23h3 Tal 11h3
21h15 22h30
1144 1147
73 74
176 177
23h00
1147
72
32/64 08/12/98
13h00 14h00
1097 1097
2h00 16h0
15h00
GOR Ct/bb
°API std
II2S PPM
Gas gr. air-1
Cumulative
45
9000
1.068
Total-4h25
45
9000
1.07
31.7 31.7
45 45
9000 9000
1.07 1.07
Cumul in 4h: 421.1 bbl Av rate 12681 bpd
864
32
39
11000
1.084
1008
685
31.8
40
11000
1.092
109 105
1027 1035
667 689
32.7 32.4
40 40
11000 11000
1.092 1.092
176
107
1018
674
32.4
40
11000
1.092
94.3 95.3
148.7 148.6
82 81.9
2714 2714
606 614
32.4 32.4
38 38
9000 9000
1098
95.7
149.8
30.1
2718
62D
32.5
38
9000
15h30
1096
95.5
151
83.7
2686
628
32.6
38
9000
Tal-04h0
16h00
1096
94.3
152.6
85.5
2706
624
32.6
38
9000
40/54 89/12/98 8h00 12h0 0910/12/98 6h00 06h0
21h00 23h45
911.1 923.8
87.2 89.1
160.2 160.5
74.6 76.5
3645 3704
610 600
32.4 32.2
38 38
8500 9000
10h00
950.3
103
180.7
82.5
3705
810
32.2
38
9000
1.071
19h00
956
105.6
163
88.7
3696
624
32.4
39
10000
1.072
23h00
959
103.4
161.2
93.5
3701
613
32.4
39
10000
1.075
02h00 05h00
951.6 964.5
113.1 115.1
165 168
102.5 109.1
3685 3682
627 634
32.2 32
38 39
9000 9000
1.074 1.074
Total – 32h
31.7
CO2 %
Total11h30 Cumul in 10h15: 438.5bbl Av rate: 1027bpd
Total-4h00 Cumul in 3h30: 395bbl Ev rate: 2708bpd 1.082
Tab. 3.2: Summary of Production Sequence 3.5 Defining the Well Test Procedure
The selection of the test type clearly depends upon the objectives of the test. Given the objectives of the test of well B-4 in our example of paragraph 3.2, the most appropriate test type will be a pressure drawdown and build-up, with fluid sampling being part of the test procedure. In general, single rate are used to measure reservoir properties. For a pressure drawdown test this means flowing at a single stable rate for a period of time. However, it can be difficult to maintain a constant rate due to fluctuations through the wellbore and surface equipment.
21
Total-h00 Cumuli 1/h30 2876.5bbl Av rate: 367bpd
The rate during a build-up period is obviously zero, but the analysis is simpler if the preceding drawdown has been carried out at a single constant rate. Build-up periods are generally considered more useful for analysis if the preceding flow rate was constant. Multi-rate test are typically used to measure rate dependent properties such as some skin effects and wellbore effects, and are more common in gas wells where skin due to turbulent flow around the wellbore is a function of the flow rate. A basic well test sequence is shown below indicating some of the different requirements from each part of the test. This test does include a multi-rate test, which would be more common in gas wells than oil wells:
Fig. 3.2: Well test sequence for an oil bearing formation
22
Fig. 3.3: Well test sequence for a gas bearing formation Clean-up
Clean-up is suggested to stress the well with different increasing chokes in order to remove non representative fluids (i.e., drilling and completion fluids). It is important to underline that a proper clean-up phase is essential for a consistent well test interpretation. The duration of the clean-up can be variable depending on the well response. In general the cleaning phase will be terminated when the main wellhead parameters (pressure and rates) are stabilised for at least 3-4 hours. The final BSW should not exceed 5%. Any evidence of sand and/or fines production must be monitored. In addition, all the physical parameters of the produced fluids such as Ph, salinity, density, gas SG, etc. must be acquired. First build-up
To measure initial reservoir pressure & temperature, restore pressure equilibrium before starting main test. The duration of the first build up should be the same .
Main flow- drawdown pressure
23
In the case of oil bearing formations a flow after flow sequence consisting of two isochronal increasing flow rates is recommended. In general each step should last 8 to 12 hours. In the case of gas bearing formations a flow after flow sequence of isochronal increasing rates is suggested. A minimum of two flow rates is necessary to estimate the turbulence factor and the flow equation. However, three flow rates are highly recommended. Each step should last 8 hours It is suggested that the maximum flow rate does not exceed the greater flow rate achieved during the clean-up phase. Final build-up
Build-up pressure analysis is used to interpret the surrounding reservoir properties (permeability, boundaries, heterogeneity) and the connection efficiency of the well to the reservoir (skin factors).
Final reservoir pressure may be observed to check for reservoir depletion.
The duration of the main build-up should be 1.5 – 2 times the duration of the main flow. Remarks
Choke sizes and testing time should be adjusted according to the well behaviour. Once the open hole logs are available and a “quick look” interpretation has been
made, an office-based operations meeting is usually called between the subsurface and operations teams to decide on the exact test procedure. Flow And Shut-In Periods Durations
The length of flow and shut-in periods are a compromise between the quantity of information required and the expense of performing the test. Longer flow & shutin periods will provide information on the reservoir more accurate. The time taken to first observe a reservoir heterogeneity at distance r from the wellbore is given for drawdown tests as:
24
T 1191.4
c t r 2
k
The time, T, is in hours all the other units are the American practical oil field units. The estimate of permeability in m D of above will need to be taken from core measurements. The open hole log interpretation estimates a permeability based on an empirical porosity -permeability relationship from the region. For a build-up, calculating the time taken is more complex depending in part on the length of the preceding flow period. Early time data is often dominated by wellbore storage effects, which make interpretation very difficult if not impossible. Consequently flow & shut-in periods should be of sufficient length to pass this period. At the other extreme, in reservoir limit testing, it may be necessary to flow for weeks or months to generate a measurable depletion of pressure. As a rule of thumb, a 50 psi depletion is significant and sufficient to estimate the connected volume. From an initial estimate of the connected volume, material balance calculations can be made to determine the produced volume required to create such a pressure drop. Well test interpretation techniques depend mostly on establishing transient flow and derivatives on type curves indicate when this flow regime commences for a given reservoir type (i.e. homogeneous, fractured). This method can be used to estimate the minimum time required for the flow and build-up periods. In general the time taken to observe all the required reservoir properties surrounding a well is best modelled using the design feature in a computerized well test package. An expected reservoir model should be constructed in liaison with the field geologist and used within the well test package to anticipate the required test duration. Although it is impossible to give a unique time for the periods, typical drawdown and build-up periods are between 6-12 hours and 12-24 hours respectively. Flow Rate
The size of flow rate has little bearing on the mathematics of well test analysis. The rate should however be sufficient to maintain stable flow.
25
A wellbore hydraulics package should be used to design tubing sizes and minimum flow rates to give an acceptable flow regime within the wellbore. Slugging should be avoided if possible. The maximum possible rate from the well is not necessary for the well test analysis, but sometimes in exploration wells there is a requirement to establish this maximum potential, especially if it is to be used as part of the information provided to a potential purchaser of the block, or in equity discussions. Location Of Measurements for Pressure, Rate and Type of Fluids
Time, rate and pressure
are the key measurements required for well test
analysis, and this data set is often referred to as the TRP data. It is essential to specify in the test proposal the frequency and location of:
pressure measurements
flow rate measurements
fluid samples.
The following schematic shows the typical points for monitoring these parameters.
26
Fig. 3.4: Typical locations for pressure and rate monitoring, and fluid sampling The exact set-up will depend upon the type of location (eg land, floater or production platform) but the main components will remain the same. The down hole pressure gauge can record pressure and time data downhole and can display this information in real time at surface using surface read out (SRO) if required.
27
The data header provides ports for monitoring flowing tubing head pressure (FTHP), temperature (FTHT), taking flow line samples, monitoring sand production, and performing chemical injection. The choke manifold controls fluid flow, and is used to establish stable flow conditions and to shut the well in. A heat exchanger may be required to prevent hydrate formation (gas testing) or to allow viscous oil to flow at surface conditions. The test separator (typically a three phase horizontal design) not only separates the three phases (oil, water, gas) but also measures the flow rate of each stream using flowmeters on each of the outlet lines. It is important to record the separator temperature and pressure to allow the rates measured to be corrected to standard conditions (typically 60°F and 14.7 psia). A test tank may be required to measure liquid flow rates if the FTHP is insufficient to allow the use of the three phase separator, and may be used as a check on the three phase separator measurements. The diverter manifold directs oil and gas to the appropriate burners, depending on the current wind direction. To keep the heat away from the installation, flare booms are used, and oil burners inject compressed air and water through nozzles to create efficient combustion and to cool the flame. Fluid samples can be taken down-hole or at surface.
28
4.0 DIFFERENT TYPES OF WELL TEST
The main objective when drilling an exploration well is to test and evaluate the target formation. There are three types of well test methods available: 1.
Wireline Formation Tester (WFT) Simple test by using wireline tools
2.
Drill Stem Test (DST) Where the drill pipe / tubing in combination with down hole tools is used as a short term test to evaluate the reservoir.
3.
Production Test (PT) Many options of string design are available depending on the requirements of the test and the nature of the fluid.
Testing is an expensive and high risk operation and, therefore, should only be conducted for essential data. The starting premise should be that testing is not required unless it is clearly justified. The second premise is that, if testing is warranted, it should be done in the simplest possible manner, avoiding any operations which entail higher risk, such as running wireline or coil tubing through the testing string. By adopting this position, the Petroleum Engineer should not appear to be negative but work towards obtaining essential data, which the company needs rather than that which is nice to have, in the most cost-effective manner. The test objectives must be agreed by those who will use the results and those who will conduct the test before the test programme is prepared. The Petroleum Engineer should discuss with the geologists and reservoir engineers about the information required and make them aware of the costs and risks involved with each method. They should select the easiest means of obtaining data, such as coring, if possible. Such inter-disciplinary discussions should be formalised by holding a meeting (or meetings) at which these objectives are agreed and fixed. The objectives of an exploration well test are to:
29
y and kh and skin value
ory nalysis
nvestigate formation characteristics
The following table indicates the typical objectives of well testing, using bottom hole pressure surveys, for various type of wells: Exploration well
Appraisal Well
Early Production well
Late Production well
Objective
Are there hydrocarbons in the reservoir?
What types of HC ? Productivity PI?
Understand productivity after stimulation job to check PI improvement.
Type of test
DST or production test
Production test
Completion efficiency. Changes in productivity. Reservoir pressure response to production. Production test
Production test
Tab. 4.1: Typical Objectives of Well Testing Exploration well On the first exploration well, well testing is used to confirm the exploration structure, establish the nature of the produced fluids as well as the initial reservoir pressure and its consistency with the RFT/MDT trend when available. Other common targets are both the evaluation of the main reservoir properties (kh, Skin) and the assessment of the well productivity. In addition, any reservoir heterogeneity as well as the presence of potential boundaries should be investigated.
30
A proper reservoir characterization through testing of an exploration well is crucial for any future action/decision and, for this reason, it is strongly recommended to maximise the value of the information achieved by the testing phase. Appraisal wells The reservoir description can be refined by testing appraisal wells to confirm average properties, productivity, reservoir heterogeneities, and boundaries as well as drive mechanism if detected. In order to identify representative reservoir fluids, surface/bottom samples are collected for PVT laboratory analysis. Production / Development wells On producing wells, periodic tests are scheduled to confirm and/or re-adjust the existing 3D-dynamic reservoir model and to evaluate the need for well treatment (re-perforation, acid stimulation, sand control, fracturing, etc) with the target to maximise the well production life. In addition, interference testing is a quite common methodology to confirm possible communication between existing wells. During the well testing time a quite large volume of reservoir rock can be investigated. As a consequence, the main reservoir parameters, such as permeability, should be considered as average values. 4.1 Wireline Formation Tester (WFT)
It is not the purpose of this manual to discuss extensively the Wireline Formation Tester (WFT) applications and, as a consequence, only some general concepts are here presented. In particular WFT is one of the most used tools in formation evaluation and reservoir studies due to its ability of:
depths.
The pore pressure regime, the fluid mobility as well as the in-situ fluid contacts within the formation are provided by WFT.
31
Due to the very short duration of WFT, generally ranging from tens of seconds up to few minutes, the investigated volume is very limited and, therefore, the major parameters (i.e. fluid mobility) are considered reliable only close to the t ool depth. Information obtained from WFT interpretation is very useful especially in designing a consistent testing programme for a new exploration and/or for appraisal wells. In particular: of the
initial PVT properties; pressure
of
the reservoir. A cross –check between this value and the extrapolated pressure from well testing analysis should always be made;
permeability; z)
can be estimated if spherical flow regime is
clearly detected from WFT analysis. 4.2 Drill Stem Testing Methods
A drill stem test (DST) is a production test in which a full completion string is not run as part of a final well completion, but a temporary test string (usually the drill pipes) is used. This avoids the cost of setting a completion string with a permanent packer. The drill stem battery includes the drill pipe, the bottom hole assembly, the packer and the fluid sample container. A drill stem test (DST) is a procedure for testing the hydrocarbons bearing formation through the drill pipe, so in open hole without casing. Mr Johnston developed in Arkansas the first drill stem tester in 1927. The test is a measurement of pressure behaviour at the drill stem and is a valuable way to obtain important sampling information, when the tool is brought to the surface, on the formation fluid and to establish the probability of commercial production.
32
The basic drill stem test tool consists of a packer, valves or ports that may be opened and closed from the surface, and pressure- temperature recording devices. The packer is set to isolate the zone from the drilling fluid column. A DST is normally used in exploration well in order to know quickly if a promising interval is hydrocarbon bearing. The pressure gauges at the bottom give important information on the initial reservoir pressure, pi. The flowing duration of the DST is very short, few hours, in order to avoid the arrival of e significant quantity of hydrocarbons on surface, where the necessary disposal arrangements have not been taken. A DST is conducted to determine the productivity characteristics of one specific zone. DST analysis can provide data to help evaluate the productivity of the zone, the completion arrangements, extension of formation damage and if there is a requirement for stimulation. This technique was quite common in the past especially for testing new exploration wells. It consisted of using a drill string (drill pipe) controlled by a down hole shut-in valve. This testing methodology is not used anymore. In most cases the testing duration was limited to few hours and, as a consequence, the production period was very short and no hydrocarbons were produced at the surface. The main targets to be achieved were basically the following:
the measurement of the static formation pressure;
the collection of a representative reservoir fluid sample.
The reservoir fluid was recovered by reverse circulation and thus the risk of contamination of hydrocarbon by mud or completion fluid was quite high. The evaluation of the other reservoir properties, such as permeability and skin, could not be very accurate because the interpretation approach was not strictly conventional. This was particularly true when tight reservoirs and/or viscous oil reservoirs were tested and when no flow at the surface was observed. In a conventional DST, flow and shut-in were operated by the down hole shut-in valve located below the drill pipe. The wellhead always remained open to the atmosphere, whether directly or through a flare.
33
If the wellhead, equipped with a pressure gauge, remains closed during the flow phase, the DST becomes a Closed Chamber Test (CCT) for the tested flow period. 4.3 Production Test (PT)
The production test is the most complete well testing procedure since implies also the presence of surface equipments for proper measurements of the rates of oil, gas and water and the separation of these three phases. The PT is usually done with tubing and not with drill pipe. The tubing string should be equal to the one foreseen for the final completion of the well. The PT is done in a cased hole and the formation is perforated with a proper gun. The PT may involve different days of execution and can last even more than a week. The purpose to organize a PT in a well is to sanction in exact way the productivity of the well, completed in the same way as when will go in the permanent production. So the interrelation of the formation with the completion string can also be studied. The possibility to run the Production Test for long time give the chance to study better the reservoir boundary. The disturb of pressure during flowing has time to go far and reach the boundary the reservoir. This can be done because the production on surface of the fluids, oil, gas and water can be properly handled and disposed. Also the build-up period can be longer than in a DST since the draw-down has been long. The PT is normally done for better planning the future development of the field. The PT is much more expensive than a DST. Production tests may be classified as follows:
Periodic production tests have the purpose of determining the relative quantities of oil, gas and water produced under normal producing conditions. They serve as
34
an aid in well and reservoir operation and meeting legal and regulatory requirements. Productivity or deliverability tests are usually performed on initial completion, or recompletion, to determine the capability of the well under various degrees of pressure drawdown. Results may set allowable production, aid in selections of well completion methods, design of artificial lift systems and production facilities. Transient pressure tests require a higher degree of sophistication and are used to determine formation damage or stimulation related to an individual well, or reservoir
parameters
such
as
permeability,
pressure,
volume
and
heterogeneities. Periodic tests Production tests are carried out routinely to physically measure oil, gas and water produced by individual wells under normal producing conditions. From the well and reservoir viewpoint, they provide periodic physical well conditions where unexpected changes such as extraneous water or gas production may highlight well or reservoir problems. Abnormal production declines may also indicate artificial lift problems, sand build-up, scale build-up in perforations, etc. On oil wells, results are reported as oil production rate, gas-oil ratio and water oil ratio as a percentage of water in the total liquid stream. Accuracy in measurement, with careful recording of the conditions is essential. Choke size, tubing pressures, casing pressure, details of artificial lift system operation and all other effects on the well producing capability should be recorded. Potential production problems should be recognised in order that they can be properly handled such as emulsions, security of power fluid or gas lift gas supply, etc. It is important that the well is produced at its normal conditions as flow rate will vary the relative quantities of oil, gas and water. On gas wells, routine are less common as each well normally has individual measuring capability. Gas production is reported as well as condensate and water. Similar to oil wells, the wells must be produced at the normal rates. Productivity or deliverability tests
35
This test is different from the periodic test in that the liquid flow performance can be determined empirically using measured flow rates at varying bottom-hole pressure drawdowns and they do not rely on mathematical descriptions of the flow process. With a limited number of measurements, they permit prediction of what a well could produce at other pressure drawdowns. This is then used to predict the PI and is successfully applied to non-Darcy conditions. They do not permit calculation of formation permeability or the degree of abnormal flow restrictions (formation damage) near the wellbore. They do however include the effects of formation damage; therefore can be used as an indicator of well flow conditions or a basis for simple comparison of completion effectiveness among wells in a particular reservoir. Commonly used deliverability tests for oil wells may be classified as:
-After-Flow
Transient tests Flow from reservoirs are characterised as transient, pseudo-steady state or steady state flow, depending on whether the pressure response initiated by opening the well had reached the drainage area boundary and on the type of boundary. Transient flow occurs when the well is initially opened or has a significant rate change, and is a result of the pressure disturbance moving out towards the outer boundary of the drainage area. During this the production conditions at the wellbore change rapidly and the FBHP pwf, decreases exponentially with time. Most DSTs and many production tests are conducted under transient flow conditions and consequently the observed productivity will often appear greater than that seen in long term production.
36
This means that corrections need to be made to compensate for transient flow behaviour as well as for skin effects. When the flow reaches the outer boundary, flow becomes steady state or pseudo- steady state. If the boundary is a constant pressure boundary, then PR will not alter with time and is termed steady state. However if it is a no-flow boundary, then P will decline purely as a result of depletion and the flow is then termed pseudo-steady state. When the FBHP appears to be constant or declining slowly proportionally with time, the well is stabilised and pseudo-steady state flow equations can be used to predict the long term deliverability of a well. Transient pressure tests are classified as:
Build-up test
Limit test
Interference (Areal, Vertical and Pulse)
Each type presents certain advantages and limitations and factors which are important for reasonable results.
Drawdown test
In the pressure drawdown test, the flowing bottom hole pressure is measured while the well is flowing, is a primary method of measuring productivity index (PI). Establishing a stable rate and a stable flowing pressure may requires a long period. With many rate it is possible to construct the inflow performance relationship (IPR curve), which is the Flowing Bottom Hole Pressure (FBHP) vs the Oil rate, see figure. With this test it is possible to compute the PI, Kh, and skin factor S.
37
PI
production rate (bbl/d) pressure drawdown (psi)
Fig. 4.1: Pressure Drawdown
Fig. 4.2: IPR Curve Build-up test
The pressure build-up test measures the bottom hole pressure response during the shut in period which follows a pressure drawdown. This is useful for measuring reservoir Kh, near well skin S, and final pressure of reservoir, equal to the initial one P i.
38
Fig. 4.3: Pressure Build-up Multi-rate test
The multi-rate test is used to determine rate-dependent properties such as skin, and are common in gas well testing. This is a form of pressure drawdown test with many rates. It is useful to determine if skin factor S is function of rate. Inflow performance curve (IPR) for oil wells, q vs flowing pressure, can be determined with accuracy, in this case the variation of PI with rate can be evaluated too.
Fig. 4.4: Multi-rate Drawdown Limit test
The reservoir limit test is designed to establish the hydrocarbon volume connected to the well. The flow rate is constant and once the limit of the reservoir has been reached the pressure drops linearly with time, indicating that the
39
reservoir is fully bounded. This “semi steady state” response can be used to
estimate the connected volume of fluid.
Fig. 4.5: Reservoir Limit Test - 1
well
fault
Fig. 4.6: Reservoir Limit The limit test is long long in duration since the disturbance of of pressure has to travel travel along the reservoir until
reach the boundary of the reservoir or some
impermeable obstacle in the reservoir reservoir such as as fault, drastic change change (drop) in permeability permeability and porosity, facies variation. Since this disturbance travels during pressure draw-down it is necessary that the well stay in production at a stabilized rate for this reason are necessary surface equipments for proper disposal of the fluid produced, oil, gas and water (if any).
40
The limit test records the return of the pressure disturb in the same well, which has generated the disturb. Actually the waves of the pressure disturb are reflected by the boundary and returns to the flowing well, well, where a pressure pressure gauge is installed. Interference test (areal, vertical, pulse)
The interference test between two wells is is used to estimate the transmissibil transmissibility ity (kh/µ) of the formation in the interval between the wells. A pressure change is created at the active active well by opening up the well, well, and a pressure pressure gauge in the closed-in observation well awaits a pressure response, the arrival time of which can be used to estimate transmissibility. transmissibility.
Fig. 4.7: Interference test It is a test usually done in a field already developed. So are not necessary surface equipments for the disposal of the fluids produced since the development already provide means to convey the production towards the production centre. The interference test is done to analyze if the reservoir has a certain continuity in its areal extension (Areal Interference Test).
Fig. 4.8: Areal Interference test
41
The procedure is to keep only one well in production and all the other wells closed. Among the closed wells is chosen one, practically far from the producing well, on which the pressure is monitor at the bottom hole with a pressure gauge descended with wireline. The active well under production generates a pressure disturb (draw-down) which travel in the reservoir and reaches the observation well. If the observation well records a certain pressure pressure drop after a certain time, this time can be also long of of days and even weeks, the reservoir has continuity between the two wells. Instead if in the observation well will not be recorded any pressure drop , even after a very long long time, the reservoir reservoir is not connected connected between the two wells. wells. Between the two wells there is an heterogeneity: heterogeneity: i.e. a fault or a facies variation. Other interference test is the Vertical Interference Test, as depicted below:
Fig. 4.9: Vertical Interference Test The layer 1 is put on production with q, while the layer 2 is shut in, but a pressure recorder is in front of the layer 2. If this latter gauge record a pressure drop it means that the two layers are in communication. Pulse test
The pulse test test is a version version of the interference interference test, test, but attempts to provide provide enough information to allow the interpreter to eliminate the effects of noise and gauge drift in pressures (to which the interference test is prone) as measured at the observation well. It determine the transmissibility transmissibility (kh/µ). This method is an effective alternative to the conventional interference test.
42
A sequence of relatively short flow (production or injection) and shut-in periods is applied to the active well. The rate and the duration of the each flow are the same. Also the shut in periods have the same duration, but not necessarily the same as the flow periods. Three or four pulses are generally enough to analyse the pressure response at the observation well. This sequence generates a pulsing pressure response at the observation well, which is analyzed in terms of amplitude and time lag. The measured parameters are compared to the theoretical simulated responses and, as a result, the average permeability and other parameters are estimated. Even if they are more difficult to interpret, pulse tests should be preferred because the oscillating response is easier to identify in a noisy reservoir environment environment (field (f ield under production).
Fig. 4.10: Pulse test Injection test
Injection well testing has its application in water injection wells for pressure maintenance as well as in water disposal wells. The main targets of this test are:
infectivity index of the well;
Injection well testing involves the following methods :
43
1.
Step rate test: these tests are specifically made to evaluate the pressure at which fracturing could be induced in the reservoir rock. A series of injection test rates are applied to the well. The rate should be constant during each step; the observed pressure is plotted versus rate. If fracturing conditions have been reached, two different straight lines are present and their intersection defines the fracturing pressure.
2.
Injectivity/falloff test: in this test, a constant flow rate is injected into the well while the downhole pressure is recorded at the sandface. Then the well is shut-in for a final falloff. The interpretation of such a test would be similar to a conventional production test provided that physical properties (viscosity, density, etc.) of the injected fluid and those of the reservoir fluid are compatible. This would be the case when water is injected into an aquifer. As a result, standard well testing objectives can be easily achieved including heterogeneities and/or permeability boundaries if investigated.
However, because the properties of the injected fluid are usually different from those of the actual reservoir fluids, the interpretation of the injection/falloff tests is much more complex than the interpretation of a conventional injection test. Moreover the pressure behaviour during the injection phase is different from the observed one during the falloff. Injection Phase
During the injection period the flooded region increases in time and a “movable front“ exists in the reservoir.
The evaluation of the skin from injection tests is difficult to interpret because the total (or apparent) skin is made of two components: the conventional well skin and the two-phase skin. As a consequence, a proper interpretation of the injection phase can only be performed with advanced tools (i.e numerical simulator) provided that the two-phase relative permeability curves are available. Artificial fractures potentially induced during the injection phase represent another important factor that heavily complicate the interpretation. To avoid fracture induction, it is strongly recommended to i nject fluid into the reservoir in “matrix conditions”.
44
Falloff phase
Due to the different pressure response during injection and falloff, the principle of superposition is, in theory, not applicable. In practice, it has been noticed that, when a Radial Composite model with stationary front is used, no significant error is introduced. As a result, the following main targets can be achieved with the usual approach:
The derivative response describes the change of saturation in the transition zone separating the inner water region and the uncontaminated, outer oil region. However, in practice, due to wellbore storage effects the response of the inner region is generally masked. Therefore, only the permeability of the outer oil region and the total skin can be evaluated.
45
5.0 GENERAL FLOW DIFFUSIVITY EQUATION IN POROUS SYSTEMS 5.1 Effect Of Hydraulic Diffusivity In Reservoir Behaviour
There are
various type of hydraulic flows, linear, radial and spherical, as
depicted below:
Fig. 5.1: Hydraulic Flows We consider the horizontal radial flow of a single phase fluid moving to centre to the wellbore. The assumption to be done are :
the formation is homogeneous and isotropic
the central well is perforated across the entire formation thickness
the pore spare is entirely saturated with the fluid.
Darcy Law
In a porous medium the linear flow rate, Q, of a f luid is proportional to the ΔP applied to the medium to the section A and inversely proportional to the fluid viscosity μ and to the length of the medium L.
The overall constant of proportionality is K, that is the permeability.
46
Where. All unit are in the so called Darcy units :
Q = fluid rate, cm3/sec
K = permeability, Darcy
A = area, cm2
ΔP = differential pressure, Pe – Pw, atm
μ = fluid viscosity, centpoise
L = length, cm
h = thickness, cm
r e = drainage radius, cm
r w = wellbore radius, cm
Ln = natural logaritmic
In case the flow is radial the Darcy formula, with the same above units,
will
change as follows:
47
Fig. 5.2 : Darcy low in case of Radial flow From Darcy law we have : p
q
2 kh
ln(
re rw
and
r
p q . r 2 kh
Te equation of mass continuity for radial flow is the following: Where:
q = fluid rate, cm3/sec
k = permeability, Darcy
ρ = fluid density, gr/cm3
μ = fluid viscosity, centpoise
Φ = medium porosity, dimensionless
r = drainage radius, cm
t = time in sec
The combination of the radial flow equation of Darcy with the mass continuity equation gives the Radial Flow Diffusivity Equation, as follows (in oil field units):
φμc p 1 p r = r r r 0.000264 k t Which is a second order differential equation with the variables: pressure p, the radius r, and the time t. All parameters of the above diffusivity equation are in oil field units, as follows:
k = permeability, mD
r = radius, feet
dp =
μ = fluid viscosity, centpoise
dt
Φ = porosity, dimensionless
differential pressure, psia
= differential time, hours
48
Unfortunately the diffusivity equation is non linear since, μ, c, Φ, k and also ρ are
dependent from the pressure. Because of this complication it is not possible to determine direct analytical solutions for use in well test analysis. But if we assume :
μ, c, Φ, k and also ρ independent from the pressure;
a pressure gradient
a single fluid flow with small and constant compressibility c.
dp/dr
small
The radial diffusivity equation of above can be accepted. The reciprocal of the coefficient on the right hand is k/Φμc is the hydraulic
diffusivity constant, which plays a major role in the in the whole reservoir engineering discipline. In the context of well testing, the higher is k/Φμc, the greater is the depth of the investigation by pressure analysis in the reservoir. The solution of the above equation by means of the Ei(x) function gives for the oil field units the following equation of flowing pressure at the wellbore for an infinite reservoir with a constant production q:
p wf = pi -
kt 162.6 q μ B 3.23 log 2 φμ kh cr w
Where : pwf = flowing pressure in the wellbore at any time t, in psi pi = initial reservoir pressure in psi q = oil rate in s.t. barrel per day B = oil formation volume factor, dimensionless k = formation permeability in milliDarcy (mD) h = formation thickness in feet Φ = formation porosity, dimensionless μ = viscosity of the oil in cent poise (cp)
49
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi r w = wellbore radius in feet
Fig. 5.3: Flowing Pressure at the Reservoir This solution can be represented with an hydraulic circular simulation as depicted below:
Tab. 5.1: Infinite Reservoir - 1 Since this hydraulic circular simulation is equivalent to an infinite reservoir with constant external pressure and constant q rate, with the only slightly difference that the pwf at wellbore in the infinite reservoir diminisnes very gently with time (with log(t)).
50
Tab. 5.2: Infinite Reservoir - 2 The plot of pwf vs time in hours is the following: 3000 Flowing Pressure at the wellbore (Pwf) in psi
2900
2800
2700
2600
2500
2400 1
3
5
7
9
11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 Time in hours
Fig. 5.4 : Pwf vs Time in Hours – Ideal and Infinite Oil Reservoir The same plot with the time in months is the following:
51
3000
2900
2800
i s p , 2700 f w P
2600
2500
2400 0
2
4
6
8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 2 8 3 0 3 2 3 4 3 6 3 8 4 0 4 2 4 4 4 6 4 8 5 0 5 2 5 4 5 6 5 8 6 0 6 2 6 4 6 6 6 8 7 0 7 2
time in months
Fig. 5.5: Pwf vs Time in Month
The same plot with the time in years is the following: 3000
2900
2800
i s p , 2700 f w P
2600
2500
2400 0
2
4
6
8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 2 8 3 0 3 2 3 4 3 6 3 8 4 0 4 2 4 4 4 6 4 8 5 0 5 2 5 4 5 6 5 8 6 0 6 2 6 4 6 6 6 8 7 0 7 2
time in years
Fig. 5.6: Pwf vs Time in Years
52
The three plot indicate the following.
the flowing pressure at the well bore p wf , being the reservoir infinite and producing at all time at constant q rate, never reaches the zero value, even after 72 years;
the value of the p wf diminishes very rapidly at the opening of the well (transient flow);
the value of the pwf continue to gently diminishe , after the transient flow, and enters into the steady state flow, which is proportional to log(t)..
53
6.0 FLOW CONDITIONS 6.1 Well Transient Testing And Analysis
During the initial pressure decline at the wellbore , the pressure recording is totally unaffected by the presence of any faults or boundaries in the reservoir. In this respect the system appears to be infinite in extent. Up to the transient period estimated in t (hours) as follows:
φ μ c r e 2 t 0.00264k
Where :
k = formation permeability in mill Darcy (mD)
Φ = formation porosity
μ = viscosity of the oil in cent poise (cp)
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi
re = reservoir external radius in feet
t = time since the well has been open with the rate q, in hours
Larger the diffusivity constant k/Φμc the sooner discontinuity in the reservoir will
influence the wellbore pressure and the therefore the period of transience will be short. Conversely, in very low permeability reservoirs, the transient phase may extend for months rather than hours. Any pressure disturbance caused in the reservoir, such as opening a well to flow, closing it in or even changing its rate will induce a transient pressure response, identification and isolation of which permits the engineer to apply the simple transient
solution of the diffusivity equation to the pressure –time record to
calculate the permeability and skin factor of the formation under test. The diffusivity equation for the determination of pwf in the transient time is still the same of the infinite reservoir:
54
p wf = pi -
162.6 q μ B kh
In the transient flow both
kt log 3.23 φμcrw 2
the infinite and limited reservoir have the same
pressure behaviour as shown in the following figure:
Fig. 6.1: Limited Reservoir-Pressure drawdown A semi-steady state in a circular bounded reservoir can be reached in time (hours) equal or greater than .
φ μ c r e 2 t 0.00088k
Where :
k = formation permeability in mill Darcy (mD)
Φ = formation porosity
μ = viscosity of the oil in cent poise (cp)
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi
55
re = reservoir external radius in feet
t = time since the well has been open with the rate q, in hours
The periods of transient and steady state should be anticipated when testing an appraisal well. The condition of steady state should hopefully not to be encountered in appraisal well testing for it implies that all the outer boundaries are influencing the pressure in the wellbore and this results in a stable rate of pressure decline throughout the system. If this observed, dp wf / dt = constant, during a well testing it means that the reservoir is small and so the calculated OOIP.
Fig. 6.2: Bounded /limited reservoir In the bounded reservoir , circular finite with a radius re, the hydraulic similitude is the following:
56
Fig. 6.3: Bounded Reservoir Hydraulic Similitude We can notice that pressure at the wellbore continuously decrease and so the rate q, furthermore the pi also decreases. But there is something similar with the infinite reservoir: the same pressure behaviour at the beginning of the flow (transient flow) before the pressure disturbance reach the boundary of the reservoir. 6.2 Steady State Flow Regime
The steady state flow regime happens only in an infinite reservoir and the pwf is proportional to the logarithm of time. This means that P wf declines in infinitesimal way with time since approximately P wf is equal to (P i –m log(t)). Being m a constant of the formation.
57
Fig. 6.4: Comparison: infinite and limited reservoir for pressure drawdown
58
7.0 GAS TESTS 7.1 Concepts of pseudo-pressure function m(p) with respect to P and P approach
The Darcy's law, and the gas equation of state must be combined to develop a differential equation for the flow of gas through porous media.
1. Darcy's law for flow in a porous circular medium is: q=
2πhk δp μ
δr
2.The mass conservation equation is : q ρ = qsc ρsc where : q = gas rate in reservoir qsc = gas rate on surface at the standard condition (sc) : psc=14.7 psi Tsc = 520°R ρ and ρsc = gas density at reservoir and standard conditions μ = gas viscosity
3.
The equation of state for a real gas is : ρ=
pM zRT
The combination of the three above equations and the integrals from wellbore pressure pw to the reservoir pressure ¯p and from the wellbore radius, r w to the drainage radius, r e, give the below equation:
59
The integration of the above equation gives the following result:
kh T pe 2 -pw 2 sc
qsc=
r e μg z T ln r w
In oil field units the above gas rate at standard conditions, qsc, will be :
(a)
qsc=
0.000305 kh pe 2 -pw 2
r e μg z T log r w
Where :
qsc = Mscf /d
k= permeability in m D
h = formation thickness in feet
pe = reservoir pressure, psia
pw = well bore pressure, psia
Tsc = Temperature at standard condition = 520 °R
T = reservoir temperature, °R
r e = drainage radius, ft
rw = well bore radius, ft
z = average compressibility factor, dimensionless
μg = gas viscosity, cP.
The above gas flow rate (a) at sc is proportional to the pseudo-pressure function m(p), defined in 1966 by Al-Hussainy: pavg
m p =2
pw
p dp z μg
where :
60
m(p) = pseudo-pressure of real gas
z = gas compressibility factor
μg = gas viscosity
p = pressure
The gas flow rate at sc (equation (a)) in terms of pseudo-pressure function m(p) becomes the following:
qsc =
(b)
0.000305 kh m(p)e-m(p)w T log r r e
w
Where :
qsc = Mscf/d
k= permeability in mD
h = formation thickness in feet
T = reservoir temperature, °R
r e = drainage radius, ft
r w = well bore radius, ft
m(p)e = pseudo-pressure function in MM psi 2/cp for the external reservoir
pressure.
m(p)w = pseudo-pressure function in MM psi 2/cp for the wellbore pressure.
For a typical natural gas at cons tant temperature we have that the product μz is constant for values of wellbore pressure less then 2000 psia and this product μz
is linear with wellbore pressure for values of pressure greater than 3000 psia. See graph below:
61
Fig. 7.1: Isothermal variation of µ Z vs pressure in linear scale This implies, being μz a costant, that the equation (b), for value of pressure less
than 2000 psia, is exactly equal to the equation (a). pw
m p =2
p0
p dp = (pw2 –po2) / (μ z) z μg
For value greater than 3000 ps ia the equation (b) has that p/μz is almost constant since the product of μz is linear with the pressure.
Therefore, when the pressure is higher than 3000 psia the pseudo-pressure m(p) becomes : pw
m p =2
po
pw
p p dp = 2 dp = (pw - po) x constant z μg μz p
o
Thus:
Below 2000 psia, either the p 2 approach or the m(p) approach can be used (usually engineers use the p 2 because is more easy).
Above 3000 psia, the m(p) function can be substituted with the linear p.
62
High-pressure gas wells behaves like a slightly compressible fluid, and therefore the pressure data, can be used directly in linear mode i.e. without being squared.
Between 2000 psi and 3000 psia, no simplification is available, and the m(p) function must be used.
The two limits of validity of the simplified forms ( p<2000 psia and p>3000 psia) are approximate, and depend upon the gas composition and temperature. When the m(p) function can be estimated with a computer program, the pseudopressure m(p) is preferably used for the complete range of test pressure. However, the practical engineers prefers to see the analysis in real pressure or even in pressure squared, rather than m(p) values. Example
Given the following data: p (psia) 400 800 1200 1600 2000 2400 2800 2970 3500 4000
µ
(cp) 0,0143 0,0149 0,0150 0,0151 0,0155 0,0160 0,0180 0,0190 0,0232 0,0246
z dimensionless 0,9733 0,9503 0,9319 0,9189 0,9100 0,9113 0,9169 0,919 0,9445 0,9647
It is possible to compute the m(p) function as follows:
63
p (psia) 400 800 1200 1600 2000 2400 2800 2970 3500 4000
µ
(cp) 0,0143 0,0149 0,0150 0,0151 0,0155 0,0160 0,0180 0,0190 0,0232 0,0246
z dimensionless 0,9733 0,9503 0,9319 0,9189 0,9100 0,9113 0,9169 0,919 0,9445 0,9647
µz
cp 0,014 0,014 0,014 0,014 0,014 0,015 0,017 0,017 0,022 0,024
p/(µz) psi/cp 28.665 56.378 85.846 115.312 141.794 164.600 169.654 170.093 160.065 168.689
m(p) (MM psi2/cp) 11,47 45,48 102,4 182,8 285,7 408,2 541,9 599,7 774,7 939,1
Tab. 7.1: m(p) pseudo pressure The above m(p) is actually calculated as follows.
Tab. 7.3: m(p) calculation From the table above it can be noted that the product μz is constant up to 2000
psia, so the m(p) can be computed directly with the p 2 approach up to the pressure of 2000 psia. After the 2000 psia the function m(p) must be used. After the 3000 psi the linear p could be used, by considering the p/μz product constant.
64
8.0 WELL DELIVERABILITY IN BOTH OIL & GAS WELLS 8.1 Productivity Index In Oil Well (Pi )
The definition of the Productivity Index (PI) of an oil well is the ratio between the oil rate, q, at stock tank condition (60°F and 14.7 psi) and the delta pressure at the bottom hole in front the perforation. The ΔP is the difference between the static pressure of the well, p ws and the
actual flowing pressure, p wf , in the wellbore for generating the oil rate q.
PI
q q p pws - pwf
Where :
pwf = flowing pressure in the wellbore stabilized , in psi
pws = static reservoir pressure, measured in the wellbore with a pressure buildup in psi
q = oil rate in s.t. barrel per day
PI = stb/psi
Example
A well produces 1000 stb/d with a p ws of 3000 psi and a flowing pressure at the bottom hole of 2500 psi, then the PI is equal to 4.0 stb/d/psi. But if the pwf was equal to 2800 psi the PI is equal to 10.0 stb/d/psi. Greater is the PI better is the well for the oil production rate. For example a well with PI of 10.0 can produce oil rate equal to: q = PI x Δp = 10.0 x Δp if the Δp, induced by the operator to the bottom hole, is equal to 200 psi the well will produce 2000 stb/dand, if the operator induces a Δp of 500 psi the well will
produce 5000 stb/d. To determine the PI of a well at least two drawdown with relatives pwf pressure and oil rate are necessary. Plus two buildup to establish properly the static well pressure.
65
8.2 Productivity Index (Pi) In Oil Well (Pseudo-Steady Conditions)
The PI equation of the previous paragraph is in transient conditions, since is determined during well test of short duration. The PI can be determined in semi-steady conditions or pseudo-steady state if all the parameters q, p wf , etc have been taken after a long time, when the semisteady state has been reached. This will happen after the time in hours equal to: t
φ μ c r e 2 0.00088k
The estimation of this time implies the knowledge of the external radius of the reservoir, re (feet). 8.3 The Inflow Performance Relation (IPR Curve) –Well Deliverability
The PI is constant for pwf not too far from the p ws, afterwards tends to decline because the pressure in the well goes under the bubble point value with gas liberation. The PI method assumes that all future production rate changes will be in the same proportion to the pressure drawdown as was the test case. This may not always be true, especially in a solution gas drive reservoir producing below the bubble point pressure. The bubble point pressure is the condition of temperature and pressure where free gas first comes out of solution in the oil. When the pressure in the formation drops below the bubble point pressure, gas is released in the reservoir and the resulting two phase flow of gas and oil around the wellbore can cause a reduction in the well productivity. Typical IPR curve, well deliverability, that is the plot of various pwf vs the oil rate Q, is as follows:
66
Fig. 8.2: Inflow performance relationship (IPR) From above curve it can be noted the non linearity of the IPR for pwf low and below the bubble point pressure. PI will no longer be constant and will start to deviate (decrease) after bubble point pressure. The closer the reservoir pressure will be to Pb, the earlier the deviation from the straight line will occur and production decrease will be consistent; this is due to the larger quantity of gas flowing, together with oil, in the formation, and to the turbulence effect. Oil viscosity will consequently decrease, while loosing its associated gas, further turbulence will occur; then Inflow Performance curve will be more like a curve than a straight line. In the following figure there are two kind of reservoirs , reservoir 1 has the bubble point pressure far from the initial pressure of the reservoir, while reservoir 2 has the pb very close to the pi. The IPR curve is good for reservoir 1 while is ba d for reservoir 2.
67
Fig. 8.3: IPR Comparison 8.4 Vogel Formula
Vogel has developed a formula useful for drawing the IPR curve from he bubble point pressure to zero pressure value. This because the portion of the curve from the static pressure to the bubble point curve is linear. The Vogel formula, to draw the IPR curve from pb to p atmospheric is the following: 2
q 1.0 0.2 pwf pwf qmax pb pb Where :
pwf = flowing pressure in the wellbore in psi
pb = bubble point pressure in the weelbore, psi
q = oil rate in s.t. barrel per day
qmax = represents the maximum oil rate obtainable from the well in a theoretical
case where the formation could be brought directly to the
atmosphere
from its depth and put on production at the atmospheric
pressure. Note: the above formula is valid for a reservoir with a static pressure already next to the bubble point pressure , s o pws = pb. 68
Example
A well has been tested with q = 65 bpd and pwf = 1500 psi, we know from PVT analysis that the bubble point pressure is equal to p b = 2000 psi. Furthermore the static pressure of the well is next already to the bubble point pressure. From the Vogel formula is possible to determine the q max. Infact : 65/qmax = 0.40 from which q max = 162.5 bpd. Knowing qmax from the Vogel formula we can determine all the other value of q f or each pwf and so drawing the IPR curve from p b to downward, which is not linear, as follows:
2500 pwf, psi 2000
1500
1000
500 q, bpd 0 160,0
140,0
120,0
100,0
80,0
60,0
40,0
20,0
0,0
Fig. 8.4: Non linear IPR curve 8.5 Oil Rate For A Damaged Well (Skin Factor, S)
In case the well is damaged, so the skin factor , S, is greater than 0, the computation of the oil rate can be done with the following adapted Darcy formula per radial flow in a porous media: Q=
7.08 x 10-3 k h Ps-Pwf μ B ln (re/rw 0.75 S
Where :
69
Q = oil rate in s.t. barrel per day
Pwf = flowing pressure in the wellbore stabilized , psi
Ps = static reservoir pressure, psi
K = permeability, m D
μ = fluid viscosity, cent poise
h = thickness, feet
r e = drainage radius, feet
r w = wellbore radius, feet
ln = natural logarithmic
B= oil formation volume factor, dimensionless
S = skin factor , dimensionless
8.6 Examples Of IPR Curves
1-IPR curve changes for different skin factors, S. The IPR, as the damage increases (S = 0; S= 10; S= 50), worsens in quality:
Fig. 8.5: IPR Curve – Skin factor S variation
70
2-IPR curve changes for different permeability k. The IPR, as the permeability decreases (k = 10 mD; k= 5 mD; k= 1 mD), worsens in quality:
Fig. 8.6: IPR Curve – Permeability, k, variation 3-IPR curve changes for different thickness
h. The IPR, as the thickness
decreases (h = 100 ft; k= 10 ft; k= 1 ft) worsens in quality:
Fig. 8.7: IPR Curve – Formation thickness, h, variation
71
4-IPR curve changes during the depletion of the reservoir (i.e.Ps decreases from 300 atm to 200 and then to 100 with the time life) The IPR, as the depletion evolves , worsens in quality:
Fig. 8.8: IPR Curve – Progressive Depletion with time 5-IPR curve changes before and after a stimulation job. The stimulation job improves the IPR curve:
72
Fig. 8.9: Well Head flowing pressure, WFTHP Vs flow rate; IPR before and after stimulation 8.7 Tubing Outflow Curves
The pressure losses in the tubing from the bottom hole up to the surface are given by the following equation: FBHP – FWHP = (Hydrostatic pressure exerted by the fluid) + ( Tubing friction losses) FBHP - FWHP = Lρ +
f ρ Q2 L D5
Where:
FBHP = Flowing Bottom Hole Pressure
FWHP = Flowing Well Head Pressure
L = tubing length
D = tubing diameter
f = tubing friction coefficient
ρ= fluid density
The following figure clarifies the tubing configuration. The FWHP, usually is almost defined since the surface facilities must work at certain pressure (being this pressure the downstream pressure after the choke). For a stabilized oil rate it is necessary that the upstream pressure to the choke be twice or more than downstream pressure. Therefore the FWHP is quite a known and well established parameter.
73
Fig. 8.10: Surface equipments configuration Typical outflow performance curves for a tubing string with various diameter D (1”, 2.5” and 4”) and for a given fixed FWHP is the following:
Fig. 8.11: Outflow Performance Curves for tubing with various diameters, D 8.8 Operating Point Of The Two Systems, Reservoir And Tubing
The two systems have the following inflow and outflow curves:
74
Fig. 8.12: Inflow and Outflow Curves The operating point, also equilibrium point, will be given by the intersection of the two curves, as follows:
Fig. 8.13: Operation Point 8.9 Flow Equation In A Gas Well (Transient Conditions)
The rigorous approach to evaluate the deliverability for gas wells relies on the pseudo-pressure function m(p). p
m p =2
p
z μ dp
po
Where :
75
m(p) = pseudo-pressure of real gas (concept introduced by Al-Hussainy
z = gas compressibility factor
μ = gas viscosity
p = pressure
Then the flow equation for a gas well with the pseudo-pressure approach (the m(p) function) should be : Δm(p) = Aq + Bq2
However for practical purpose s, the difference of the square pressure Δ(p 2) is generally preferred (being : Δ(p 2) = pws2 - pwf 2 ), the flow equation for the gas
flow in a porous media is proportional to the gas rate q for the laminar flow and to the square of q 2 for the turbulent flow, as follows. Gas Well deliverability equation in transient state: Δ(p2) = Aq +Bq2 But the laminar flow in the formation not damaged is: A‟q = m x n and the turbulent flow in the formation not damaged is B‟ q 2.
Then we should add the laminar flow in the formation damaged (A s q) , and the turbulent flow in the formation damaged (B s q2). Therefore the above equation becomes in broad sense: Δ(p2) = (A‟ + A s)q +(B‟+Bs)q2
Where (see also next chapter to understand the symbols):
A‟ q= squared pressure drop of the linear flow in the formation not
damaged; A‟ = m x n/q.
Asq =
squared pressure drop of the linear flow in the formation
damaged (skin effect); but A s is unknown.
B‟ q2= squared pressure drop of the turbulent flow in the formation not
damaged.
76
4.7 x10
-10
ND m K
in American units
B‟ =
Bs q2 = squared pressure drop of the turbulent flow in the formation
h r w q
damaged by skin effect ; but Bs is unknown. To determine A‟ , A s, B‟ , Bs is necessary to have a full well testing interpretation,
but in case we have two flow rates of gas and two flowing pressures as shown in figure below:
Fig. 8.14: : Gas deliverability well test with two rates and two buildups where:
q1 and q2 are different ;
Kh of the two buildup are equal i.e. m 1/q1 = m2/q2.
the two rates and the two buildup must be done sequentially or with time interval very narrow;
the two pseudo times of flowing to be equal to 1 = to2; i.e. Gp1/q1 = Gp2/q2
Therefore we have two flow equations with two rates q 1 and q2, and relative pressures constituting a system with two unknowns A and B:
we can determine A and B , which must be always positive, from the system of two equations as follows:
77
Example
Tab. 8.1: Gas flow equation calculation The gas flow equation Δ(p 2) = Aq +Bq 2 represented in log-log scale is linear as
follows:
Fig. 8.15: Diagram of Gas Flow Equation in log-log scale 7.10 Absolute Open Flow (AOF)
The gas well if is left to flow at the atmospheric pressure, will produce the maximum flow rate, since the counter pressure is the minimum vailable in nature.
78
This condition does not exist since the flow equation of the gas has been determined at the bottom hole, but ideally if the reservoir could be brought on surface the maximum gas rate in direct flow into the atmosphere will be reached. From the formula, Δ(p2) = Aq +Bq2 , the AOF can be derived by putting in the : Δ(p2) = pws2 - pwf 2
pwf = atmospheric pressure, i.e. equal to 14.7 psi. But it is not easy to solve the equation Δ(p 2) = Aq +Bq 2 for q .
To find the AOQ it is easy to use the log-log diagram and find for Δ(p2) = pws2 - 14.72
in the linear curve the corresponding value of q, which is the AOF. The above example gives Δ(p 2) = 50002 - 14.72 = 24,999,784 psi2, by entering
with this value in the diagram we find AOF =4.24 MScf/d
4.24
Fig. 8.16: AOF determination 8.11 Flow Equation In A Gas Well (Pseudo-Steady Conditions)
79
The flow equation in transient conditions : Δ(p 2) = Aq +Bq2 can be applied in
pseudo-steady state if all the parameters q, pwf, etc have been taken after a long time, when the semi-stedy state or pseudo steadi state has been reached. This will appen after the time in hours equal to: 2
t
φ μ c r e
0.00088k
The estimation of this time implies the knowledge of the external radius of the reservoir, re (feet). 7.12 Gas Back Pressure Curve (Mainly For Well Head Flow)
Another gas flow equation beyond the Δ(p 2) = Aq +Bq 2 named the empirical
relationship by Rowlins-Schellardt is the Back-Pressure Equation: qgas = C (Δp2) n Where:
q = gas rate
C = Constant to be determined by production test
n= flow coefficient depending on flow type, laminar, intermediate and turbulent
Δ(p2) = pws2 - pwf 2
This equation can be used mainly for the well head pressure. This equation to be determined needs a Back Pressure Test with two gas rates and relative well head pressures (static and flowing).
n , flow coefficient, n = 1 : is an indication of laminar flow
n =< 0.5 : is an indication of turbulent flow.
0.5
The above equation in log-log scale is a straight line : logq = logC + n log Δp 2.
80
With two flow rates q 1 and q2 and two flowing pressures p wf1 and pwf2, by having always the same static pressure pws , that can be done even in the transient time it is possible to determine the C and n values of the back pressure equation. C
n
q1 ( (P 2 )1 ) n
(log q 2 - log q 1) (log ( p 2 ) 2 - log ( p 2 )1 )
Example
Tab. 8.2: Back pressure equation calculation From the above example it is clear that we can compute the AOF at the bottom well the same way of the equation Δ(p 2) = Aq +Bq2. The flow in the above example, being n =0.89, is intermediate between laminar and turbulent.
81
9.0 DRAWDONW AND BUILD-UP TESTS 9.1 Reservoir Properties Estimate Geometrical data
r w = well radius It is the radius of the bit that has drilled the producing formation. This is valid both in the case of cased hole and open hole wells. hp = flowing interval The flowing interval shall coincide with the lenght of the perforated interval in cased hole or with the formation thickness in open-hole wells. If several perforated intervals are open to production, the distance between the top of the first perforated interval and the bottom of the last one is considered. However, if direct well information is available, the actual flowing thickness shall be used. Lh = horizontal length In horizontal wells it defines the horizontal length drilled in the producing formation. The whole length of the perforated portion shall be used for cased hole wells. If several perforated intervals are open to production, the distance measured between the first perforated interval and the last one will be considered. The whole open hole length will be used in the case of open hole wells. D = distance between the wells Distance between the producer and the observation well. It is used only in the case of interference tests. Petrophysical data
When defining petrophysical parameters, it is important to stress that, though evaluated at the well, they are considered as average reservoir values. The hypothesis of homogeneous formation might be in contrast with the actual reservoir characteristics. Only numerical models allow the discretisation of the
82
reservoir volume into blocks, to which specific values of petrophysical parameters and saturations can be assigned. Only large scale reservoir heterogeneities can be taken into account in both analytical and 2-D numerical models. Φt = porosity (%)
Total (communicating) porosity of the producing formation. In the case of fractured carbonate formations, total porosity is defined as the sum of primary (or matrix Øm) and secondary (or fracture Øf) porosity: Øt = Øm + Øf Matrix porosity is generally higher than the fracture one. Fracture porosity is generally lower than 1.0% of the total porous volume. Depending on the type of rock, degree of fracturation and fracture spacing the most probable Ø f values are as follows:
es : 0.01 - 0.5% - 1.5%
When the total porosity is greater than 5-6%, as a first approximation, it can be assumed: Øt = Øm When the test investigates several layers with different petrophysical characteristics (multilayers) or zones inside the same producing formation, it is possible to define an “average” porosity value calculated as follows: Øm = (Ø1 h1 + Ø2 h2 +… +Øn hn) / hTot The porosity value is evaluated on the basis of the compared analysis of logs and cores. hn = net-pay Net thickness of the producing formation. It is defined as the sum of the single layer thickness actually contributing to production. It is evaluated starting from the total reservoir gross thickness „„h GROSS”
considered as the difference between the bottom and the top of the structure.
83
The average net thickness is calculated by multiplying the total vertical thickness by the net/gross ratio, which is evaluated by the compared analysis of logs and cores. However, when interpreting a test, the total net-pay (orthogonal with respect to the dip of the formation) must be used. When present, the dynamic response of a PLT (Production logging tool) represents a further information to characterize the actual producing pay. This parameter enables the user to evaluate the effective permeability of the fluid considered when the kh of the formation is known. Fluid saturations data
Sw = water saturation (%) Water saturation of the producing formation. It is evaluated by log analysis. When the test involves several layers with different petrophysical characteristics (multilayers with or without cross-flow) or zones inside the same producing formation, it is possible to define an average water saturation value calculated as follows: Swm = (Sw1 Ø1 h1 + Sw2 Ø2 h2 + … + Swn Øn hn) / hTot Øm where hTot and Øm represent the net total thickness and the average formation porosity. So = oil saturation (% ) Oil saturation of the producing formation. It is evaluated by log analysis. As before, in the case of multilayers formations, an average oil saturation calculated is calculated: Som = (So1 Ø1 h1 + So2 Ø2 h2 + … + Son Øn hn) / hTott Øm Sg = gas saturation (%) Gas saturation of the producing formation. This case is similar to oil saturation. i.e.: Sgm = (Sg1 Ø1 h1 + Sg2 Ø2 h2 + … + Sgn Øn hn) / hTott Øm When the producing formation is characterized by the co-existence of the three phases, the following equation must be satisfied:
84
Sg + So + Sw = 100% Compressibilities data
Fluid saturations are used to define the total compressibility of the system, i.e.: Ct = Co So + Cg Sg+ Cw Sw+ Cf being Co, Cg, Cw the oil, gas and water compressibility, respectively. These values are evaluated by PVT analyses or by suitable empirical literature correlations. Cf represents the actual formation compressibility. For reservoirs with primary or matrix porosity: Cf = Cfm where Cfm is the pore volume compressibility by lab measurements. If no experimental data are available, interpretative softwares directly calculate the pore compressibility as a function of the matrix porosity (Hall diagram). In fractured reservoirs with secondary porosity, the formation compressibility takes into account the contribution of the matrix, fractures and possible communicating vuggy systems (Karst phenomena): Cf = Cfm + Φfrac Cfrac +Φv Cv where:
fm :
frac :
matrix pore compressibility fracture compressibility in the range 1.0 – 6.0 x 10-4 (kg/cm2)-1
Cv : vug compressibility
frac :
v :
secondary porosity (fractures)
vuggy porosity (vugs) comprised in the range 0.1 – 3.0 %.
In this case, the total value of the rock compressibility shall be evaluated with respect to its components and shall be manually introduced into the interpretative softwares. All the compressibility values are referred to the average static conditions of reservoir
85
pressure and temperature. As a first approximation it can be assumed C v = 3 Cfm. PVT data
The evaluation of PVT parameters is always made based on reservoir bottomhole pressure and temperature: Reservoir pressure It defines the average static pressure of the reservoir during the test. When a remarkable depletion occurs during the test, average PVT parameters are calculated on the basis of an average pressure value, comprised between the initial and the final value. Reservoir temperature It defines the average reservoir static temperature. It is assumed that reservoir phenomena are isothermal. As a consequence, the reservoir temperature is always considered constant. In the case of gas wells, the average static value shall be considered. The highest value measured during the test (usually recorded during the drawdown phases) shall be used for oil wells. Reference depth The above defined average static values, as well as all the other pressure and temperature values recorded during the test, are referred to the depth at which the gauge is located. On the other side, all the corresponding PVT parameters must always be referred to the pressure evaluated at the depth of the middle point of the producing interval. (Reference depth). The PVT parameters should be corrected also for the temperature of the middle point depth. However, the PVT corrections due to temperature variations are negligible in most practical cases. The gauge is generally located close to the producing zone and hence the variation of the PVT parameters is absolutely negligible. When the producing
86
formation has a remarkable thickness (order of magnitude of many hundreds of meters) there can be significant differences in the PVT. This is particularly evident in the case of oil bearing formations where it is also possible to encounter a vertical distribution of the oil physical properties due to gravitational effects. Correction of pressure "at well level" If the actual vertical distribution of the fluids inside the well is not known, the correction of the reservoir average static pressure at a conventional depth can generate remarkable errors. The error is directly proportional to the distance between the measurement point and the reference depth at which the static pressure and the corresponding PVT parameters are evaluated. In these cases, it is recommend to verify the fluid nature and the possible fluid distribution in the wellbore and to carry out some static profiles, generally at the end of the final pressure build-up, with numerous steps along the producing formation. The knowledge of the parameters (P,T) measured during the static profiles allows the evaluation of the distribution, nature and density of the different phases in the well. Where no information on the real well fluid distribution is available, the pressure at which the PVT parameters are referred can be calculated according to different hypotheses (two simple situations are generally considered): 1.
Single phase oil: it is assumed that the fluid is single phase oil from the measurement point to the reference depth. Based on the average oil gradient γo (kg/cm2/m) and on the difference Δh (m) between the
reference and the gauge depths the reference pressure P r is calculated as follows: Pr (kg/cm2) = Pgauge + γo Δh 2.
Single phase gas: assuming that the fluid is dry gas and based on the average gas gradient the reference pressure Pr is calculated as follows: Pr (kg/cm2) = Pgauge + γg Δh
87
A significant control on field data and particularly on the nature of the produced fluids can be useful to support the adopted hypothesis. For example, the presence of water, also in minimum percentages, found during the flowing phases can be (but not necessarily) a sign of the presence of liquid levels in the well. On the contrary, dry flowing phases do not a priori exclude the presence of liquid levels. However, the assumptions made to calculate the reference pressure and the value of the average gradient of the fluid must be expressly mentioned. Use of PVT reports data (laboratory analysis)
When laboratory fluid analysis are available, the required parameters to be used in the interpretation can be directly obtained from PVT reports. These parameters are specific and representative of the reservoir fluids at different pressure and temperature conditions. For this reason, they replace any empirical correlation. The PVT parameters to be used during an interpretation are those obtained in laboratory tests and particularly: Oil volume factor, B o Considered as the ratio between the measured oil volume at static reservoir conditions at the time of the test and the corresponding oil volume measured at Stock Tank conditions (P = 1.033 kg/cm 2 , T = 288 °K ). The oil volume factors at Stock Tank conditions are obtained in laboratory by flashing a sample at the bubble pressure through Test Separator (Flash Liberation) . Between the different tests of pressure separator, the oil volume factor must be selected on the basis of the field separator data, possibly interpolating laboratory data. The correct Boi value at the reservoir pressure must be calculated through the equation: Boi (corrected) = Bob (flash) x [ B oi ( diff ) / Bob (diff ) ] where:
88
oi (diff
): differential volume factor at reservoir P i and T
ob (diff
ob (
): differential volume factor at P b and reservoir T
flash ): flash volume factor at P b ( Separator Test )
The “Differential Liberation” is representative of the phenomena which take place
in the reservoir at a constant temperature and is characterised by gas development and production due to the progressive pressure depletion. In contrast, “Flash Liberation” is more consistent with the production process. In
fact the oil (and the possible free gas) is produced from the reservoir at surface with a gradual decrease of both pressure and temperature. Then the oil is sent to one more separators in sequence (high and low pressure) and it is then measured, completely deposited, in storage tanks at atmospheric pressure. In each separation stage, gas is separated from oil and measured. Oil viscosity, μo Oil viscosity at static reservoir conditions at the time of the test. The viscosity value obtained by a transformation of the type “Differential Liberation" at reservoir temperature shall be used in the interpretation. In the case of saturated oil (P i = Pbubble), the oil viscosity at the saturation pressure μob shall be used.
Oil compressibility, C o Oil compressibility at static reservoir conditions at the time of the test. Laboratory analyses measure the average value of oil compressibility from the initial static pressure (Coi) to the saturation pressure (C ob) at reservoir temperature. For pressures lower than P bubble, taking into account that there are generally no Co laboratory measurements, the oil compressibility can be preliminarily evaluated according to the following equation: Co (P) = Cob x [ Pb/P ] x [ Rs (P)/Rsb]0,5 where:
o(P):
ob:
oil compressibility at the pressure P < P bubble
oil compressibility at the pressure P bubble
89
s(P):
sb:
laboratory value at the pressure P ("Composite" transformation)
laboratory value at the pressure P bubble
The value obtained by the equation is to be considered as a first approximation since the Specific gravity variations of the gas separating from oil due to reductions of pressure
dp
= Pbubble - P and oil density in API degrees are not
taken into account. The empirical correlation used for the evaluation of the oil compressibility below the bubble point is the following: Co = 6,8257 x 10 -6 x Rs
0.5002
x P-1 x T 0.76606 x S.G -0.35505 x API 0.3613
where the parameters are expressed in the Oil Field System, except for temperature T expressed in F degrees. Use of empirical data from correlations
Field Data Due to the lack of PVT reports, the reservoir fluid parameters are obtained from empirical correlations provided by the literature. In any case, field data evaluated at the surface during the test and presented in the test reports of the Service Companies are used. The reports provide:
for gas wells: the average value of the Specific Gravity ( air = 1.0 ) of the gas mixture at Standard Conditions (P = 1.0 atm, T = 288 °K).
for oil wells: the oil density expressed in API degrees, the Specific Gravity of the gas separating from oil and the GOR gas/oil ratio from test at Standard Conditions. They also include the separation conditions at different stages. The oil Shrinkage coefficient for converting the measurements from separator conditions to ST conditions, is also presented.
In both cases, the field evaluation of the Specific Gas Gravity is referred to the total gaseous mixture, i.e. the measurement takes specifically into account the presence of H 2S, CO2, N2. Gas correlations
90
Starting from field data, both the software programmes, Interpret/2003 and Saphir, directly calculate all the PVT parameters necessary for the test analysis (z factor, Bg volume factor, Cg compressibility) based on the static reservoir pressure and temperature and the Specific Gravity by using their internal correlations. In the case of laboratory analysis the gas composition shall be directly introduced. The Lee- Gonzales - Eakin correlation must be used for the calculation of gas viscosity. Condensate correlations In gas condensate reservoirs, the fluid at reservoir pressure and temperature conditions is in the gaseous phase. At constant reservoir temperature and after pressure depletion, the dew point can be reached and the liquid phase can precipitate (retrograde condensation phenomenon). During production gas is the dominant phase. The liquid phase which condensates at the surface is gathered and measured in the test separator. In the gas condensate test, the GOR has a wide range (from 5000 to 10000 Scf/STb) while the Specific Gravity of the condensate is generally greater than 45 API degrees. The PVT calculation for tests in condensate gas reservoirs with retrograde condensation is made by using the average specific gravity SGaverage at initial reservoir conditions. From a conceptual point of view, this value is completely different from the one measured at the surface since its composition varies after the separation of the liquid component. The average specific gravity is defined by the relationship ("Applied Petroleum Reservoir Engineering" - Craft and Hawkins): SGaverage = (GOR x SGgas + 4584 x SGoil) / (GOR + 132800 x SG oil / Moil) where:
GOR : test gas-oil ratio, Scf for Stb of condensate
SGgas: specific - gravity of the surface gas (air = 1.0)
91
SGoil: specific - gravity of the surface condensate (water = 1.0)
Moil : molecular weight of the condensate
where:
SGoil= 141.5 / (131.5 + API cond)
Moil = 6084 / (API cond - 5.9)
For the evaluation of the PVT parameters (z, B g, μ g), the interpretative softwares consider the SG average value calculated using the internal correlation of LeeGonzales-Eakin. Oil correlations The correlation internal to the softwares, Interpret/2003 and Saphir, are deemed insufficient to cover all the different types of reservoir oils. The most reliable correlations with the name of the Author, for each physical property, as a function of the oil API gravity are given in the following table: Type of
API range
oils Super
Bubble
Solution
Volume
Compressibilit
pressure, PB
gas, RS
Factor, BoB
y Co
< 10
Standing
Standing
Glaso
Vasquez-Beggs
10
Standing
Vasquez-
Vasquez-
Vasquez-Beggs
Beggs
Beggs
Kartoatmodjo
Kartoamodjo
Kartoamodjo
Vasquez-Beggs
Glaso
Kartoamodjo
Kartoamodjo
Labedi
Heavy Heavy
Medium
22.3
Light
>33.1
Tab. 9.1: Oil Properties Correlations In the above table the evaluation of the oil volume factor Bob is referred to the reservoir
temperature
and
bubble
pressure.
Since
both
programmes
,Interpret/2003 and Saphir, requires the Bo,volume factor at the reservoir average static pressure at the time of the test, the following relationship shall be used ( P ≥ Pb ):
92
Bo = Bob x e-Co (Pi - Pb) where Co represents the oil compressibility at the average reservoir static conditions. The oil viscosities under the different conditions have been evaluated through the following correlations:
93
Type
of
API range
Dead
oils
oil
viscosity, µod
Saturated
Unsaturated
oil
viscosity, µo
oil
viscosity, µob
< 10
Egbogah-Jack
Kartoatmodjo
Labedi
Heavy
10
Egbogah-Jack
Kartoatmodjo
Kartoatmodjo
Medium
22.3
Kartoatmodjo
Kartoatmodjo
Labedi
Light
>33.1
Egbogah-Jack
Beggs-
Labedi
Super Heavy
Robinson
Tab. 9.2: Viscosity Correlations where:
μod: viscosity of the dead oil at the atmospheric pressure and at reservoir
temperature;
μob : viscosity of the saturated oil at bubble pressure and at reservoir
temperature;
μo: viscosity of the undersaturated oil at reservoir pressure and
temperature. The application software "Predator ver. 1.0" ( APSERIIN - 9/94 ) allows the evaluation of the PVT parameters by automatically selecting the option which always gives the most reliable correlation. The parameters obtained are manually introduced into the interpretative software independently from its internal correlations. In particular, the program requires: μo, Bo, Co
at the average reservoir static conditions at the time of the test. The
Bo value is calculated based on the B ob via equation ,B o = Bob x e-Co (Pi - Pb) . Note: The correlations developed by Gorini-Palma, which give both the B o and μo curves as a function of pressure and temperature, can be used as an alternative.
94
These correlations shall be introduced into a program already existing in MODI ("Mbal") or developed in an ad hoc application. Two phase/three phase flow In addition to the single phase flow condition, there is the possibility of analysing tests with multiphase flow both at the well (i.e.: flowing pressures lower than Pb with gas phase development) and in the formation (i.e. gas development, as mobile phase, in the reservoir where S g > Sgcritical). In all cases the PVT calculation imposes the selection of the dominant flow phase. It is important to underline that the test interpretation will have to be reviewed afterwards when the PVT data obtained through laboratory analysis are available. In fact, as previously stated, the empirical correlations cannot replace laboratory parameters. An additional further weakness of any correlation is due to the fact that correlations are based on field measurements affected by uncertainty. In particular, the GOR evaluation, from which the bubble pressure value Pb and the oil volume value Bo depend, can be difficult due to the instability of the gas phase during the flowing periods. The average error on the measurement is at least 5% in the case when the surface equipments are perfectly calibrated. The phenomenon is remarkable in the case of saturated or very volatile oils due to the high gas rates. This results in large errors in the determination of the other PVT parameters and, as a consequence, in the evaluation of the results of the interpretation. Other data from PLT, RFT, MDT, LOGS and Cores
The well test interpretation must be integrated with other information provided by measurements taken before and/or after the production test. This data allow a complete validation of the well test results. The main additional information are obtained by the following tools:
the real fow profile vs depth at different rates and the presence of possible cross-flow under shut n conditions. The PLT is strongly recommended when testing heterogeneous reservoirs (multi-layer or multi-zone formations, etc...)
95
RFT (Repeat Formation Test) and MDT (Modular Formation Dinamic Tester): they are used to collect reservoir fluid samples and to measure reservoir pressure at different depth along the well profile.
sedimentology, stratigrafy etc. are useful for a correct
interpretation
and must be taken into account when available.
CORES: all the information obtained from lab analyses on cores must be integrated with other available information for a complete rock characterization.
9.2 Main Drawdown (Oil Well)
A pressure drawdown test is simply a series of bottom-hole pressure measurements made during a period of flow at constant production rate. Usually the well is closed prior to the flow test for a period of time sufficient to allow the pressure to stabilize throughout the formation, i.e., to reach static pressure. The transient flow in the wellbore area has the following limit time, t, in hours: t
r w
2
0.000264k
Where :
k = formation permeability in milli Darcy (mD)
Φ = formation porosity
μ = viscosity of the oil in cent poise (cp)
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi
re = reservoir external radius in feet
t = time since the well has been open with the rate q and has completed the transient flow , in hours
96
Fig. 9.1: Flowing times In the transient flow time the diffusivity equation for an infinite reservoir is still valid for a finite reservoir; the diffusivity equation is the following : p wf = pi -
162.6 q μ B kh
kt log φμcr
w
2
- 3.23
Skin Effect, S
In many cases it has been found that the permeability of the formation near the wellbore is reduced as a result of drilling and completion practices. Invasion by drilling fluids, dispersion of clays, presence of a mud cake and of cement , presence of a high gas saturation around the wellbore, partial well penetration, and plugging of perforations are some of the factors responsible for this reduction in permeability. Since this effect is close to the well, transient pressure caused by the damage is of small duration and may be neglected. Hence the effect of a reduction in permeability near the well can be taken into account a san additional pr essure drop Δp proportional to the rate of production q. The zone of reduced permeability, where the original fluid has beeen contaminated by the mud, has been called “skin” and the resulting effect “skin effect”.
97
Fig. 9.2: Formation damaged by mud filtration
Fig. 9.3:Damaged area near wellbore shows permeability Ks inferior of formation permeability K. The skin effect shall be understood as a further pressure drop in the formation near the wellbore. According to Van Everdingen such a pressure drop can be estimated with the following equation (oil field unit):
p skin 141.4
q B S kh
Where :
Δpskin = pressure drop near the wellbore for damaged caused while
drilling, psi
k = formation permeability in mD 98
h = formation thickness in feet
q = oil rate in s.t. barrel per day during the flowing time
B = oil formation volume factor, dimensionless
μ = viscosity of the oil in cent poise (cp)
S = skin factor, dimensionless
Being the slope of the flow equation: m=
162.6 q μ B kh
The above Δpskin equation can become the following in oil field units:
p skin 0.87 m S q B S in the diffusivity equation : p 141.4 kh
By introducing the
p
wf
=
p-
162.6 q μ B
i
kh
skin
kt log φμcr
p wf = pi -
w
2
- 3.23 we have
162.6 q μ B kh
the diffusivity equation with skin effect:
kt log 3.23 + 0.87 S φμcr w 2
Where :
pwf = flowing pressure in the wellbore at any time t, in psi
pi = initial reservoir pressure in psi
q = oil rate in s.t. barrel per day
B = oil formation volume factor, dimensionless
k = formation permeability in milliDarcy (mD)
h = formation thickness in feet
Φ = formation porosity, dimensionless
μ = viscosity of the oil in cent poise (cp)
99
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi
r w = wellbore radius in feet
t = time since the well has been open with the rate q, in hours
S= skin factor, dimensionless
Drawdown test interpretation (oil well)
The diffusivity equation for an infinite reservoir still valid for a finite reservoir (provided to be utilized in the transient time) with the skin effect is re-written as follows: p
wf
=
pi
k 162.6 q μ B log t + log kh φμcr
2
w
- 3.23 + 0.87 S
This equation in semilog scale, p wf vs log t, represents a straight line of the type: pwf = p1hr + m log t with intercept p 1hr and slope m. The intercept, p 1hr , will happen for log t = 0, which means time, t = 1 hour. Knowing p1hr at t = 1 hr and then log t=0 from the above formula it is possible to determine the Skin factor, S, as follows:
S=1.15
pi-p1hr k -log m φμcr 2 w
+3.23
Where :
S = Skin factor, dimensionless
pwf = flowing pressure in the wellbore at any time t, in psi
p1hr = flowing pressure at 1 hour time, psi; or the intercept for log t = 0
pi = initial reservoir pressure in psi
k = formation permeability in milliDarcy (mD)
Φ = formation porosity, dimensionless
100
μ = viscosity of the oil in cent poise (cp)
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi
r w = wellbore radius in feet
And from the slope m given by : m=
162.6 q μ B kh
It is possible to determine the Capacity of the formation, k h, as follows:
k h=
162.6 q μ B = mD x ft m
Where :
k = formation permeability in milliDarcy (mD)
h = formation thickness in feet
q = oil rate in s.t. barrel per day
B = oil formation volume factor, dimensionless
μ = viscosity of the oil in cent poise (cp)
m = slope in the semilog graph , p wf vs log t
The radius of investigation, ri during the transient time can be estimated with the following formula:
ri =
kt 848 μ c
Where :
Ri = radius of investigation at time, t, in feet
k = formation permeability in milliDarcy (mD)
h = formation thickness in feet
101
μ = viscosity of the oil in cent poise (cp)
Φ = formation porosity, dimensionless
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi
Example
A drawdown test has been done on a well. The stabilized oil rate and other parameters were the following:
Tab. 9.3: Stabilized Oil Rate and Other Parameters The bottom hole pressure, p wf recordings while the well was flowing at oil rate q were as per the following graph: 3500
3000
2500
2000
i s p , f w p
1500
1000
500
ime, hrs 0 0
10
20
30
40
50
60
70
80
90
Fig. 9.4: Drawdown Test We woul like to determine, kh, h, S and ri of this well. 102
Procedure of solutions: 1.
Plot the above data in semilog scale, i.e pwf vs log t:
2.
Draw the straight line (red) and compute m and pihr for t = 1hr
1.100
pwf
1.050 1.000 950 900
p1hr= 1022 psi
850
m= 142 psi/cycle
800 750 700 650 600 550 500 450 400 350 300 250 200 150 100
logt
50 0 1
10
100
Fig. 9.5: Drawdown interpretation in semilog scale 162.6 q μ B m
3.
Compute k h =
4.
Compute Skin factor, S=1.15
S=
5.
and k :
pi-p1hr k -log m φμcr 2 w 10,1
+3.23
adim
Compute radius on investigation, ri, for the time of the drawdown, almost 80 hours: r i =
kt 848 μ c
9.3 Main Build-Up Test (Horner Method) (Oil Well)
103
Pressure buildup testing is the most familiar transient well-testing technique, which has been used extensively in the petroleum industry. The test is conducted by producing a well at constant rate for some time, shutting the well in (usually at the surface), allowing the pressure to build up in the wellbore, and recording the down-hole pressure in the wellbore as a function of time. The pressure buildup test use the principle of superposition. To develop this principle we shall consider a well which flows at rate q for a fixed time to as follows: p wf = pi -
162.6 q μ B kh
kt log φμcr
o 2
w
- 3.23 +0.87 S
Then the well is closed and we continue we the same equation of above but the time is to + Δt: p ws = pi -
k to + t log 3.23 +0.87 S φμcr
162.6 q μ B kh
2
w
This equation by the principle of superposition (see fig below) must be counter balanced by a negative flow –q from the time to onward (the new shut-in time is called Δt) as follows :
Fig. 9.6: The principle of superposition p ws = pi -
k to + t 162.6 q μ B log kh φμcrw 2
162.6qmB k t S log 3.23 0.87 cr kh
- 3.23 +0.87S
2
w
104
The simplification of the above equation gives the following equation for the buildup test::
p ws =pi +
t 162.6 q μ B log kh tot
Where :
pws = bottom hole well shut in pressure at any time Δt, in psi
pi = initial reservoir pressure, or static reservoir pressure, in psi
q = oil rate before closing the well, in s.t. barrel per day, stb/d
B = oil formation volume factor, dimensionless
k = formation permeability in milliDarcy (mD)
h = formation thickness in feet
μ = viscosity of the oil in cent poise (cp)
Δt = closing time , after to, in hours.
to = time since the well has been open with the rate q and then closed, in hours practically: to
(cumulative well production since the opening of the well) (production rate q, before closing in)
This equation tells us that if we plot the pressure pws observed during a closed-in period vs logarithm of Δt/(t o + Δt) we should obtain a straight line, as follows.
Furthermore we have (see following figure) :
If Δt/(to + Δt) tends to 1 , that is after an infinite time of closure (Δt = ∞ ) log( Δt/(to + Δt)) will be equal to z ero, the extrapolation of the pressure plot
will give us the p*, the static pressure of the reservoir. p* is a check for the depletion of the resrervoir: if p* is less than pi the reservoir has started its depletion.
105
The slope of the pressure curve vs log ( Δt/(to + Δt)) is equal to: m=
162.6 q μ B kh
From wich we derive the Capacity of the formation, k h, as follows:
k h=
162.6 q μ B = mD x ft m
Where :
k = formation permeability in milliDarcy (mD)
h = formation thickness in feet
q = oil rate in s.t. barrel per day during the flowing time
B = oil formation volume factor, dimensionless
μ = viscosity of the oil in cent poise (cp)
m = slope in the semilog graph , p ws vs log( Δt/(to + Δt))
Fig. 9.8: Slope m We can derive the skin effect from the equation of diffusivity while the well is closed: p ws = pwf
162.6 qmB kh
k t 0.87 S log cr - 3.23 2
w
106
If in this equation if we put Δt = 1 hr, we will have (being logΔt =0) :
k S p 1hr - pwf = m log 3.23 0.87 2 cr w
and so the skin factor S:
p1hr-pwf k S=1.15 -log φμcr 2 m w
+3.23
Where :
S = Skin factor, dimensionless
pwf = flowing pressure in the wellbore before closing the well, in psi
p1hr = flowing pressure at 1 hour from closing time, psi; to be taken on the straight line portion of the buildup
k = formation permeability in milliDarcy (mD)
Φ = formation porosity, dimensionless
μ = viscosity of the oil in cent poise (cp)
c = oil compressibility or overall compressibility, c = coSo + cwSw +cf, in 1/psi
rw = wellbore radius in feet
m = slope in the semilog graph , pws vs log( Δt/(to + Δt))
The graph below shows where to take p 1hr :
107
Fig. 9.9: P1hr determination S is the skin effect around the wellbore, is a number without dimension, to be determined by the equation above mentioned. In particular we have if:
S = 0, no skin , no damage around the wellbore, permeability of the skin a ks is equal to the formation permeability k.
S > 0, there is formation damage around the wellbore k s<
S < 0, there is no damage around the wellbore, exceptionally we have i improvement. That usually happen after a stimulation job, such as acidizing, fracturing etc. In this case the permeability of the wellbore area is even better than the permeability of the formation one , k s>k.
S can vary between -5 and 500. Sometimes a better relative index than the skin effect for deciding upon the efficiency with which a well has been drilled and completed is provided by the “Flow Efficiency” (FE), denominated also “Completion Factor” (CF).
The Completion Factor is defined as the ratio of actual Productivity Index, PI, over the PI ideal, calculated as if there were no skin (S=0). Completion Factor, CF = PI actual / PIideal = % Since : PIactual = q/ΔP = q/(p*-pwf ),
where sometimes p* = pi
108
and PIideal = q/(p* - (pwf +Δpskin)) We obtain: Completion Factor , CF
p * - p wf - pskin p * - p wf
%
CF should be around 70%-90% to represent a good well with little damage due to the drilling. If CF is less than 70%, we should improve the flowing around the wellbore with a stimulation job. Example of Pressure Buildup Test Interpretation.
An exploration well has flown for 48hrs with an oil rate of 4600 BPD, afterwards was closed for 38 hours to record the bottom hole pressure buildup. By knowing all the relevant input parameters such as:
Φ = formation porosity, dimensionless
μ = viscosity of the oil in cent poise (cp)
c = oil compressibility or overall compressibility, c = coSo + cwSw +cf, in 1/psi
rw = wellbore radius in feet
B = oil formation volume factor
we must deteminre from the buildup test the followings:
m = slope of the buildup in psi/cycle
k h= formation capacity , mD x ft
k= formation permeability
p* = static pressure of the reservoir, (sometimes equal to pi), psi
if p* is less than pi the reservoir has already started the depletion.
S = skin factor
FE = flow efficiency
109
The following shows the table of the calculation and the plot of pws vs log( Δt/(to+Δt)) :
Tab. 9.4: Pressure Buildup Interpretation 4500
P* =4450 4400 4300 m= 200 psi cycle 4200
P1hr =4090
4100
BHP, psi
4000 3900 3800 3700 3600 3500 3400 1,00E-03
1,00E-02
1,00E-01
1,00E+00
Delta t /(delta t +t)
Fig. 9.10: Pressure Buildup plot (Horner plot)
110
10.0 INTRODUCTION TO TYPE CURVE AND PRESSURE DERIVATIVE APPROACH 10.1 Type Curve Matching Methods
The diffusivity equation:
p wf = pi -
162.6 q μ B kt 3.23 +0.87 S log 2 φμ kh cr w
can be expressed in form of dimensioless pressure pD and time tD Dimensionless pressure, pD is given by :
Δp kh pi - pwf pi - pwf = = 162.6 q μ B 162.6 q μ B m kh
pD=
Dimensionless time, tD is given by: tD
0.000264 kt c rw 2
Where :
pD = dimensionless pressure
tD = dimensionless time
pwf = flowing pressure in the wellbore at any time t, in psi
pi = initial reservoir pressure in psi
q = oil rate in s.t. barrel per day
B = oil formation volume factor, dimensionless
k = formation permeability in milliDarcy (mD)
h = formation thickness in feet
f =
μ = viscosity of the oil in cent poise (cp)
formation porosity, dimensionless
111
c = oil compressibility or overall compressibility, c = c oSo + cwSw +cf , in 1/psi
r w = wellbore radius in feet
t = time since the well has been open with the rate q, in hours
S= skin factor, dimensionless
m = slope in the Horner plot, psi/cycle
The two last equations introduced in the first one give the following:
pD = logtD + 0.35 +0.87S This pD and tD are the basis for the “Type Curve” analysis commonly used in the petroleum industry up tp to 80‟s.
The Type Curve matching consisted in drawing many curves in log-log scale for pD vs tD with different ideal reservoir carachteristics (literature curves). Then the actual field data generated by a well testing are transformed in a curve of pD vs tD. When the field data curve in transparent paper matched the basic curve of literature it was possible to define further reservoir parameters such as: reservoir size, wellbore storage, recalculation of permeability (already perhaps computed with the Horner method), etc. This traditional Type Curve analysis has two major drawbacks: 1.
Non uniqueness: many reservoir models have similar looking type curves, and different values can be obtained by choosing the different type curves.
2.
Drawdown and not Buildup: the numerous type curves that have been published are for drawdown data. They can be used for buildup data in approximate manner, only under certain circumstances. Where boundaries or reservoir heterogeneities are evident and buildup data are more available than
drwadown data the type curve method is
misleading.
10.2 Derivative Analysis
112
In the 80‟s Bourdet and others introduced the “derivative” concept which helped
to make type curve matching more easy than the previous one and sometimes also without the need of the matching procedure. This new method of analysis has its basis in the published literature, and is rooted in the recognition and behaviour of various flow regimes. The pressure derivative application in oil well test analysis involves the combined use of existing type curves in both the conventional dimensionless pressure form (pD vs logt D) and the new dimensionless pressure derivative (p D‟ vs logtD) Thus, this new approach has combined the two approaches, (p D vs logtD ) and (pD‟ vs logt D ), in a powerful method of well test interpretation. Use of the pressure derivative p D‟ with pressure p D type curves reduces the uniqueness problem in curve matching and gives greater confidence in the results. Features that are hardly visible on the Horner plot or that are hard to distinguish because of similarities between a reservoir system and another are easier to recognize on the pressure derivative plot. In the Horner plot pressure changes are plotted vs logarithm of time and if the reservoir is infinite the plot has a straight line with a certain slope, m, as in the following equation: p
wf
=
p - m log i
kt φμcr
2
- 3.23 +0.87 S
w
From this equation we can derive the following by simplification: Δp = pi-pwf = m log(t) + b And by taking derivative of Δp in log(t), the symbol is Δp‟ but is commonly used p‟, we obtain, (being b a constant) :
p '
p log t
m
The plot of p‟ vs log t gives a zero slope line of a constant value m. Having m i n
oil field units always the same meaning :
113
m=
162.6 q μ B kh
Example of Δp And Derivative of Δp Plot Vs Log(T)
We have the following well testing data:
114
Tab. 10.1: Well Testing Data The plot of normal ΔP vs log(t) as per Horner Method is indic ated in blue in the
following graph, while the plot of the derivative of ΔP is indicated in red:
901
801
701
' P e P a t l e d
601
501
401
301
201
101
1 time Delta P
Derivative
Fig. 10.1: Plot of Delta P and P‟ vs log (t)
100 ' P e P a t l e d
m
1 time Delta P
Derivative
Fig. 10.2: Plot of Delta P and P‟ vs log(t) The Horner plot in blue above does not give many information except for the slope m, while the derivative give more information.
115
Usually the plot of the two, ΔP vs log(t) and derivative of ΔP vs logt (t) is done in
scale log-log, Because the log-los plot manifests many carachteristics of the reservoir, as follows (the same graph of before from semi-log scale to log-log scale). 10.3 Pressure Derivative Applications In Well Test Analysis
The application of the pressure derivative plot is that it is able to display in a single graph many separate characteristics that would otherwise require different plots. For example can shows:
Finite conductivity fracture
Infinite conductivity fracture
Dual porosity behaviour
Closed outer boundary
Linear impermeable boundary
Constant pressure boundary
Infinite acting radial flow shows on a log-log derivative plot, the following a.
A flat region of the derivative curve, Δp‟,: This indicates the radial flow.
b.
The two curves, Δp and Δp‟, at the beginning with the same slope. This
indicates the wellbore storage effect. c.
A peculiar different shape of the derivative tail after the flat region. This may indicates: dual porosity zone, drawdown or buildup test, boundary etc
d.
The 1.5 log cycle from the end of the wellbore storage effect. After that time can be taken the linear portion for the m determination in the Horner plot.
See figure below for all carachteristics of the derivative curve.
116
Fig. 10.3: Characteristics of derivative curves 9.4 Type Curves Of Dervative
Fig. 10.4 : Infinite acting radial flow regime
117
Fig. 10.5: Radial flow in composite reservoir
Fig. 10.6: Radial Flow for dual porosity/dual permeability
118
Fig. 10.7: Flow regime for Layered Reservoir
Fig. 10.8: Flow regime for Fault boundary
119
Fig. 10.9: Flow Regime for Horizontal Well with Strong skin effect 7- Drawdown and Buildup derivative curves. Comparison of the buildup and drawdown responses for the same closed reservoir :
he drawdown behaviour has an upward trending in the derivative;
the buildup behaviour has a downward trending derivative.
Fig. 10.10: Comparison of buildup and drawdown responses 9.5 Software Packages Dor Derivative Anaysis
120
The two main software package available in the market are the following:
Both of them are Pressure Transient Analysis softwares. Their methodology is based on the use of the pressure derivative. This methodology consists in matching the pressure data using simulation models which takes into account the detailed well production history. Interpret 2003 (Paradigm)
Tool description Interpret/2003 is based on the conventional Horner analysis and advanced type curves analysis techniques which use the pressure derivative curves as the main diagnostic tools. The analysis is performed using analytical models for early, middle and late time effects. The software is structured into 6 functional sections: 1.
Data section: allows the input of basic data, fluid type and PVT parameters (including a simple window for PVT estimation via correlations), bottom hole pressure and temperature data from multiple gauges, produced fluids rates. Options such as multiphase flow at the wellbore and in the reservoir are also available. Temperatures can also be loaded. Rates can be loaded as measured rates or analysis rates, the latter being the averaged values to be used for the interpretation
2.
Validate gauges: if more than one gauge is loaded, this section allows the user to compare recorded pressures and to perform a pressure and time shift on gauges. Gauge combination is also allowed. Gauges can also be compared by displaying pressure differences and, if rates are already loaded, log-log and superposition function for different flow periods.
121
3.
Validate rates: for a single selected gauge diagnostic plots (Log-Log and Superposition function) relative to different flow periods (draws-down and builds-up) can be compared for consistency. Automatic rate adjustment can be performed even on subset data.
4.
Diagnose: this section presents on the same windows the main diagnostic plots (Log-Log and Horner) and the diagnostic tools (trend lines for pressure derivative for early, middle and late time models). Partial results are also presented.
Fig. 10.11: Interpret 2003 – Diagnostic plot 5.
Matching: after setting the diagnose lines the matching option generates the corresponding analytical model. Real data and model lines are compared in Log-Log, Horner and Pressure History plots. Interpretation refinement can be done using the Model Controls window where different combinations of the interpretation models can be chosen and model parameters can be manually set. Variable storage and variable skin options are also available. Regression in the different plots and for selected parameters and data subsets can be done in order to automatically improve the match. Using
122
the regression option care should be taken to the meaningfulness of output parameters, even if the matching is satisfactory. Results of the model can be viewed also. Different analysis can be saved, re-loaded (file menu) and compared (select display).
123
Fig. 10.12: Interpret 2003 – Plots 6.
Design: this section allows generation of the theoretical pressure response of a test performed in a well with certain characteristics, inputting flow rates and test sequence. Input data (see data section)
124
must be loaded. The Model Control window allows selection of the model, input model parameters and definition of the gauge properties. The designed test is plotted in the usual Log-Log, Horner and Pressure History plots. The pressure response can be saved as a gauge (file menu) for conventional analysis. Saphir (Kappa Engineering)
Tool description A well test analysis performed by Saphir software may enhanced its reliability by using the following features:
a wide QA/QC section with, in particular, the tidal effect correction tool;
development of a numerical linear model based on under structured automatic grids with a modelling flexibility far beyond that of an analytical model;
(pressure/saturation fields);
development of a numerical non-linear model with advanced features near to reservoir simulation.
Saphir analytical method The analytical method of Saphir software is based on the same approach as in Interpret/2003 Nevertheless it presents some additional features as:
the regression on zone contributions.
Multiple flow period analysis: allows to analyze multiple flow periods considering superposition effect.
rd Log-Log- and superposition function, Saphir offers the possibility to plot Horner and user defined graphs. Analysis can also be performed on selected plots.
125
production history of a well whose behaviour has changed at a certain time (due to acidizing, fracturing, etc in a single analysis.
petrophysical or fluid properties in a linear direction.
126
127
Fig. 10.13: Saphir graphs Numerical analysis (linear) The 2-D numerical module can extend the modelling capabilities to simulations which takes into account a number of factors that cannot be considered in analytical analysis. The model is set up defining a under structured grid scaled on the reservoir map. The numerical model allows to consider the following items:
irregular outer boundary shape;
fault trajectories and leakage factor of each fault;
irregular composite zones;
reservoir thickness and porosity variation by a discrete set of values using
grids and other interpolations;
evaluation
of
the
pressure
response
of
a
well
when
other
production/injection wells are active in the same reservoir at the same time;
2-D and 3-D display and animations of pressure and/or saturation fields.
128
Numerical analysis (non-linear) Saphir
covers
also the assumption of slightly compressible fluids and the
pseudo-pressure function and takes into account non- linearities such as:
-Darcy flow into gas reservoir;
permeabilities curves;
Fig. 10.14: Example of Saphir numerical model
129
Fig. 10.15: Example of Saphir numerical model
130
11.0 EARLY TIME MODELS 11.1 Buildup Division Into Three Time Models
We can divide a buildup curve into three regions : 1.
Early-Time Region (ETR). In this region, a pressure transient is moving through the formation nearest the wellbore.
2.
Middle-Time Region (MTR). In this region, the pressure transient has moved away from the wellbore into the bulk formation.
3.
Late-Time Region (LTR). In this region, the pressure transient has reached the drainage boundaries well.
Fig. 11.1: Behaviour of shut-in pressure in an oil well during buildup The reasons for the distortion from a straight line in the ETR portion are the followings:
Wellbore storage effect.
Altered permeability near the wellbore, skin effect.
The MTR portion of the curve should be a straight line. This is the portion of the buildup curve that we must identify and analyze. Analysis of this portion will provide reliable reservoir properties of the tested well with Horner method. Some distortion from a straight line occurs also in the LTR portion. Here the pressure is influenced by boundary configuration, interferences from nearby
131
wells, reservoir heterogeneities, and different fluid contacts i.e. water/oil or gas/oil). The analysis of the pressure buildup in the transient time, MTR region, using the Horner method involves the following steps:
Find to, the equivalent flowing time before the well is shut –in, which is equal to the cumulative production of oil divided the oil rate q.
Plot pws versus log (Δt/(to+Δt)) in semilog paper.
Plot Δp = pws-pwf versus Δt to identify wellbore storage effects; identify
ETR and beginning of MTR.
The MTR ends when the radius of investigation begins to detect the drainage boundaries of the tested well; at this time the buildup curve starts to deviate from the straight line.
Once the MTR is identified, determine the slope and intercept. The slope m is of the straight-line portion of the Horner plot (MTR region), by extrapolating this line to infinite time at log Δt /(to+Δt) = 1 it is possible to
determine the intercept, p*.
On the straight-line portion of the curve (MTR region) we should read p 1hr at Δt = 1 hour, which is useful for the skin factor determination..
11.2 Wellbore Storage Effect
When a well is opened, the production at surface is initially due to the expansion of the fluid stored in the wellbore, and the reservoir contribution is initially negligible. This characteristic flow regime, called the wellbore storage effect, can last from a few seconds to a few minutes. Then, the reservoir production starts and the reservoir rate increases until it becomes the same as the surface rate. When this condition is reached, the wellbore storage has no effect any more on the bottom hole pressure response, the pressure data describe the reservoir behaviour and can be used for transient analysis. During shut-in periods, the wellbore storage effect is also called after flow: after the well has been shut-in, the reservoir continues to produce at the layer face and the fluid stored in the wellbore is recompressed. See figure below.
132
Fig. 11.2 : Wellbore W ellbore Storage Storage Effect Wellbore storage coefficient, C
For a wellbore full of a single-phase fluid, the expansion expansion of this compressed compressed fluid when the well is open to the surface is given by the following equation: ΔV = -coVw Δp
Where:
ΔV = volume variation variation of the fluid inside inside the wellbore, wellbore, bbl
co = fluid f luid compressibility, compressibility, psi -1
Δp = pressure variation variation in the wellbore, wellbore, psi
Vw = volume of the wellbore, bbl
The Storage coefficient, C, is defined as the vriation of the Volume per unit of pressure drop, as follows: C = - ΔV/Δp = coVw
p - V/C
q B t 24 C
Where:
ΔV = volume variation variation of the fluid inside inside the wellbore, wellbore, bbl
C = storage coefficient in bbl/psi
133
Δp = pressure variation variation in the wellbore, psi
q= oil rate at the surface, stb/d
Δt = time in which the flow q is happened, happened, hours
B =oil formation volume factor, dimensionless dimensionless
During the pure wellbore regime, the well is acting as a closed volume and, with a constant surface rate condition, the pressure changes linearly with time. The wellbore storage coefficient, C, can be estimated on a plot of the pressure change Δp versus the elapsed time Δt on a linear scale.
The response follows a straight line of slope m wbs, intercepting the o rigin of Δp axis. Therefore C, wellbore wellbore storage coeficient is equal equal to: C=
qB 24 mwbs
Where:
C = storage coefficient in bbl/psi
q= oil rate at the surface, stb/d
B =oil formation volume factor, dimensionless dimensionless
mwbs = slope in the graph Δp vs Δt, psi/hrs
Example Of Wellbore Storage
A well with a production of 4600 Bpd (formation volume factor, B = 1.47) has been shut in and has the following Horner plot of buildup:
134
4500
P* =4450 4400 4300 m= 200 psi cycle 4200
P1hr =4090
4100
BHP, psi
4000 3900 3800 3700 3600 3500 3400 1,00E-03
1,00E-02
1,00E-01
1,00E+00
Delta t /(delta t +t)
Fig. 11.3: Pressure Buildup Analysis-Wellbore storage in the early time It is evident evident that the well well in the ETR ETR region has not a linear pressure pressure this id due to the wellbore storage effect beside the skin effect due to to a damaged well while drilling. If we plot pws-pwf vs Dt in the ETR region we have a straight line as follows:
135
Fig. 11.4: Pws-Pwf vs delta time Then the value of the wellbore storage coefficient is : C = 4600 x 1.47 / (24 x 1600) =0.176 bbl/psi When a well is opened, the production at surface is initially due to the expansion of the fluid 10 11.3 Skin Effect
As already said in capter 8 the skin effect eff ect around the wellbore is caused by the drilling activity. activity. This causes a pressure pressure drop in the buildup, as shown in the following graph:
136
4500
P* =4450 4400 4300 m= 200 psi cycle 4200
P1hr =4090
4100
BHP, psi
4000 3900 3800 3700 3600 3500 3400 1,00E-03
1,00E-02
1,00E-01
1,00E+00
Delta t /(delta t +t)
Fig. 11.5: Pressure Buildup analysis – The Skin effect The skin effect around the wellbore, is a number without dimension, to be determined by the equation equation above mentioned. mentioned. In particular we have if:
S = 0, no skin , no damage around the wellbore
S > 0, there is formation damage ad the wellbore
S < 0, there is no damage around the wellbore, exceptionally we have improvement. W
See figure below:
137
Fig. 11.5: Skin Effect, S 11.4 Infinite And Finite Conductivity In Vertical Fracture
A common well stimulation method consists of creating a hydraulic vertical fracture from the wellbore to the formation . The reservoir / well surface of contact is significantly increased, thus producing a negative skin factor. Two main types of fractured well behaviour are observed: infinite and finite conductivity fractures.
138
The fracture is symmetrical on both sides of the well and it intercepts the complete formation thickness, xf is the half fracture length. With the infinite conductivity fracture model, it is assumed that the fluid flows along the fracture without any pressure drop. The hydraulic fracturing technique has been used from the 1950's to improve the productivity of damaged wells, or wells producing from low-permeability reservoirs. By injecting fluid into the formation, a vertical plane fracture is created and filled with propping agents to prevent closure. Infinite conductivity vertical fracture model
The well intercepts a symmetrical vertical plane fracture of half-length xf (see figure below).
Fig. 11.7: Vertical Plane Fracture The well and the fracture penetrate totally the reservoir thickness and there is no pressure loss along the fracture plane. Wellbore storage effects can be present in the well, and the fracture can be affected by a skin damage. Two characteristic regimes can be observed after the wellbore storage in the early time (ETR) , linear flow and pseudo radial flow, as illustrated in the following figure. Linear flow, with Δp proportional to Δt
0.5
and a half unit slope straight line on
pressure and derivative log-log curves (see figure below). The linear flow regime defines the hxf product, and therefore the fracture half-length x f .
139
Fig. 11.8: Derivative log-log curve in red From the following formula it is possible to determine x f : Δp'=2.03
qB hxf
μ
Δt 0.5 c k
Then xf is equal to:
qB μ Δt 0.5 xf = 2.03 h c k p' Where:
Δp‟ = derivative of the pressure vs logΔt, for Δt value
q= oil rate at the surface, stb/d
B =oil formation volume factor, dimensionless
h = vertical fracture thickness, feet
xf = half fracture length, in respect to the well, feet
μ = oil viscosity, cp
Φ = formation porosity, dimensionless
c = fluid compressibility , 1/psi
k = formation permeability, mD
140
Δt = shut-in time, hours
Pseudo-radial flow regime when the flow lines converge from all reservoir directions. During the pseudo-radial flow regime, the pressure follows a semi-log straight-line behaviour, as during the usual radial flow regime towards a cylindrical vertical well. The fracture influence is then described by a geometrical negative skin and the pseudo-radial flow analysis provides the permeability thickness product kh and the skin factor S. Finite conductivity vertical fracture model When the pressure gradient along the fracture length is not negligible, the low conductivity fracture model has to be used for the analysis of hydraulically fractured wells. This may happen for example when the permeability of the fracture is not very high compared to the permeability of the formation, especially when the fracture is long. With the finite conductivity fracture model linear flow is produced within the fracture, in addition to the linear flow regime from the pay zone into the fracture plane. The fracture geometry is defined as bilinear flow in following figure :
Fig. 11.9: Fracture Geometry When the pressure drop in the fracture plane is not negligible, a second linear flow regime is established along the fracture extension. Before the two ends of the fracture are reached, this well configuration produces the so-called bi-linear flow regime. During the bilinear flow, the pressure change, Δp, is proportional to the fourth
root of the elapsed time since the well was opened . With w f the width of the finite
141
conductivity fracture and k f the permeability in the fracture, the formula for Δp vs Δt is the following:
Δp = 44.11
qBμ h
kf w
4 f
φμck
4
Δt
Where:
Δp = pressure drop during drawdown, pws-pwf, psi
q= oil rate at the surface, stb/d
B =oil formation volume factor, dimensionless
h = vertical fracture thickness, feet
μ = oil viscosity, cp
f =
c = fluid compressibility , 1/psi
kf = fracture permeability, mD
wf = fracture width, ft
k = formation permeability, mD
formation porosity, dimensionless
Δt = elapsed flowing time, hours
On a plot of the pressure change Δp versus the fourth root of elapsed time Δt,
pressure response follows a straight line of slope m BLF, intercepting the origin, during the bilinear flow regime as in following figure:
142
Fig. 11.10: Plot of the pressure change vs time From the slope m BLF it is possible to calculate the fracture conductivity, kf wf, as follows:
qBμ k w = 1944.8 φμck h m 1
f
2
f
BLF
Where:
kf wf = frctiure conductivity, mD ft
q= oil rate at the surface, stb/d
B =oil formation volume factor, dimensionless
h = vertical fracture thickness, feet
μ = oil viscosity, cp
f =
c = fluid compressibility , 1/psi
kf = fracture permeability, mD
wf = fracture width, ft
k = formation permeability, mD
formation porosity, dimensionless
11.5 Partial Well Penetration
143
In the case of limited entry or partial penetration into the formation, the well communicates with only a fraction of the producing zone thickness. This could be due to plugged perforations, or for production problem to stay away from Gas Oil contact (GOC) or from Water POil Contact (WOC).
Fig. 11.11: Partial Well Penetration The distortion of the readial flow creates additional pressure drop in the well and increase the skin effect. The skin factor determined by a well test, S w will comprise both the Skin factor due to the permeability change, S k ,and the skin due to the partial penetration SB: Sw= Sk + SB The geometry of the partial penetration shall be the following:
Fig. 11.12: Geometry of partial penetration The sequential flow regime that happen in a well with partial penetration are the followings:
144
1.
Radial flow over the open interval hw, generates a Δp proportional to log(Δt) and a first derivative plateau (see below figure).
Analysis of the initial radial flow regime gives the kH x hW for the open interval and the skin factor of the well, Sw.
Fig. 11.13: Derivative plateau in red 2.
Spherical flow with Δp proportional to Δt
-1/2
and a negative half unit
slope straight line on the derivative log-log curve (see below figure). The spherical flow regime lasts until the lower and upper boundaries are reached. Analysis yields to the permeability anisotropy k V/kH .
Fig. 11.14: Derivative slope in red
145
3.
Radial Flow over the entire reservoir thickness with Δp proportional to log(Δt) and a second derivative stabilization. The reservoir permeability-
thickness product k H x hW , and the total skin Sw can be estimated from the second radial flow regime. The following figure represent the sequence of the three flow regime above described:
Fig. 11.15: Three Flow Regime in the derivative mode (red) 11.6 HORIZONTAL WELL
Advances in drilling and completion technologies have placed horizontal wells among the techniques used to improve production performance. For example in the case of gas cap or bottom water drive, horizontal wells prevent coning without introducing the flow restriction seen in partial penetration wells. Horizontal drilling is also efficient to increase the well surface area for fluid withdrawal, thus improving the productivity
146
Fig. 11.16: Horizontal well The geometry of an horizontal well is the following:
Fig. 11.17: Geometry Horizontal Well
In the above figure we consider :
homogeneous reservoir with sealing upper and lower boundaries,
the well is strictly horizontal,
the total penetration is 2L and half penetration length is L,
zw defines the distance between the drain hole and the bottom-sealing boundary,
the vertical part of the well is not perforated,
147
there is no flow towards the end of the well,
kH and kV are the horizontal and the vertical permeabilities.
In an infinite system, the geometry of the flow lines towards a horizontal well produces a sequence of three typical regimes, as described below: 1.
The first regime is radial flow in the vertical plane. On a log-log derivative plot, the wellbore storage effect is followed by a first stabilization.
During this radial flow regime, the permeability-thickness product , , is defined with the average permeability in the vertical plane, the average horizontal permeability and the well effective length 2L.
Fig. 11.18: Radial Flow Vertical Plan 2.
When the sealing upper and lower limits are reached, a linear flow behavior is established, as per following figure. The derivative follows a half-unit slope log-log straight line, from where kHL2 can be derived.
148
Fig. 11.19: Linear Flow 3.
Later, the flow lines converge from all reservoir directions towards the well, producing a horizontal radial flow regime (see figure below). The derivative stabilization corresponds to the infinite acting radial flow in the reservoir, the permeability-thickness product is kHh.
Fig. 11.20: Horizontal Radial Flow The three above regime, which happen in sequence in an horizontal well, since its opening to the production are depicted below in logΔp and derivative logΔp‟ vs log Δt, time elapsed since the opening of the well:
149
Fig. 1.21: Log Dp and derivate Dp‟ Vs log Dt
150
12.0 MIDDLE TIME MODELS 12.1 Homogeneous Reservoir
The Middle Time Region (MTR) of a buildup curve (see below figure), after the Early Time (affected mainly by the wellbore storage and skin effects) is the typical transient regime of an homogeneous reservoir. In this portion of the buildup curve we must identify and analyze the straight line with Horner method, in order to identify the slope, m, the kh, the skin factor S, the Flow efficiency etc, as already seen. Homogeneous reservoir means:
one porosity,
one permeability,
constance of fluid carachteristics with pressure variation as, viscosity and compressibility,
one reservoir uniform thickness,
homogeneous radial flow.
Fig. 12.1: Behaviour of shut-in pressure in oil well during- The Middle Time Region (MTR) 12.2 Heterogeneous Reservoir
151
Heterogeneous reservoir models have attracted a lot of attention from petroleum engineers since advanced software were available (last 30 yrs). Reservoir heterogeneities are identified by variations in the pressure response. Sometimes the pressure data deviates from the homogeneous behaviour only during the first minutes of the test period under investigation, in other cases it takes from several hours to several days before the heterogeneity becomes evident. The introduction of high accuracy pressure measurements and computerized loglog
analysis
technique
explains
today's
recent
use
of
heterogeneous
interpretation models. In addition, the derivative of pressure exaggerates the characteristic features of the response. The basic heterogeneous solutions assume two different behaviours are combined in the reservoir response. They are described as double porosity models and/or
double permeability models and composite systems, radial
composite system and linear composite system. These three models will be analyzed in details:
double porosity- double permeability model
radial composite system
linear composite system.
12.3 Double Porosity- Double Permeability Model
Among the different heterogeneous interpretation models, the double porosity model has been the most frequently discussed in the technical literature. They assume the existence of two porous regions within the formation. One region, of high conductivity, is called the fractures whereas the other, of low conductivity, is called the matrix blocks. As described in below figure the concept of double porosity is representative of the behaviour of fissured and multiple-layer formations, when the permeability contrast between layers is high.
152
The "fracture system" describes the high permeability layers, and the "matrix blocks" the tight zones. The matrix blocks are not producing to the well, but only to the fissures. In all cases, the fissure network provides the mobility, and the matrix blocks supply most of the storage capacity. A double porosity response depends upon the storativity contrast between the two reservoir components, and the quality of the communication between them. ACTUAL RESERVOIR
MODEL RESERVOIR
Fig. 12.2: Double porosity model Basic assumptions of the double porosity model
1.
Each point in the reservoir is associated with two pressures, namely pf the pressure of the fluid in the fractures, and pm, the pressure of the fluid in the matrix pore volume.
2.
The fluid flows to the well through the fractures system only; the matrix blocks are not connected.
3.
Most of the reservoir fluid is stored in the matrix blocks porosity, the storage of the fractures network is only a small fraction of the reservoir storage.
153
4.
Three matrix block geometries are usually considered, depending upon the number n of fissure plane directions:
5.
•
n = 3, the matrix blocks are cubes
•
n = 2 , the cylinder matrix blocks are cylinder
•
n = 1 matrix blocks like slab.
Two different types of matrix to fissure flow have been considered:
6.
•
flow under pseudo-steady state conditions.
•
flow under transient flow conditions.
In the double porosity models, all matrix blocks are homogeneous, and they have the same size.
Behaviour of the double porosity model
When a well is opened in a fractures reservoir, a rapid pressure response occurs in the fractures network due to its high diffusivity. A pressure difference is created between matrix and fissure, and the matrix blocks start to produce into the fractures. The pressure of the matrix blocks pm decreases as flow progresses and, finally, tends to equalize with the pressure of the surrounding fissures pf. Definitions
For the permeability thickness product kh, an equivalent permeability is used. I The equivalent distributed permeability (bulk fracture permeability) kf is a function not
only of the actual fissures thickness and intrinsic
permeability, but also of the fractures network characteristics (such as tortuosity and fissure connectivity when material
separates
individual
fractures). k h = kf hf
Two porosities are defined in double porosity systems. We call Φf a nd Φm, the ratio of pore volume in the fractures and in the matrix, to the total
reservoir pore volume (V f + Vm)
154
The average reservoir porosity Φ is given by: Φ = Φf Vf + ΦmVm In n fractured formations, both
Φ f and Vm are close to 1 . The average
porosity of above equation becomes then : Φ = Vf + Φm
The storativity ratio
w
expresses the ratio between the two porous
systems, the fractures system and the reservoir system in broad sense: w =
fractures storativity / reservoir storativity = dimensionless
Where: Fracture storativity = (fluid pore volume in the fracture ) by (fluid
compressibility). Reservoir
storativity
=
(fluid
pore
volume
in
the
reservoir,
matrix+fractures ) by (fluid compressibility) Usual values of
w are
0.1, sometimes down to 0.01 or even 0.001.
A second heterogeneous parameter, called interporosity flow coefficient λ,
is used to describe the ability of the matrix blocks to flow into the fissures. λ as expressed by Warren and Root is a function of the matrix blocks
geometry and permeability km: Where:
λ = interporosity coefficient, dimensionless
α is related to the geometry of the fractures system
r w is the wellbore radius, feet
km is the matrix permeability, mD
k is the reservoir permeability in broad sense, mD
λ defines the communication between the matrix blocks and the fissures. When λ
is small, the fluid transfer from matrix to fissure is difficult, and it takes a long time
155
before the double porosity model behaves like the equivalent homogeneous total system. Such behaviour is obtained for example, when the matrix is tight, and the permeability k m, is small. Low density of fissures is another example of poor matrix communication: the characteristic block size is large, and α is small. Usual values for λ are in the range of 10 -4 to 10 -10. A typical buildup diagram for an infinite fractures reservoir with a
λ
of 5 x 10
-6
and various w is the following:
Fig. 12.3: Buildup of fractured reservoir with various In the above graph it appears as the
w,
w
varies from 1, all fractured reservoir with
very little matix, to zero, i.e. no fractures in the reservoir, the beavihour of the buildup moves from a straight line to a double slope curve. The two slopes m are equal.
156
From the distance of the two slopes in psi , Δp, and from the slope m, as depicted in the following figure it is possible to derive the value of
as follows:
2.303 p
e
w,
m
Fig. 12.4: Buildup of fractured reservoir with double identical slope
157
Fig. 12.5: Buildup of fractured reservoir and derivative with variuos w 12.4 Radial And Linear Composite Systems
For the composite reservoir models, like all
heterogeneous reservoirs two
geometries are considered for the interface between the reservoir areas. 1.
Radial composite system assumes that the well is at the center of a circular zone, the outer reservoir structure corresponds to a second element (see figure below). This geometry is used to describe a radial change of properties, resulting from a change of fluid or formation characteristic. Such change can be man-induced in case of injection wells and in some cases of damaged or stimulated wells. It can also be observed when oil and gas saturations vary around the wellbore, for example when the reservoir produces below bubble point or dew point .
158
Fig. 12.6: Radial composite system, as two reservoirs 2.
Linear composite system assumes a vertical plane interface between the two reservoir regions:
the reservoir is divided into two semi-infinite
zones, the well is located in one of them . This composite configuration can be observed for example when a linear fault separates two different reservoir elements with different characteristics, or when a water drive is active in one direction of the producing zone.
Fig. 12.7: Linear composite system, as two reservoirs Composite reservoir assumptions
A discontinuity defines two distinct homogeneous regions in the infinite reservoir. The interface has no thickness.
159
The mobility (k/μ) and storativity (Φct) are different on each side, but the reservoir
thickness h is constant. The change of reservoir properties is abrupt, and there is no resistance to flow between the two reservoir regions. The well, affected by wellbore storage and skin, is located in the region 1: with the radial composite model, it is at the center of a circular zone of radius R, with the linear composite, the interface is at a distance L. Definition
The changes of reservoir mobility (k/μ) and storativity (Φct) are expressed with
the mobility ratio M and storativity ratio F of the two systems, defined as follows:
M =
F=
k/μ 1 k/μ 2 φct 1 φct 2
A mobility ratio M greater than 1 indicates a better fluid
mobility (higher
permeability and lower viscosity) in region 1 compared to region 2. A storativity ratio F greater than 1 indicates an better fluid storage (by higher compressibility and space) in region 1 in respect to region 2. 12.5 Radial Composite System
With the radial symmetry of the system, the two reservoirs have the pressure responses in sequence: 1.
First, the pressure response depends upon the inner zone 1 characteristics, and the well behavior corresponds to a homogeneous reservoir 1 response.
2.
When the circular interface is reached, a second homogeneous behaviour, corresponding to the outer region 2, is observed.
160
Fig. 12.9: Radial composite system, as two reservoirs figura uguale e allora ? The pressure response at the well in derivative analysis for different values of mobility ratios, M, and constant storativity ratio (F assumed equal to 1) is as follows:
Fig. 12.9: Derivative analysis for a radial composite system - The M influence. The pressure response at the well in derivative analysis for different values of storativity ratio, F, and constant mobility ratio (M assumed equal to 1) is as follows:
161
Fig. 12.10: Derivative analysis for a radial composite system - The F influence. The duration of the first homogeneous regime is a function of the inner region radius: with a large R, the transition occurs later. Before the transition, the early time response corresponds to the behaviour of a well with wellbore storage and skin in a homogeneous reservoir. The shape of the transition is a function of M and F. When both the mobility and the storativity change, the two transitions illustrated in the two above figures are superimposed on the response. 12.6 Linear Composite System
Two homogeneous pressure responses happen
in the
linear composite
reservoirs, but the second homogeneous reservoir describes an equivalent total system: 1.
The region around the well is producing alone, and the pressure behaviour corresponds to a homogeneous reservoir 1.
2.
When the linear interface is reached, the two regions are producing together. A second homogeneous response is observed due to the equivalent homogeneous system, which is defined by the average properties of the two regions.
162
Fig. 12.12: Linear composite system, as two reservoirs figura uguale e allora ? The pressure response at the well in derivative analysis for different values of mobility ratios, M, and constant storativity ratio (F assumed equal to 1) is as follows:
Fig. 12.13: Derivative analysis for a linear composite system - The M influence.
163
LATE TIME SCHEDULE In the Late-Time Region (LTR) the pressure transient has reached the drainage boundaries of the well. The effect of boundaries has been considered from the earlier studies of pressure transient analysis. In 1951, when presenting his historic paper, Homer discussed the response due to a single linear sealing fault on a build-up example. Today, complex boundary systems are used in well test interpretation, with sealing or constant pressure conditions. Partially sealing and conductive linear boundaries can also be identified and interpreted on well pressure responses. 13.1 Sealing Boundary : Fault Concept
The sealing fault model consists of a linear no-flow boundary, which closes the reservoir in one direction (see figure below). Such a configuration is encountered in faulted reservoirs but it can also be considered, as an extension of the linear composite system already discussed in paragraph 12.6.
fault
pressure waves
L = distance well-fault
well
Fig. 13.1: Fault boundary as a linear no flow boundary At the time t2 ,on above figure, the pressure wave has reached the boundary and the pressure image his rturned to the well.
164
As the flow time increases, the radius of investigation of the theoretical infinite reservoir curve continues to expand, and the image curve reaches the well and go further (time t 3) . The well bottom hole pressure in its drawdown starts to deviate from the infinite reservoir response, and drops faster. Ultimately, when the well has been flowing long enough (after time t4 ), the hemiradial flow regime is reached: the flow lines converge to the well with a half circle geometry. During the hemi-radial flow regime, the pressure at the well varies with the logarithm of the elapsed time and the slope of the semi-log straight line is double (2m) that of the infinite acting radial flow, as per following figure:
Fig. 13.2: Drawdown pressure with fault – Double slope indicates the presence of a fault The interpretation of the drawdown follows the Horner method , with the m slope and the other parameters, furthermore the distance L of the well from the fault is given by : Where:
L = distance well-fault, feet
k = formation permeabilità, mD
Δtx = is the time corresponding to the intercept of the two slopes, hour
165
f =
porosity, dimensionless
μ= oil viscosity, cp
ct = total compressibility, 1/psi
Also a pressure buildup test will give the two slope as follows:
Fig. 13.3: Buildup pressure test with fault – Double slope indicates the presence of a fault. In the derivative analysis a typical drawdown response is presented on the following figure for a well with wellbore storage and skin in a homogeneous reservoir limited by a sealing fault.
Fig. 13.4: Drawdown pressure test with fault in derivative model In the above typical derivative analysis with a sealing fault we note the following:
166
The early time part of the well response corresponds to the infinite reservoir behaviour.
During radial flow, the pressure response follows the first semi-log straight line as illustrated above in semi-log plot and, and on the derivative plot follows the first stabilization.
When the influence of the sealing fault is felt, the flow becomes hemiradial. On semi-log scale, the slope of the straight line doubles and, with the derivative, the curve follows a second stabilization at a level twice the first. In dimensionless terms Δp D‟ and
tD, the
first derivative plateau is at 0.5
and the second at 1. 13.2 Parallel Boundaries: Channel Model
The well is located between two parallel sealing boundaries, like a channel sand deposit. The well can be equidistant from the sealing faults as per following figure:
Fig. 13.5: Parallel Boundaries Or the well can be closer to one boundary than the other as per following figure:
167
Fig. 13.6: well cluster to one boundary This type of configuration corresponds to long narrow reservoirs such as channel sands. Typical example of parallel sealing boundaries (sand channel type) responses are presented in the following figures in drawdown semi-log plot and in derivative analysis log-log plot..
On the semi-log plot , after wellbore storage and skin effect the curve present the usual straight line slope m of the Horner method. Afterwards the curve tends to increase significantly without showing particular carachteristics.
In the derivative analysis the curve describes first the wellbore storage and skin effect, then it follows the 0.5 stabilization (dimensionless plot with ΔpD‟ and t D).
Later, when the two reservoir boundaries have been reached, the flow lines become parallel to the reservoirs limits, and a linear flow regime is established .The pressure changes proportionally to , and the derivative follows a ½ slope straight line.
The shape of the transition between radial and linear flow is a function of the well location in the channel. When the well is equidistant from the two boundaries, such as the well A, the transition between radial and linear flow regimes is short. If the well is closer to one of the two boundaries (well B), the characteristic behaviour of one sealing fault is seen before the linear flow.
The derivative stabilizes first at 0.5, then at 1 and finally it reaches the half unit slope straight line. solution, the well is located between two parallel sealing faults.
168
Fig. 13.7:Drawdown pressure test with a channel.
Fig. 13.7: Drawdown pressure test in a sand channel in dimensionless derivative analysis 13.3 Intersecting Boundaries: Wedge Model
With the intersecting sealing faults model, two linear no-flow boundaries limit the reservoir drainage area, the wedge is otherwise of infinite extension. The angle of intersection between the two faults can take any value smaller than 180° . See below figure:
169
Fig. 13.9: Wedge model The effect of two intersecting sealing faults on semi-log plot (Horner method) and in derivative analysis (log-log plot) is illustrated in the following two figures for a well with wellbore storage and skin in a homogeneous reservoir. In the following example, the angle between the faults is 60°.
The pressure response first describes the infinite reservoir behavior and later, when the two faults are reached, a fractional radial flow limited by the wedge.
When two intersecting faults limit the drainage area, a smaller fraction of the plane produces: on the semi-log scale, the slope of the straight line is increased by a factor of 360°/ q° ( in this example 6m, being
q
= 60°) and,
with the derivative, the curve follows a second stabilization at a level equal to 180°/ q ° (in this example 3, being
q=
60°).
170
Fig. 13.10: Drawdown pressure test in a well boundary
in dimensionless
derivative analysis
Fig. 13.11: Drawdown pressure test (dimensionless) with a wedge boundary in the Horner plot(1) 13.4 Multiple Boundaries: Closed System
A closed system behaviour is characteristic of limited reservoirs but it can also be encountered in developed fields, when several wells are producing and each well drains only a certain volume of the reservoir . It is important to note that the responses are different for a drawdown and a build up.
-During drawdown periods, when all boundaries have been reached after the infinite acting behaviour, the reservoir starts to deplete. The response follows the pseudo steady state flow regime, and the well flowing pressure becomes proportional to time. Pressure and derivative log-log curves merge on a straight line of slope unity at late time.
During build-ups, the shape of the well response is different. After shut-in, the pressure starts to build-up during the initial infinite regime but, later, it stabilizes and tends towards the average reservoir pressure .
In a closed reservoir, when all boundaries have been reached, the flow regime changes to pseudo steady state: i.e at any point in the reservoir the rate of
171
pressure decline is proportional to time, i.e.
dp/dt
is equal to a constant, as per
following figure:
Fig. 13.12: Closed reservoir : pressure behavior from infinite to closed reservoir The above graph with the logarithm of r, radius from wellbore axis, will be the following:
Fig. 13.13: Closed reservoir : pressure behaviour from transient to pseudo steady state. The pwf , measured at the bottom well, during drawdown will be as follows (as already seen:
172
Fig. 13.15: Closed reservoir : pressure behaviour at the bottom well . OK
As long as the closed reservoir behaves as an infinite one (time from t1 to t3 on the above figure), the pressure at the bottom well drops with the logarithm of time. When the well after a certain production is closed for a shut-in period, the pressure builds up until the average reservoir pressure p is reached, and the curve flattens, as per below graph:
Fig. 13.15: Closed reservoir : pressure behaviour at the bottom well in drawdown and buildup
173
The depletion, expressed by the difference (pi - p ) between the initial pressure and the final stabilized average pressure, is proportional to the cumulative production. The longer the duration of the drawdown period, the lower is the final average reservoir pressure p . Typical delta pressure and derivative curves (dimensionless) in log-log scale are presented on the below figure:
Fig. 13.16: Closed reservoir : pressure and derivative behaviour for drawdown and buildup The pseudo steady state flow regime, characterized by a straight line of slope unity on the late time, is seen only during drawdown test (dotted line). As opposed to drawdown responses, the derivate build-up response, as in the above figure, flattens and then drops. This illustrates the particular behaviour of closed systems, where derivative drawdown and derivative build-up curves have totally different late time responses. Calculation of reservoir volume
In a circular reservoir closed after a long drawdown with constant production rate it is possible to calculate the Original Oil in Place (OOIP).
174
Fig. 13.17: Calculation of reservoir volume (slide 171) The Δp = pi-pwf , during the pseudo steady state , after a long time of flowing, is
given by the following equation:
p 0.234
qB c1hA
t 162.6
qB A log 2 - log (C A ) 0.351 0.87S Kh r w
Where :
Δp = pi-pwf,
reservoir static pressure less bottom hole flowing pressure,
psi
A = circular area of the reservoir, feet 2
q = oil rate in s.t. barrel per day
B = oil formation volume factor, dimensionless
k = formation permeability in milliDarcy (mD)
h = formation thickness in feet
Φ = formation porosity, dimensionless
μ = viscosity of the oil in cent poise (c p)
ct = oil compressibility or overall compressibility, c t = coSo + cwSw +cf , in 1/psi
r w = wellbore radius in feet Δt = time since the well has been open with the rate q, in hours
S= skin factor, dimensionless 175
C A= shape factor of the area A of the reservoir, dimensionless..
The above equation plotted in linear scale as p wf vs Δt is a linear function of slope m* in the pseudo steady state as per following figure:
Fig. 13.18: Drawdown test up to the semi steady state in a circular limited reservoir- m* From the above equation:
m* =0.234
qB hA ct
From which:
OOIP =
hA B
= 0.234
q ct m*
Where :
OOIP = Original Oil In Place, without the water saturation S w,
at stock
tank condion, ft 3
A = circular area of the reservoir, feet 2
q = oil rate in s.t. barrel per day
B = oil formation volume factor, dimensionless
Φ = formation porosity, dimensionless
176
ct = oil compressibility or overall compressibility, c t = coSo + cwSw +cf , in 1/psi
m* = slope at pseudo steady state of pwf vs Δt, psi/hour
The shape factor CA can be determined by the following equation:
CA = 5.456
m m*
e
- 2.303 p - p* /m t int
Where :
CA = shape factor of the reservoir area, dimensionless
m* = slope at pseudo steady state on linear scale of p wf vs Δt, psi/hour (see below graph).
pi= initial reservoir pressure, psi
m = normal slope in the transient time of pwf vs lo g Δt, psi/cycle (see below graph)
pi-p*int = intercept for Δt = 0 on the y axis (see below graph)
177
Fig. 13.19: Drawdown curves to determine, m*, m, and p*int
178
13.0 WELL TEST EQUIPMENT 14.1 Surface Equipment
Fig. 14.1: General Layout of surface equipment Well head
Fig. 14.2: Well head Well head pressure
179
The pressure measurement is made through a Dead Weight Tester (DWT) which hydraulically balance the well pressure. Its accuracy is of the order of 0.1% of the measured value. Electric sensors are seldom used. The wellhead pressure data are not directly used in analyzing the test, but they allow a comparison with the pressure data recorded by bottom hole electronic gauges (Quality Control). Well head temperature The measurement of the temperature of the produced fluid is generally made by thermometers located on the production line. The surface temperature measurement at static conditions is not meaningful from a physical point of view. The temperatures measured under dynamic conditions have a low degree of reliability since they are affected by the external temperature. However, they have not a specific value in the interpretation. In the absence of measured gas rates, the T wf is used only to estimate the theoretical gas rate at critical flow conditions. The dynamic wellhead temperatures have a remarkable importance in dimensioning and planning the surface facilities. Safety valve
180
Fig. 14.3: Safety Valve Choke
Fig. 14.4: Choke Critical flow occurs when the pressure downstream of the choke is one-half or less than the pressure upstream from the choke. In this case, the flow rate through the choke depends only on variations of the upstream pressure and on choke setting. Changes in the separator pressure within the critical flow range does not affect the rate of flow through the choke. Flow rates can be estimated from choke coefficient tables in the critical flow condition range. Non-critical flow occurs when the downstream pressure is more than half of the upstream pressure. In this case, changing the separator pressure downstream from the chocke will affect the flow rate through the chocke. In the non-critical flow condition, estimation of flow rate cannot to be made from choke coefficient tables.
181
Flowlines
TUBING O.D. inches 1.900 2.375 2.875 3.500
lbs/ft E.U. 2.90 4.70 6.50 9.30
Nom. Up 2.75 4.60 6.40 9.20
Internal pressure J55 N 80 6,870 psi 8,980 psi 7,180 psi 9,380 psi 6,800 psi 8,900 psi 6,560 psi 8,580 psi
DRILL PIPE O.D. inches 2.375 2.875 3.500 4.500
lbs/ft
Internal pressure Grade D Grade E 11,350 psi 15,470 psi 12,120 psi 16,530 psi 10,120 psi 13,800 psi 7,210 psi 13,830 psi
6.65 10.40 13.30 16.60
All dimension in inches NOMINAL SIZE
0D
1 2 2½ 3 4 6 NOMINAL SIZE
1.315 2.375 2.875 3.500 4.500 6.625 0D
1 2 2½ 3 4 6
1.315 2.375 2.875 3.500 4.500 6.625
ID 1.049 2.067 2.459 3.068 4.026 6.065 ID .815 1.689 2.125 2.626 3.438 5.189
SCHEDULE40 Thickness Ibs/ft
Allowable WP, psi 2110 1470 1850 1640 1440 1210
-133 1.678 -154 3.652 -203 5.793 -216 7.575 -237 10.790 -280 18.974 SCHEDULE 160 Thickness Ibs/ft Allowable WP, psi -250 2.85 5720 -343 7.45 4600 -375 10.00 4200 -437 14.30 4130 -531 22.60 3980 -718 45.30 3760
ID .957 1.939 2.323 2.900 3.826 5.761 ID -599 1.503 1.771 2.300 3.152 4.897
XH – SCHEDULE80 Thickness Ibs/ft Allowable WP, psi -179 2.171 3470 -218 5.022 2490 -276 7.661 2820 -300 10.252 2560 -337 14.983 2280 -432 28.573 2070 XXH Thickness Ibs/ft Allowable WP, psi -358 3.659 9540 -436 9.029 6290 -552 13.695 6850 -600 18.583 6090 -674 27.541 5310 -864 53.160 4660
Tab____. Flowlines specifications Example Given: Flowline 2 ½” SCH 160 - Long 1000 ft.
Oil : Specifc gravity = 0.82; Viscosity: 60 cStokes; Oil rate = 800 BPD. Find : Pressure drop in the flowlines. Result: from graph. Pressure drop = 0.48 psi/100 ft ;Effective pressure drop = 0.48 x 10 x 0.82 x 2.48 = 9.76 psi
182
Heater
1.
Prevent hydrating at separators.
2.
Compensate for heat loss through a flow control throttling device (choke which consumes a large amount of well stream heat through free expansion.
3.
Assist in separation of water in oil or in water emulsions.
4.
Lower the oil viscosity to promote better Oil Measurement.
5.
Prevent waxes from coming out of solution in wax bearing oils, which
would foul the separator. For safety reason only indirect heater are used. Indirect heaters are heaters that transfer the heat, generated by gas or diesel combustion, indirectly via hot water (steam) to the hydrocarbon fluid
Fig. 14.6: Heater
183
Fig. 14.7 : Indirect Heater
Fig. 14.8: Flopetrol indirect heater specifications
184
Hydrates
The gas hydrates are generated by the contact of the light hydrocarbons (mainly methane) with water at liquid state. They are solid crystals, compact, and looks like snow. Hydrates occur with low temperature and high pressure. Hydrate Formation Conditions: This free water gas is a serious problem because it tends to freeze in the field equipment in the form of hydrates making maters and valves inoperative and even plugging chokes or pipe lines, This is why gas transmission companies require that most of the water vapour be removed (7 lbs of water vapour per MM cu.ft is a typical requirement). Natural gas hydrates have the appearance of hard snow and consist of chemical compounds of light hydrocarbon Ns dissolve in liquid water under certain low temperature and high pressure conditions. This formation process is accelerated where there are high gas velocities, pressure pulsations or other agitation, such as at elbows, which cause mixing of the hydrate components. These conditions can be predicted using the chart by D.L. The higher the gas pressure, the higher the temperature at which hydrates form. The higher the specific gravity of the well stream, the higher this temperature is, at equal pressures, ethane, propane, H 2O and CO2 form hydrates at higher temperatures than methane – Hydrate formation is therefore promoted by the presence of these components in the gas, whereas N 2 and penthane plus have no noticeable effect.
185
Fig. 14.9: Hydrates
186
Fig. 14.10 Temperature at which gas hydrates will freeze (from KATZ) Hydrates will form above each curve of typical gas (SG). Below the typical gas curve the hydrates will not form.
Fig. 14.11: Gas type curve and hydrates area
Fig.14.12 :KATZ Diagram for hydrates formation Separation
The separation is a receiver vessel where the total well-stream effluent is separated into liquid and vapour by using the large difference in densities of the
187
two phases. In principle, the separator could be simply a sufficiently large pressure vessel which would lower the velocity of the stream passing through it enough for complete separation. In order to reduce the dimensions and cost of separator, several flow system or device are used in assisting separation.
Fig.14.13- Separator Inside an horizontal test separator the velocity of the flow stream is considerably reduced due to his large size. Liquid mist extractor are installed in the gas chamber. The mist extractor are usually baffles, plates mesh contractors, etc. The advantage of a contracts separator is good handling of liquid slugs, some foam and heavy crude oils. Their disadvantages are their relatively large size, weight and cost in relation to their separating capacity. The liquid phase may itself be separated into the lighter density oil and the heavier water, thus obtaining three phase separation. This liquid separation requires relatively large water-oil contact area to be effectively carried out, making the small diameter, vertical separator somewhat unsuitable for three phase separation. Water in oil or oil in water emulsion will not separate readily by gravity from each other. In this case, heating of the wellstream and the use of emulsion breaking chemical injected into the stream may be of assistance. Condition to be met by a test separator 1.
A test separator should permit the separation, the metering and the sampling of all elements or phases of the effluent.
2.
It should be able to separate very different types of effluent.
3.
It should also accept products containing quantities of impurities such as muds, acids, etc…, particularly when the well is being cleaned up.
188
4.
Another requirement concerns the physical dimension of the separator: it should be as compact as possible facilitate transportation to the site, and to be easily accommodated on offshore platforms.
5.
Adequate protective frames for transportation also necessary, as well as protection against corrosion (tropical climate, marine environment, etc…)
6.
Lastly, a test separates should have appropriate ancillary piping to enable quick connection on site.
On account of the versatility necessary, test separators are not expected to achieve as perfect a separation as production station separators, but rather to separate in such a way that the separated elements can be reliably metered.
Fig 14.14: Haliburton Test Separator
189
(a)
ENTRY
(b)
(c)
ENTRY
HUILE LIBRE DU GAZ
ENTRY
(a)Spoon
(b) Plate
(c) Cyclon
Fig. 14.15: Different types of fluid entries
The volume of gas that a separator will remove is dependent upon: 1-Physical and chemical characteristics of the crude 2-Operating pressure 3-Operating temperature Vsep= Qoil x time = Q x 3 min = m 4-Rate of throughput Being Qoil in m3/min. 5-Size and configuration of the separator 6- Retention time SPEED
EXTRACTION OF
GAS
Q ENTR
GA
HUIL
EAU
LI UID
WATER
OIL
To measure rates oil, gas and water It is necessary to have: -No liquid foaming -Steady flow -Adequate lines and valve capacities
Fig. 14.16:Horizontal test separator - 3 phases (Oil, Gas and Water)
190
gas Diameter
D in inch x Length in ft
“
”
Liquid level d “
”
in inch
oil
D
+6 . ”
d
Fig. 14.17: Technical specifications Flopetrol Test Separator
0
Fig. 14.18- Surface sampling of Oil and Gas at the separator Rate measurements
Oil Rate
191
0 -6 . ”
The oil rate, in case of lack of surface measuring systems, could be estimated by an empirical relationship (W.E.Gilbert) which relies on the GOR and on the wellhead flowing pressure: Qoil = Pwf x D1.89 / [ 435 x ( GOR/1000 )0.546 ] where:
Qoil: oil rate at Stock Tank conditions (STb/day);
D dimensionless choke diameter, expressed as a fraction of 64” (i.e.: for
choke 12/64", Ø = 12),
Pwf : wellhead flowing pressure (psig);
GOR: gas-oil ratio from test (Scf/STb).
Above equation is valid in the case of critical flow. Critical flow conditions occur when the upstream pressure is at least twice the downstream pressure. In this case the fluid velocity reaches a maximum value and then keeps constant, apart from pressure variations downstream. However, the values obtained from this equation must be considered as a first approximation of the real rate data: the average error can be greater than 10 12%. In particular, this equation is affected by the uncertainty in the T wf temperature measured at well head under dynamic conditions. When temperature dynamic profiles are available, the wellhead T wf value obtained by interpolating the average temperature gradient in the string shall be used. In the absence of any measurement, the value of 520° Rankine (15.5°C) can be assumed as a first approximation. In all other cases, when surface measuring systems are available, the following measuring tools are used:
192
Oil Metering manifold
separator
On all types of separator the oil outlet is fitted with two meters in parallel with isolating valve. The ranges of these two meters make it possible to cover all oil rates.
FLOCO Flowmeter
Fig.14.19 Oil Metering manifold Gas Rate The gas rate is estimated according to the following equation : Qgas = C x Pwf / ( SGgas x Twf )0.5 where:
Qgas: gas rate at standard conditions (MScf/day);
diameter (inches);
Pwf : wellhead flowing pressure (psia); gas:
gas specific gravity (air = 1.0);
193
wf :
wellhead flowing temperature (Rankine degrees).
In the table below the values of the C coefficient are reported as a function of the diameter of the measurement line and as a function of the calibrated orifice (choke): Choke diameter
C coefficient
C coefficient
inches
2 inches pipe
4 inches pipe
1/16
1.524
3/32
3.355
1/8
6.301
3/16
14.47
7/32
19.97
1/4
25.86
5/16
39.77
3/8
56.68
7/16
81.09
1/2
101.8
100.2
5/8
154.0
156.1
3/4
224.9
223.7
7/8
309.3
304.2
1
406.7
396.3
11/8
520.8
499.2
24.92
56.01
Tab.: C Coefficient for Gas Rate
194
Gas Metering manifold
A calibrated orifice plate is used to meter the gas. Orifices can be changed also with the vessel under pressure. Differential and static pressure are recorded on a Barton recorder.
h
Fig. 14.20 Gas Metering manifold Oil burner
The burner is a main element of well testing set-up designed to dispose of the separated oil by combustion without air pollution by smoke and without site pollution by fall out of unburnt residues. Description The burner main piece is the combustion head, the combustion head comprises:
195
An atomizer to reduce oil into fine droplets
A cylindrical hearth to stabilize the flame and facilitate the circulation of the combustion air
A pilot flame fed with either propane, butane, or with gas from the separator.
A spark plug to ignite the pilot flame.
A water rig with special nozzles to inject water into the flame.
The burner is mounted on a standard FLOPETROL rotation device with its vertical swivel joint distributor, allowing a suitable orientation of the hearts according to the direction of the wind. Correct burning is obtained with an aft wind or a three-quarter aft wind and the maximum angle of rotation is 80° on each side of the burner‟s axis. Therefore, in order to continue testing, even when abrupt changes in the wind‟s direction occur, it is recommended that two burners be
installed on opposite sides of the offshore drilling unit. When requested. FLOPETROL can provide the standard supporting boom, complete with independent gas flare, and necessary piping to supply the burner. The boom is an assembly of several elements which are easy to carry and set up. Once mounted on the standard boom the BURNER is about 50 feet away from the structure of the barge.
Fig. 14.21 Seadragon
196
Flopetrol Seadragon Oil atomizer
Fig. 14.22 Oil burning capability 14.2 Down Hole Tools Wireline
Wireline is the technology that allows internal measurements of producing wells without interruption of production from the well. The aim of the wireline is to operate inside the producing tubing under producing pressure. It allows safe and rapid intervention with a minimum amount of preparation and minimum interruption of production. It can use AMERADA bomb as pressure and temperature gauge. DST
The DST is usually conducted with a downhole shut-in tool that allows the well to be opened and closed at the bottom of the hole with a surface-actuated valve. One or more pressure gauges are customarily mounted into the DST tool and are read and interpreted after the test is completed.
197
The tool includes a surface-actuated packer that can isolate the formation from the annulus between the drill-string and the casing, thereby forcing any produced fluids to enter only the drill-string. Occasionally, operators may wish to avoid surface production entirely for safety or environmental reasons, and produce only that amount that can be contained in the drill-string. Drill Stem Tests are typically performed on exploration wells, and are often the key to determining whether a well has found a commercial hydrocarbon reservoir. The formation often is not cased prior to these tests, and the contents of the reservoir are frequently unknown at this point, so obtaining fluid samples is usually a major consideration.. DST String The well testing objectives, test location and relevant planning will dictate the most suitable test string configuration to be used. Some generic test strings used for testing from various installations are shown below. In general, well tests are performed inside a 7” production liner, using full opening test tools with a 2.25” ID. In larger production casing sizes the same tools will be
used with a larger packer. In smaller casing sizes, smaller test tools will be required, but similarly, the tools should be full opening to allow production logging across perforated intervals. For a barefoot test, conventional test tools will usually be used with a packer set inside the 95/8 inch casing. If conditions allow, the bottom of the test string should be 100ft above the top perforation to allow production logging of the interval. In the following description, tools which are required both in production tests and conventional tests are included. The list of tools is not exhaustive, and other tools may be included. However, the test string should be kept as simple as possible to reduce the risk of mechanical failure. The tools should be dressed with elastomers suitable for the operating environment, considering packer fluids, prognosed production fluids, temperature and the stimulation programme, if applicable.
198
Two packers DST
Fig. ____ DST in open hole
Fig. 14.23 Running in the wellbore the DST
199
Fig. ___ : Typical Jack Up Test String with TCP Guns on Permanent Packer
200
Fig. ___ : Typical Test String with TCP Guns Stabbed Through Production Packer
201
Fig. ___ : Typical Jack Up test String Retrievable Packer
202
Fig. ___ : Typical Semi-Submersible Test String Retrievable Packer The most common test sequence consists of a short flow period, perhaps five or ten minutes, followed by a buildup period of about an hour that is used to determine initial reservoir pressure. This is followed by a flow period of 4 to 24 hours to establish stable flow to the surface, if possible, and followed by the final shut-in or buildup of a duration almost double of the flowing time. Basis of DST operations In simple terms, a DST is carried out by running test tools in a BHA on a test string in the hole . When the string is successfully installed and all pressure and function testing is completed, a fluid is circulated into the tubing to provide an under balance to allow the well to flow after perforating. The down hole tester valve is opened to flow the well to clean up perforating debris and invasive fluids from the formation, the tester valve is then closed to
203
allow the formation fluids to build-up back up to reservoir pressure which is recorded on pressure recorders or gauges. After a suitable time (usually 11/2 times the flow period), the tester valve is then reopened to conduct the planned flow and shut-in periods in accordance to the programme requirements to obtain other additional data and verification. Following figure shows a typical schematic of a simple single flow operational sequence.
Fig. __ : DST Typical Sequence of Events
204
Static pressure profile Bottom hole pressure
time
Bottom hole pr essure essure
Fall-off test
Temperature profile
Fig. ____ Typical Pressure Charts with DST Pressure Gauge (amerada) Completion type- test string
Cased hole
Open hole
h reservoir thickness
Sand face
Sand face
Fig. Type of completion
205
The test string refers to the subsurface equipment run below the well head. The configuration will depend upon the type of well, the installation, and the type of test. The simplest test string is that required for a DST (drill stem test) in which a temporary test string is run in the hole and set using a retrievable packer. The following diagram shows different versions of the DST.
Fig. Different type of DST Configuration Configuration The concerns with a DST are as follows: f ollows:
will the inflatable packer seal,
is the formation strong enough at the casing shoe to withstand withstan d gas influx,
will the borehole remain stable throughout throughout the duration of the test (all for open hole DSTs),
will the tubing joints joints have have sufficient sufficient integrity integrity to stop gas leaks leaks around around the joints or to withstand the corrosion of of fluids such as H2S? H2S?
The latter can be managed by specifying sufficient quality tubing materials and gas tight connections, but the packer integrity and borehole stability may be of sufficient sufficie nt concern that the policy may be not to run open hole DST. At the other extreme from a DST, the following figure shows a full production test string for a well test of a cased and lined well.
206
Fig. __ : Test String Details Note in this test string design that tubing conveyed perforating (TCP) guns are run on the bottom of the test string to reduce the need for wireline perforation, and allow underbalanced perforating. Note also the position of the gauge carrier relative to the TCP guns which will be at the reservoir depth. The test string should be designed to acquire sufficient data to meet the objectives of the test with the simplest equipment to run and operate (minimum wireline requirement), in a safe manner. Some of the constraints which will affect the test string design include :
the expected flowrate
the test pressures and temperatures
207
sufficient tubing tubing size size to run required required wireline wireline tools (i.e. a 2 1/8" through through tubing perforating gun)
the ability ability to carry carry sufficient sufficient pressure and temperature recorders
the ability to control the well safely
the ability ability to withstand withstand corrosive fluids (i.e. CO 2, H2S) - in new areas where
the presence presence of of it is uncertain the string string will normally be for sour service. service.
A typical test string is 3 1/2" tubing for a standard well test, with 5" tubing for high rate tests. Remember that the rate has little material effect on the well test interpretation, unless one objective is to establish the maximum potential of the well. Perforations
Perforations procedure Here are some general guidelines on perforating procedures. Intervals should be perforated bottom up to avoid wireline passes across perforated intervals and thus reduce the potential for stuck tools. Perforating is generally performed underbalanced in order to clean debris out of the perforated channels. This can be optimised by perforating all intervals together through the use of tubing conveyed perforating (TCP) guns. Perforating can also be performed over pressured in order to fracture the reservoir as the well is perforated. This can help to increase well productivity, but is a special application. As soon as the first perforating run has taken place a clean up period should be considered to displace wellbore completion fluid from the well and avoid it slumping into the formation f ormation and potentially causing wellbore damage. Perforation gun type and ratings should be designed to maximize well productivity.
208
Wireline Gun
Through tubing Gun
Tubing conveyed Gun
Fig. Perforating gun
Packer
Function To isolate the production layer from mud weight Types of packers, according to utilization. utilization.
For Open hole test
For Cased hole test
For Tubing completion
Principal components
Friction slips
Rubber packing element for sealing
By-pass
209
Fig. _Packer Tester valve
Operated with annulus pressure
Dual Valve Tester + Circulating Valve operated by pressure pulses in the annulus (Pulse Operated Tool)
210
Fig. Tester valve Circulating valve
Function It permits the mud circulation after the test and the fluid recovery inside the tubing Types
operated with pressure from the tubing (MIRV and MCCV).
operated with pressure from the annulus (OMNI )
single shot (SHRV, APR-A)
211
Fig. Circulating valve Bottom hole fluid sampler
The sampler has two valves and a small tank (capacity from 2 to 10 litres). It is mounted above the Tester valve. The sampler takes a fluid sample (for PVT study) during the flowing of the well in dynamic conditions. The formation fluid sample is taken at the end of the test.
212
Fig. Bottom hole fluid sampler 14.3 Measuring Equipments Gauge Carrier
The Memory Gauges are located in the DST assembly by means of a Gauge Carrier (Bundle Carrier). The carrier can take up to four gauges for pressure and temperature recording. The reading are done through dedicated ports, located inside the tubing.
213
Fig. Gauge carrier
214
Fig. Memory gauges
215
15.0 DOWN HOLE GAUGE 15.1 Down Hole Data
The Well Testing principle is to analyse the reservoir response to an input signal (the imposed rate) to which an output signal (the bottom hole pressure) corresponds. The identification of the flow regimes in the formation, the main petrophysical properties, the potentialities and the physical limits of the reservoir are based on the bottom hole pressure response. 15.2 Bottom Hole Pressures And Temperatures
The recording of bottom hole data during the test is possible by using electronic gauges. Mechanical gauges (Amerada) are obsolete due to their poor performances with respect to electronic gauges. In accordance with the operation constraints and depending on the string-well system, the gauges are located as close as possible to the producing formation in order to reduce errors in referring the values from the measurement point to a reference depth. The electronic gauges can provide the bottom hole temperature and pressure with variable sampling rates (from a minimum of 2 seconds between the measured data). However, it is recommended to use sampling intervals higher than 5 seconds since lower values can lead to wrong temperature measurements. This is due to the thermal inertia of the tool which is not able to adapt to fast temperature variations. A general criteria to be followed is to decrease the sampling rate at each phase modification (from flowing to shut-in and vice versa). At longer times, the sampling interval can be gradually increased. However, for tests shorter than 10-15 days it is suggested not to select sampling rates longer than 15 minutes so as to have a suitable data management without affecting the continuity of the measurement. As a common procedure, the Service Company provides the results of the instrumental monitoring such as cumulative times, pressure and temperatures..
216
15.3 Pressure Measurement
Pressure measurement can be made at surface and down hole. For production tests down hole gauges are invariably run, and these are of three main types:
mechanical gauges (eg Amerada type)
electronic memory gauges
electronic surface readout (SRO) gauges
Electronic gauges have now replaced mechanical gauges for most down hole applications. There are two main categories of electronic gauge; strain gauge and quartz crystal gauge. Gauge selection is based on the criteria listed in the following table, plus price. The following table gives the manufacturers specifications for a typical strain gauge and quartz crystal gauge: MSG-S-20 strain
MQG-20 Quartz
Pressure
range (psi) accuracy (psi) resolution (psi) Temperature range (°F)
accuracy (°F) resolution (°F) memory type memory capacity O.D./ lengh (in)
0-2000 p ±4 0.6 32-347 1.08 1.08 EEPROM (electrically) 32768 (temp+press) 11/4 / 56.3
0-20000 ±4 0.2 32-347 0.36 0.036 EEPROM 32768 press, 8192 temp 1 11/16 56.3
Tab. ____ Geoservice gauges In general, the gauge accuracy, resolution, robustness for the reservoir conditions and price are prime considerations in gauge selection. Resolution refers to how small a difference in pressure the gauge can detect. 15.4 Technology
Apart from technical specifications, the measurement instruments can be subdivided into two main types:
217
1. SRO Gauges (Surface Read Out), 2. Memory Gauges Surface Read Out (SRO) gauge
During the various test phases SRO gauges allow a real time monitoring of the data being measured. This is because the gauges are run in the well by a mono conductor cable allowing the transmission of the signal from the bottom to the surface. Advantages:
minimization of costs (very high if the tests are carried out with the rig on site or in offshore operations). or
static profiles to assess the real flow distribution and the nature of the fluids along the wellbore. It is also possible to have relevant information on the portion of the formation that actually contributes to production (thermometry).
Direct action on sampling times during the data acquisition when the original test programme needs to be modified.
Disadvantages:
personnel provided by Service Companies.
consequence, in the case of malfunctioning, there is no way to verify the reliability of the instrument response. Use: Recommended in exploration wells to have an immediate verification of the reservoir response and in all situations requiring real time monitoring and immediate decision making for peculiarity and importance. Other memory gauges
218
These gauges are run in the well by a harmonic steel cable (slick-line) and placed in nipples in the completion string. Alternatively, they are directly run with the testing string in a tool called "bundle carrier”.
At the end of the test they are retrieved from the well. The main difference with respect to the SRO gauges is that it is not possible to monitor the pressure response in real time; only at the end of the test the collected data can be analysed. In fact, the acquisition is guaranteed by a battery pack (generally lithium based) located below the gauges. All the relevant data are stored in the tool internal memory. Only at the end of the test the gauges are retrieved and the data recorded unloaded and available for interpretation. Advantages:
data acquisition.
redundancy in some cases (especially in exploration wells) a third memory gauge is added, also to solve potential inconsistencies between measurements.
and support personnel during the test. Disadvantages:
the test in real time.
Limited test time due to the battery efficiency that depends on the sampling rate and on the number of data recorded (times, pressures and temperatures). Moreover, the battery duration is a function of the bottom hole temperature: the higher the temperature the lower the duration of the batteries.
Use:
219
Conventional memory gauges can be combined with SRO in exploration wells and, generally, in development wells with definitive completion. 15.5 Main Gauges Specifications
When planning a test, the gauge is the key element to reach the designed targets. The main gauge properties are: stability, resolution, accuracy, and stabilization time. Stability
Property related to the drift phenomenon. It defines the shift in measured pressure compared to the actual value. Drift phenomena tend to amplify in time and are generally positive. Drift does not depend on the magnitude of the measured pressure. The importance of the drift varies from gauge to gauge and for the same type of gauge there are different types of drifts. As an example, the indicative laboratory drift values for different types of gauges are reported:
psi/week;
s: ~0.2 psi/first week, then < 0.1 psi/week.
It can be noted that quartz gauges are very stable and do not have drift problems. Long tests, of the order of several weeks, require the application area of Quartz Gauges. Resolution
The resolution of an instrument represents the amplitude of the smallest step detectable in monitoring the real pressure. Thus, all the gauges reproduce the real physical pressure behaviour in a reservoir by steps. The resolution is a property varying from gauge to gauge. A high resolution gauge can be an efficient choice for tests carried out in very high permeability formations. Indicative laboratory resolution values are as f ollows:
220
psi @ 10000
psi);
Accuracy
For a given pressure, it defines the relationship between the gauge pressure measure and the actual value. Accuracy laboratory values for the different gauges are as follows:
psi);
psi);
Quartz Gauges: 0.02% full scale (i.e. 2 psi @ 10000 psi).
Stabilization times
Time necessary to stabilize a gauge after abrupt pressure and temperature variations (i.e. during the steps when carrying out static and/or dynamic profiles). It is defined as the time necessary so that the difference between the gauge value and the actual value is smaller than 1 psi. It can vary from more than 10 minutes in the case of Amerada to less than 1 minute (quartz gauge). All the above values provided by Manufacturers were obtained under laboratory conditions by submitting the gauges to increasing pressure steps from 1000 psi to 10000 psi. The temperature was kept constant at a value of 150 °C for the testing time. Based on the existing technology, all the electronic gauges are suitable to work at reservoir temperature up to 150°C. Special gauges must be required when testing HP-HT environment with reservoir temperature greater than 150°C. The current technological limit is some of 185-190°C. HP Quartz pressure gauge
The quarts technology that Hewlett-Packard first introduced to the oil and gas well industry an 1970 is still the standard for pressure measurement applications
221
requirement applications requiring extremely high accuracy resolution and repeatability. These features combined with its rugged construction, make the probe ideally suited for petroleum applications, oceanographic research and subterranean hydrodynamic studies. High Precision, Resolution and Repeatability Capable of sensing wellbore pressure changes as small as 0.001 psi, the probes measurements can be instantly observed and recorder on the surface. With an accuracy better than ± 1.0 psi and ± 0.01% of the precision of you measurements. How it works.
The essential pressure-measuring components of the HP 2813C Quartz Pressure Probe are its sensor crystal, reference crystal and electronics pc board. The sensor crystal, which is an direct fluid communication with the well changes the frequency of its oscillations in response to pressure. The reference crystal, which is protected from applied pressure, subtracts the effect of the temperature changes from the sensor crystals frequency The resulting frequency is then transmitted by the electronics pc board through a centre conductor, armoured electric line to an HP 2816A Signature Process on the surface. This processor conditions the pressure-related signal frequency counter. The counters signal can be converted to a pressure reading when processed with the calibration data in a desktop computer. The sensor crystals high resolution is essentially constant and independent of operating pressure and temperature. It‟s in essentially minimizes hysteresis and
zero draft thus estimating the need for frequent recalibration
222
Fig. Truck with winch Double drum winch
Electrical line 20000 feet of 7/32” electrical line
Cable rating: Maximum pull: 4.500 lbs. Maximum temperature:
250°F ; 390°F
Slick line 25000 feet of .92” slick line
On both drums hydraulic power provided by hydraulic pump controlled by progressive servo valve
Winch accessories Adjustable depth meter, CCL box, Martin Decker
Hp quartz pressure gauge Resolution = 0.001 psi Accuracy = + - 1.0 psi or + - 0.01 % of the reading Downhole gauge location
Pressure gauges are best located down hole close to the reservoir, but there is an extra cost compared to surface location. Gauges can be located above or below the perforations. Advantages below the perforations are the ability to perform wireline logging, avoiding constricting the wellbore and minimising turbulence. Disadvantages are that perforating debris or sand may fall onto the tools making them difficult to subsequently retrieve. In any case, the gauge is unlikely to be located at the reservoir datum depth (which is a specific reference depth for any one reservoir) and reservoir pressure measured at the gauge will need to be corrected to the datum depth.
223
This requires knowledge of the fluid gradient inside the wellbore as well as the reservoir fluid gradient. The former can be determined from the gradient stops performed when the gauge is run in the hole and the latter from the RFT pressures. Real time pressure readout (surface read out: SRO) If down hole pressure gauges are monitored in real time, the test can be interpreted in real time, giving the opportunity to extend or curtail the test as the opportunity arises. There are large potential rig time savings in being able to do this. For example, if the test objectives can be met after 12 hours of build-up, there is no additional benefit of remaining shut in for another 12 hours, even though the well test design specified a 24 hour build-up. SRO gauges are powered from surface, so the need for battery packs is eliminated, making them attractive for harsh conditions (HP/HT wells). SRO is achieved by passing a signal from the down hole gauge to surface using electric cable run inside the production tubing to surface. Typically a down hole gauge will record the drawdown data and upon shut in, the wireline will be run into the well to transmit the stored drawdown data and then transmit the build-up data in real time. If a PLT is run with the well test, the electrical wireline provides the real time data. The downside to such operations is the increased potential for fishing jobs du e to wireline operations.
224
16.0 WELL TEST DESIGN AND COSTS
After the well test objectives have been defined, the following steps are required to design a test. 16.1 Well Test Design For An Oil Well
1.
Acquisition of input data: Input data
Source of input data
Geological information
Geologist
Sedimentological information
Sedimentologist
Petrophysical data
Geologist
PVT data
PVT analyst
P,T reservoir
Subsurface geologist/Reservoir Eng.
Tab . 16.1 Input data for Well Test For production wells additional information is necessary:
tion history,
-fracturing
operations, well
washing
due
to
asphaltens & paraffins presence,
2.
Acquisition of information about possible constrains relative to:
225
3.
Selection of the optimal test sequence; Generation of the theoretical pressure response to be used as the reference case with the interpretation software (i.e. Interpret/2003 and/or Shaphir);
4.
Performance of sensitivity analyses by modifying the relevant parameters:
-up (drawdown),
l reservoir model.
5.
Display the obtained results (oil well), i.e.: (k) (Fig. 16.1) at different skins
(S); permeability (k) (Fig. 16.2) considering
different models; s (Ri) vs. time (t) (Fig. 16.3) at different permeability
values.
Fig 16.1 Sensitivity PI vs k at different skin S
226
Fig 16.2 Sensitivity PI vs k with different model
Fig 16.3 Sensitivity Ri vs time at different k 16.2 Well test design for a gas well
After the well test objectives have been defined, the following steps are required to design a test:
227
1.
Acquisition of input data: as per Tab 6.1
For production wells additional information are necessary: er rates),
asphaltens/paraffins presence,
2.
Acquisition of information about possible constrains relative to:
3.
Selection of the optimal test sequence;
Generation of the theoretical pressure response to be used as the reference case with the interpretation software (Interpret/2003 and/or Shaphir); 4.
Performance of sensitivity analyses by modifying the relevant parameters:
in, -up (drawdown),
5.
Display the obtained results, i.e.:
2
) vs. Log(q) (Figure 16.4) at different skins;
228
Log (Δp2) vs. Log(q) (Figure 16.5) considering different
models;
Fig 16.4 Log (Δp2) vs. Log(q) at different skin (S)
Fig 16.5 Log (Δp2) vs. Log(q) with different models 16.3 Design Appproach
229
230
17.0 FLUID SAMPLING 17.1 Surface Sampling
The objective of reservoir fluid sampling is to collect representative samples of the reservoir fluids at the time of sampling. In general terms oil, gas and even water samples are required to properly characterise the formation fluids. Sampling is generally performed in the initial exploration and/or appraisal phase when the fluid is still characterized by its original composition. This is a crucial step for reliably predicting the future reservoir behaviour. Two methods are used for sampling reservoir fluids. They are referred to as “subsurface sampling“ and “surface sampling”. In this second method, sampling
can be made at the separator (most likely) as well as at the wellhead. When sampling exploration wells, subsurface sampling is always associated with surface sampling. As a general procedure, sampling operations can be planned either during the main flow phase or at the end of the test after the final build-up. The choice of the sampling method is influenced by several factors, such as :
type of reservoir fluid;
volume of sample required by lab analysis;
mechanical conditions of the well;
limits of the available gas-oil separators equipment.
The key factor to collect a representative reservoir fluid sample is the preliminary conditioning of the well. This consists of producing the well, for a certain time, at a rate which removes all the altered (non representative) fluid from the wellbore. The recommended procedure to reach such a situation, consists of producing the well in a series of “step by step” flow rate reduction. A stabilized gas -oil ratio
(GOR) should be achieved and measured at each step. The well is considered to be sufficiently conditioned when further rate reductions have no effect on the GOR which remains constant over time. Monophasic flow conditions are then basically achieved and sampling can be successfully performed.
231
Special attention must be dedicated when sampling oil reservoirs (light - volatile oil) if the saturation pressure (or dew point pressure for gas condensate) is closed to the initial static pressure. During the sampling phase the following parameters should be stabilized and properly monitored:
Fluid flow rates (Qoil, Qgas, Qwater ),
Bottom Sediment & Water (BSW),
Gas Oil Ratio (GOR),
Wellhead pressure and temperature,
Separator pressure and temperature.
In addition, the main physical fluid properties, such as oil and average gas gravity as well as the presence of CO 2/H2S, should be carefully evaluated. As a general procedure, all the surface/downhole samples collected during the production test must be properly validated at the wellsite before they are sent to the labs. In the case of samples inconsistency, the operation must be repeated. Sampling Recommended Practice
Recommended Practices: 1. Single phase downhole samples may be taken for oil wells with low water cut to compare with surface recombination samples. These samples are particularly useful for wax or asphaltene detection. 2. Ensure that sufficient separator gas is acquired when taking surface separator samples to allow recombination with the oil samples to replicate reservoir conditions. 3. Samples of separator water and any solids production should always be taken. 4. If samples taken during the main test are thought to be poor, further samples can be taken by reversing out the tubing contents. Maintaining back pressure at the choke manifold may allow single phase samples to be collected, however this will not always be possible.
232
5. Chemical injection (e.g. Methanol, Glycol) should not take place upstream of the sampling point, when samples are being collected, unless operationally unavoidable. If chemicals are injected upstream of the sampling point, this should be noted on the sampling tag and samples of the injected chemicals should be sent with the hydrocarbon samples to the laboratory. 6. Dead oil and condensate samples should be taken for assays. Small separator gas (300cc), stock tank oil or condensate (lOOcc) and water (lOOcc) should be taken for geochemical analysis. 7. Duplicate samples should always be taken as a minimum, however triplicate are preferred. 8. Cylinders should never be shipped liquid full (it is illegal). If a piston cylinder is used, an inert gas should be used on the backside of the piston. Cylinders should be liquid filled to no more than 90% of capacity. All cylinder pressure ratings should be checked to make sure they are appropriate. The pressure rating needs to have been certified within the last five years and this will be shown by a stamp on the cylinder. Sampling Programmes
The particular sampling programme for a test must be designed taking account of the expected fluid type, reservoir conditions and overall aims of the test. Many test procedures include a low rate sampling flow period after completion of the main test. If sampling is preceded by a high rate test with flowing bottom hole pressures below the saturation pressure of the fluid, then the well needs to be conditioned prior to sampling. This conditioning period is designed produce out all the fluid that has been below the saturation pressure. If the saturation pressure is not known or the reservoir fluid is saturated, all fluid that has seen more than about 100 psi drawdown needs to be produced before sampling, in order to ensure that the sample is representative. In all cases it is important that the samples are collected under stable flowing conditions. For fluids near their saturation pressure, it is sometimes recommended that the low rate sampling flow period should be early on in the test, to assure that the reservoir has not been dropped below the dew point or bubblepoint pressure during previous flows. Unfortunately, information on saturation pressures are in many cases unknown before conducting the test.
233
The typical sampling requirements for oil, gas and gas condensate reservoirs are as shown in table 17.1. These are however only a guideline and depending on the detailed objectives of the test additional samples may be required.
Sample Type
Source
Container Type
Typical Size
Separator Gas
Separator Gas Line
Pressure Bottle
20 Litre 3 per 300cc for period geochem. Analysis
Separator Separator Oil or Oil Line Condensate
Pressure Bottle
600 cc
Single Phase Oil
Pressure Bottle
600 cc
Metal Drum
45 gallon
Plastic Bottle
1-5 Litre 100 cc for Geochem Analysis
Wellhead
Dead Oil or Separator Condensate or Gauge Tank Produced Separator Water water outlet Gauge Tank or Reverse Circulation
Number Requires
Remarks
flow Used for PVT recombination of reservoir fluid. Geochemical analysis 3 per flow Used for PVT period recombination of reservoir fluid. Condensate also for Geochemical Analysis As Required Only if saturation pressure is less than the flowing wellhead pressure 2 per test For crude assay Water salinity every hour and retain 2 to 3 samples per flow period
For water salinity analysis, Rw and Geochemical analysis
234
Produced Solids
Wellhead or Plastic Separator Bottle
Bottom hole Bottom of Pressure Fluid test string Bottle Samole
1 cubic centimetre packages. 5 litre well head fluid samples for filtration Tool Size dependant
Check every hour or more frequencly if high solids production retain 2 to 3 samples per test Minimum of 3, but as required to obtain representative reservoir samples
For gravel pack sizing
Reservoir PVT fluid analysis
Tab. 17.1: Typical T ypical Sampling Requirements It is important that all necessary flowing conditions are noted whilst sampling. This information is used for the recombination of samples and for fluid analyses. Indeed this data is as critical to good sampling as the t he samples themselves. The conditions to be noted are as follows: Upstream of Choke
Downstream of Choke
Reservoir
Flowing Pressure
Separator liquid rate
Reservoir Pressure
Flowing Temperature
Separator gas rate
Reservoir Temperature
Separator temperature
Flowing Pressure
Bottomhole
Separator pressure Stock tank liquid rate Stock tank temperature t emperature Stock tank pressure
Other:
Choke size
Gas lift rate and composition (if used)
235
Gas gravity used to calibrate orifice meter readings (Fg)
Stable Flow
When sampling to obtain representative samples for PVT analysis the following producing conditions are needed:
Stable flowing bottom hole pressure with minimum drawdown.
Stable flowing wellhead temperature and pressure.
Well producing producing clean, fresh fluid from reservoir.
Separator flowrates stable and measurable. measurable.
No major fluctuations in separator levels.
Stable separator pressure.
Flowrates not too high - avoid liquid carry over.
Although separator samples are the most convenient convenient samples to take, the accuracy of the PVT analysis on these samples relies on a large number of conditions being stable. Even fairly small disturbances to separator the equilibrium or stability will effect the sample quality. The absolute accuracy of any PVT data derived from these surface samples is a direct function of the accuracy of the measurement of oil and gas production rates. Since oil rates can generally be measured to no better than 5% accuracy and gas rates to no better than 3 % accuracy, recombination GOR's are always uncertain. This means that bottomhole sampling will usually always give more representative representative fluid f luid samples. Surface Sampling Points
The choice of sampling location will depend on the fluid properties and the flowing conditions of the well. In general the following will apply: Well head pressure greater than bubble point pressure - surface samples upstream of choke.
236
Well head pressure less than bubble point pressure - downhole sampling or separator recombination sampling The following guidelines should be followed f ollowed in selecting sample points: Separator Gas Line
Upstream of the Orifice Plate / Daniels Box.
As close to the separator vessel as possible.
Not immediately immediately downstream downstream of thermo thermo well or other tappings tappings in the flowline.
Not immediately after a bend in the flowline.
Ideally the sampling point should protrude into the centre of the gas flowline and face upstream. However, a pipe into the stream is acceptable.
Note: The sampling point should not be on the lower half of the flowline cross section, due to any possible free liquid or liquid carryover being present. If the sampling point has to be fitted flush to the inside surface of the flowline, then it is preferable that it is on the tog of the line and not on the side. Separator Oil Line
As close as as possible possible to the exit of the oil flowline from the separator separator vessel.
Not immediately immediately downstream of the thermo wells or bends bends in in the flowline.
Ideally the sampling sampling point point should should protrude protrude into into the centre of of the flowline, with the mouth facing upstream. However a pipe into the stream is acceptable.
It should should be upstream of any increase in flowline diameter.
Note: The sampling point should not be on the upper half of the flowline cross section, due to any possible free gas. If the sampling point is on the wall of the flowline, then it is preferable that it is on the side, rather than the top or the bottom - due to possible free gas or water in the flowline. Wellhead
237
Upstream of the choke choke manifold manifold as close close to the flowhead flowhead as possible. However, in practice samples are normally taken at the data header, upstream of the choke.
Ideally the sampling sampling point point should protrude into the centre of of the flowline with the mouth facing upstream. However a pipe into the centre of the flowline is acceptable.
Not immediately after a bend in the flowline.
Ensure that that the sampling point is chosen where where the main flow is passing through. Not in any dead legs or alternative flow paths.
17.2 Downhole Sampling
As previously discussed in order to obtain representative downhole samples it is important that the well should be produced at a minimum drawdown (to maintain bottom hole pressure above dew point or bubble point) but still allow continuous flow of produced liquids (oil or water) to surface. Surface readout pressure gauges should most definitely be considered to monitor flowing bottom hole pressure to ensure the draw down is low. Samples should always be collected at a known downhole depth, temperature and pressure. Samplers run into the well on either electric line or slickline should always be run in association with a pressure and temperature gauge (SRO or memory gauge). Samplers should normally be positioned within 300 ft of the perforations, never attempt to take samples below the perforations as the samplers may become stuck in debris and samples recovered may be of mud or water from the rat hole. A gradient survey may be required prior to sampling to locate any oil water contact in the well. The sampler should be placed at the very least 30 ft above the oil water contact to avoid sampling emulsion. Formation-fluid samples can be acquired using one of three main techniques. First, wireline formation tester deployed in open hole can acquire fluid samples and also perform downhole analysis of fluids, ensuring optimal sample acquisition and the possibility of analyzing fluids early in the life of well. These testers provide a cost-effective method of acquiring early fluid samples, with
238
performance now often equal to or above that achievable with the second method, drillstem tests (DTSs). In the past, DSTs, typically designed to test production and investigate reservoir extent, have produced samples with less contamination that openhole sampling. DSTs require early planning and a well completion that withstand production pressure, and can cost much than t han openhole sampling, especially in offshore wells. In a third method, samples can be acquired by wireline tools deployed in a cased, producing well. An important aspect of fluid sampling is analysis analysis of the fluids at reservoir conditions. This helps validate sample quality during the sampling process, but also enables the mapping of vertical variations in fluid properties, allowing interpreters to determine zonal connectivity and define reservoir architecture early in field life. Uncontaminated fluid samples allow accurate measurement of fluid properties both dowhole and at the t he surface. After samples are acquires, they typically are analyzed in laboratories, where they undergo undergo a series of of tests depending depending on what what the client client
needs to
understand. Standard analyses for hydrocarbon samples include chemical composition to C 30+, gas/oil ratio (GOR), density, viscosity, and phase properties such as saturation pressure, bubble point, puor point and stability of asphaltanes. Several measurements can now be performed downhole, using optical spectroscopy to characterize formation fluids under reservoir conditions. These include density, optical density, GOR and chemical composition to C 30+. Laboratory and dowhole fluid measurements both require pure, uncontaminated samples. Contamination occurs when miscible drilling fluid filtrate that has invaded the formation mixes with the formation fluid being sampled. For istance, hydrocarbon samples samples are contaminated by oil-base mud (OBM) filtrate, and water samples are contaminated by water-base mud (WBM) filtrate. To reduce contamination during sample collection, engineers rely mostly on increasing the volume of fluid pumped from the reservoir by pumping longer or at a higher rate. Downhole analysis of contamination level can determine when fluid f luid flowing through the sampling-tool flowline is clean enough to be collected. However, long pumping time increases rig time and associated costs, and may increase the risk of downhole tool sticking. Depending on the reservoir permeability, high pumping rates can cause the reservoir permeability, high pumping rates can cause the reservoir fluid to drop below saturation pressure. If
239
this happens, the downhole samples will not be representative of the reservoir fluid. In the case of unconsolidated formations, high pumping rate may induce sand production. Also, in settings involving high vertical permeability, even long pumping times and increased pumping rates do not clean samples. Fluid-analysis experts have worked to understand and mitigate the effects of contamination on samples. Some methods attempt to derive the composition or GOR of a pure sample knowing the composition of the OBM contaminating the collected sample. However, uncertainties and errors accompany fluid properties estimated in this manner. Researchers have quantified the errors caused by contamination on some measurements. For example, the pressure at which asphaltenes precipitate from solution in crude oil decreases in the presence of OBM contamination by weight caused asphaltene precipitation onset pressure to decrease by 100 to 150 psi
[0.7 to 1.0 MPa]. Thus, measurements on
contaminated samples under-estimate asphaltene-precipitation onset pressure, and may negatively affect flow-assurance and production predictions. These results emphasize the need for extremely low-contamination samples. Downhole Fluid Analysis
In most reservoirs, fluid composition varies with location in the reservoir. Fluids may exhibit gradations caused by gravity or biodegradation, or they may be segregated by structural or stratigraphic compartmentalization. One way to characterize these variations is to collect samples for surface analysis. Another way is to analyze fluids downhole, without bringing them to surface. Downhole fluid analysis (DFA) is emerging as a powerful technique to characterize fluids downhole. DFA helps determine the best intervals for sample collection, if necessary. Analyzing fluid composition while the tool is still in the hole also allows more detailed fluid characterization, because interpreters can modify the fluidscanning program in real time to investigate unexpected results. The ability of the Quicksilver Probe module to supply uncontaminated fluids ensures optimal DFA results, and the faster cleanup time allows several DFA fluid-scanning stations to be conducted efficiently without the long station times associated with conventional sampling. A combination of DFA and sample collection helped a Norwegian operator understand fluids in a well drilled on the Norwegian Continental Shelf.
240
The well was drilled ad a final appraisal before development of an oil field. Because of environmental restrictions, a production test was not planned, so it was critical to obtain uncontaminated samples and fully characterize fluid variation within reservoir. The fluid analysis would be used in the material selection of subsea pipeline and surface facilities, process design and production planning. Because of the high priority to capture representative hydrocarbon sample within miscible contamination, the well was drilled water-base mud (WBM). The Quicksilver Probe tool was run in the 12 ¼ in, and 8 ½ in section to collect samples of gas condensate, oil and formation water, and filled 19 samples chambers from many levels. An example from one of the more challenging zones, sampling oil in a relatively tight zone with mobility of 17 mD/cP, shows hoe the focusing technology results in an uncontaminated sample. Fluid cleanup began with commingled flow first through the guard flowline, then through the sample flowline. After 1,300 seconds, flow is split and focusing is achieved by increasing the flow rate in the guard probe. The real-time GOR detected by the CFA module stabilized at around 2,300 seconds, indicating that the fluid was clean. However, pumping continued, and a sample was acquired at 2,800 seconds. The spikes in the GOR curve indicate the presence of produced fines from the formation, confirmed later when the samples was analyzed at surface. Wellsite analysis showed some sand in the samples, but no detectable level of WBM filtrate. In the same well, the focusing method created optimal conditions for DFA. The spectroscopic analyzers that indicate when fluid in the flowline is pure enpugh to sample also characterized the fluid composition in terms of three component groups: methane (C 1), ethane to pentane (C 2 to C5), and hexane and heavier (C6+). This allows in situ-compositional analysis without collecting a sample and retrieving it to surface.
241
18.0 TEST IN AGGRESSIVE ENVIRONMENTS 18.1 Environmental Concerns Dangerous Substance
During the implementation of a well test a wide variety of hazardous substances may be used. Prior to commencing a well test operation, consideration should be given to each of these substances to ensure that safe handling procedures are drawn up. Steps should be taken to ensure that exposure of personnel to these substances is minimised. Any personnel handling these substances should be aware of the required precautions and associated hazards. These substances must only be handled in strict accordance with specified procedures and by authorised personnel. Typical hazardous substances used during well testing are listed below: Acid additives
Hydraulic oil
Benzene
Cement
Biocides
Diesel
Glycol
Oil based muds
Hydrochloric acid
Crude oil
Hydrofluoric acid
Hydrocarbon gas
Hydrogen Sulphide Scavengers
Scale Inhibitors
Methanol
Mineral Oils
Oxygen Scavengers
Brines
Pipe Dope
Mercury
Toluene
Xylene
Daylight/High Visibility Working
Some operations undertaken during a well test will require good visibility in order to be carried out safely. Moreover, some operations which could be carried out without problems on offshore rigs, may not be possible on onshore rig sites where the overall site lighting may not be so good. Whenever possible, additional 242
explosion proof lighting should be installed at onshore wellsites for testing operations. Generally the following activities will become considerably more hazardous if carried out under low light conditions:
Acidizing and fracturing.
Coiled tubing work.
High pressure pumping.
Igniting flares.
Initial perforating and flowing the well.
Slickline and electric wireline work.
Well kill.
Where possible these activities should be avoided during the hours of darkness. In certain circumstances some of the operations listed above will be required during the hours of darkness. In these cases it is imperative that adequate lighting is provided and that personnel involved in this phase of the test are alert and aware of any increased hazards. Restrictions on work requiring daylight or high visibility should be indicated within the detailed well testing programme. Oil Spill Contingency
An oil spill contingency plan will normally be formulated prior to the drilling of any well. This plan should be written so as to include any possible spillage occurring during well testing. Oil spills associated with testing would probably come from the following:
Unburned oil falling from burner booms, " Fall Out".
Separator rupture disk blowing causing liquid dumping.
Well blowout.
Catastrophic failure of production piping or vessels.
Unburned oil based mud falling from burner booms.
243
Leakage of crude oil storage vessel or tanks.
Rupture of tanks or compartment during crude oil transport.
Each of the above types of oil spill would vary in severity depending on the nature of the test or on the amount and type of oil storage at the well site. The oil spill contingency plan for a particular well should be reviewed prior to writing the detailed well test programme. If necessary, additional section should be added to the detailed programme to cover specific oil spill contingency planning. 18.2 Detection System Breathing Apparatus
Basic breathing apparatus will normally be available on the rig. However, this equipment may be limited in quantity. Therefore, if a well test is to be carried out and H 2S is 'possible' in the well, sufficient breathing apparatus sets must be made available for all personnel on the rig. This breathing apparatus should be self contained and allow at least 20 minutes usage without refilling air bottles. For tests where H2S is 'expected' a full cascade air system should be installed in preference to the use of air bottles. The use of breathing apparatus (particularly air bottles) to carry out normal well testing operations in an H 2S contaminated environment is more physically taxing to workers. Therefore, regular rotation of test crew working in contaminated areas will be required. Spare sets of breathing apparatus should always be maintained in a safe area close to the work sites to enable rapid change over in the event of an equipment failure. Emergency escape air packs found on the rig should only be used for this purpose and not for normal working. Further information on H 2S is contained in section 4.9. However, the advice of a specialist H2S contractor is recommended when planning a sour well test. Gas Detection
244
Gas detection equipment is used during well testing to monitor the area around the test equipment for hydrocarbon gas emissions and to ensure that H 2S concentrations in the atmosphere are safe for working. Portable gas detection equipment is used for both monitoring contaminants in produced gases and to ensure a safe working environment exists before carrying out work in enclosed areas. Permanent gas detection equipment should be located in areas where releases of gas are likely i.e. on rig floor, in the vicinity of the test separator and close to the choke manifold (if this is located away from the rig floor). Permanent gas detectors should be located so that normal operations will not cause the detectors to be activated, however they must be located such that they will detect quickly any abnormal quantities of gas. For this reason the detectors should be located close to the equipment but away from "dead air" areas. During the well test flow periods, regular sampling of the produced gases should be carried out. These tests are carried out using various hand held analysers, either disposable, or re-usable. Gas measurements are usually in parts per million (ppm) by volume of H 2S and percentage of CO 2. These measurements are particularly important on wildcat wells to ensure that test equipment is not exposed to unsafe operating conditions. Regular measurements will ensure that personnel are continuously aware of any changes required in the safety procedures, should high concentrations of H 2S be present. The measurements obtained from portable gas detection devices are generally regarded as a more qualitative than quantitative. To obtain quantitative analytical results more sophisticated equipment such as a gas chromatography may be used. However, for the purposes of establishing a safe working environment prior to entry into any confined space e.g. test separator, gauge tank, mud pits etc. these portable devices are adequate. Work permits should be obtained where applicable and breathing apparatus must be available at the work site. Hydrogen Sulphide
Hydrogen sulphide gas is extremely toxic and in relatively low concentrations can quickly cause unconsciousness and death.
245
At concentrations in the range of 1 - 30 ppm it can easily be identified by its characteristic smell of rotten eggs. However, a noticeable odour can be detected even at concentrations as low as (0.01 ppm) At higher concentrations the smell becomes sweetish and at about 150 ppm olfactory paralysis occurs when the sense of smell can no longer be relied upon. Table 18.1 provides a summary of the hazards and precautions to be taken if H 2S is expected. Each operating centre should produce a detailed H2S procedures document, specific to the rig, wellsite location and operating environment. The procedures described below are recommended for use in the testing section of these H2S procedures document.
246
HAZARDS TO LIFE
PRECAUTIONS/TREATMENT
1. Highly
Monitoring H 2S concentration with detectors during flow 2. At low concentration dulls the sense If H2S levels in the gas stream reach 10 of smell ppm the test will have to be terminated unless sour service equipment is being used. 3. Higher concentrations- paralyses When testing sour wells (with sour the olfactory nerves at about 150 ppm service equipment) inform the drilling supervisor if H2S concentration in the well stream exceeds 20 ppm 4. Can be masked by other odours If H2S is detected around the rig, locate (such as Butane and Propane) and repair leaks. If H2S persists, terminate test and bullhead fluids back into the formation. 5. Heavier than air (S.G. 1.185) – it First treatment for those affected by can accumulate H2S 6. Flammable gas (burn with a blue Remove person to fresh air flame) Maximum allowable concentration Resuscitate if required 10ppm (Safe up to 8 hours) 100ppm – May sting eyes and throat Oxygen may help 200ppm – Kills sense of smell rapidly, sting eyes and throat 300ppm- Severe headache, eyes and lungs affected – over 1 hour exposure may cause death 500ppm – Loose sense of reasoning and balance. Respiratory paralysis in 2 – 15 minutes. Casually will need prompt artificial resuscitation 700ppm-Breathing will stop and death will occur if not rescued promptly. Requires immediate artificial resuscitation 800ppm- Fatal after few minutes Requires immediate artificial resuscitation 1000ppm – unconscious at once. Permanent brain damage may result unless rescued promptly
Tab. 18.1: Hydrogen Sulphide- A summary of hazards and precautions
247
Fire Fighting Equipment and alarm system
Fire fighting equipment should always be available at the rig site. Prior to commencing a well test this equipment should be thoroughly checked to ensure it is in good working condition. Fire extinguishers of the appropriate type should be moved to within easy reach of the test separator area. All testing personnel should be trained in the use of hand held fire extinguishers. The rig fire crew should be briefed on the layout of the test equipment and the location of any portable fire fighting equipment in the area. For onshore testing the local fire department, where appropriate, should be made aware of the nature and timing of test operations. 18.3 Prevention Measures General Provisions
Certification, Service Testing and Safety Review All equipment purchased, or rented to carry out a well test should be fit for purpose. This means the safe working limits in terms of pressure, temperature, flowrate, nature of services and tensile/compressive loading should not be exceeded during any part of the well test operation. Well testing operations are often planned with very little information available about the reservoir and these tests carry the greatest risk. The worst case scenario cannot routinely be designed for and therefore a risk analysis, safety review or HAZOP should be undertaken in such circumstances. This will determine if worst case scenario is required in the well test design. It is recommended that all equipment supplied for a well test should have current certification available prior to being sent to the rig. Full pressure testing and function testing of equipment should be carried out prior to and on arrival ot the rig. These tests should be witnessed by a competent person. Records of all certification and preservice tests should be retained on the rig until completion of the test. For North Sea tests it is a requirement that the equipment layout and P. & I.D. meets with the approval of the rig contractors certifying authority e.g., DNV, ABS,
248
Lloyds. Generally, this will be organised by the surface testing contractor. The equipment should be placed such that it complies with any zoning requirements. Although this may not be a regulatory requirement elsewhere, it is recommended that this be considered as part of a safety review. It is strongly recommended that a HAZOP or safety review be conducted to ensure that the equipment and procedures are fit for purpose. The extent of the review will be dependent upon the complexity of the operation. At the very minimum, a safety review orchestrated by the surface testing contractor should be undertaken. HP/HT tests for example, would justify a formal HAZOP. These Safety Review/HAZOPS should follow a format. Restricted Access
During certain parts of the well test operation it is important to restrict access of personnel to specific areas of the rig or well site. This may require using signs and barriers and making PA announcements. Principally the times when restricted access will be required are as follows:
Offloading areas while loading or backloading equipment.
The rig floor or catwalk while making up perforating guns.
The rig floor while running the test string.
The rig floor and test separator area etc. while pressure testing.
The rig floor, derrick and cellar, separator area, burner booms and flare area during perforating and flow periods.
The rig floor, derrick and cellar while killing the well and pulling the DST string.
These restrictions are related to the type of work being carried out and details should be given in the detailed well test programme. Typically these restrictions would only allow access to personnel directly involved with the procedures. The start and end of operations requiring restricted access should be announced where possible by using the rig PA system. The status of existing restrictions should be retransmitted over the PA system at crew changes, so that on-coming personnel are made aware of the situation. Safety Meetings
249
Safety meetings should be carried out prior to each critical phase of the well test operations. These meetings are required to inform all relevant personnel of the work being carried out, specific hazards and hazardous areas. Safety meetings are also a good opportunity to ensure that all personnel are aware of their individual responsibilities. Typically safety meetings would be held at the following times.
Prior to picking up the test string and perforating guns.
Prior to perforating or initial flowing of the well.
Prior to carrying out a stimulation treatment.
Prior to killing the well and pulling the DST string.
Additional meetings may well be required depending on the nature of the test and in certain circumstances these may have to be held at the beginning and end of each tour with both day and night crews being represented. Some points, which may be discussed at a safety meeting are detailed below. Pre-Test Considerations A pre test meeting should take place with all supervisory personnel present and all the points below which are applicable to the operation should be addressed.
All breathing apparatus should be checked for serviceability.
All personnel should be trained in use of B.A. sets if H2S is expected.
Carry out an H2S drill, if H 2S is expected.
All fire fighting appliances to be inspected.
Fire pumps to be functioned and the system pressurised.
Fire drill to be carried out in test area.
Fire appliances to be positioned next to the test spread.
Gas detectors to be function tested.
Explosion meter to be function tested.
Lifeboat engines to be function tested.
250
Lifeboat launching equipment to be function tested.
Radio and telephones to be function tested.
Life-jackets and survival suits checked.
Drill floor sprinkler system to be function tested.
List of safety muster points to be posted on the notice boards and reviewed at the
pre-test safety meeting.
Firedrill and abandon rig drill to be held prior to testing.
Schematic showing hazardous areas to be posted.
ESD system to be installed and tested. Personnel to know location and function.
All rig cooling systems to be function tested.
Kill fluid to be prepared in suitable quantities. Kill lines and pump to be tested and manifold correctly lined up to kill wing.
All diesel units to be checked for spark emissions.
All pressurised bottles to be stowed away from hazardous areas.
All unnecessary electrical appliances to be disconnected.
All annulus monitoring sensors to be checked and purged.
Escape routes to be clearly marked and kept clear of obstructions.
Personnel Briefing: Overleaf on a single page is a personnel briefing that may be copied and distributed to personnel at the well site prior to holding the pre-test safety meeting. 18.3 Prevention Measures Prevision in Presence of H 2S
251
The risk of encountering H 2S must be assessed from available information relating to the current well and other wells in the area. Danger signals must remain displayed while also carrying out production testing where a presence of SO 2 greater than 5ppm is expected (due to the combustion of layer fluids). H2S Emergency Provisions In the event that the occurrence of H2S is a possibility, provisions must be made as follows: 1.
Detection - any H2S fixed or portable detectors which may be required in addition to those already on board.
2.
Wind Direction - indicators such as pennants or socks will be positioned in at least 4 locations such that the movement of H 2S can be foreseen and its impact on escape
3.
routes/ systems and support vessels/helicopters can be assessed.
4.
Any person working on a rig that is drilling in a known H 2S area or which encounters H2S while drilling must be clean shaven.
The Production Superintendent and representatives from the Drilling and Safety Departments must inspect the rig and ensure that a safety meeting has been held before the well test operations start. General Procedures The requirements below must be reconciled with the Drilling Contractor‟s onboard
equipment and emergency procedures to ensure that they cover all the points addressed. In the event that there are deficiencies they must be dealt with to ensure that the overall provisions are at least equivalent to the requirements of this manual. Detection Detection is accomplished by smell, mud analysis, fixed detectors and hand held detectors.
252
Only the fixed detectors will automatically provide an alarm, all other detection methods require personnel to raise the alarm. Brine/Mud Analysis In the event that brine/mud analysis shows the presence of H 2S, the logging engineer will immediately raise the alarm indicating the level of gas. He must be provided with adequate means of communication. Personal Monitor Any detection of H2S by personal monitor must be reported immediately to the senior drill floor personnel or Production Test Supervisor during well test operations and the OIM, giving the location of detection and the concentration measured. Fixed Detectors Provision of the fixed detectors must be such that detection of H2S will result in a suitable alarm being raised in all areas manned during drilling or well test operation, i.e. in the mud treatment room and in the control room, and also give the detector position and the concentration detected. Safe Breathing Areas (offshore rigs) The OIM will designate at least two Safe Breathing Areas (SBAs) of which one will be in the open air upwind of any incident. The second SBA will be inside the accommodation in the gallery/cinema/recreation area. An H 2S detector will be provided in the inside SBA and must be switched on when the alarm is given. If deemed necessary, a second open air SBA will be designated to ensure that at least one SBA will be upwind of any incident. H2S Detection While Tripping Prior to pulling out of the hole, circulate the brine/mud system. Circulating All drill floor and mud room personnel will wear SCBA and be masked up (f ireman‟s sets or tied into the cascade system) immediately. At the same time the mud/gas separator (degasser) will be started and all non essential personnel will be warned to stay away from the drill floor and mud treatment areas.
253
Mud logging personnel will inform the Tool pusher and the OIM when the trip gas is up and when the H 2S level falls below 10ppm. Logging When pulling out of the hole, all tools and cable must be washed with scavenger and spray inhibitor. Persons handling repeat formation tester (RFT) samples/chambers must wear SCBA until the chamber has been vented and purged. Flow Testing During this phase (time from first opening of test tools until tools are recovered to surface) H2S will be produced to the surface for the first time with consequent increase in risk. To counter this the following precautions are required: 1.
Before well test operations take place:
2.
The onshore Drilling Manager/Drilling Superintendent, Snr Production Engineer, Workover Superintendent and Safety Superintendent in consultation with the Rig OIM, the offshore Eni- Agip Senior Drilling Supervisor and Production Test Supervisor must agree whether or not it is necessary to specify any in stream concentration of H 2S at which the well test crew and other essential personnel involved in well test operations must mask up.
3.
A safety meeting prior to opening the well must be held to inform all personnel of the increased risk of the presence of H 2S.
4.
All testing equipment and systems must be capable of withstanding the effects of H2S.
5.
All critical activities such as the first opening of downhole tools must be performed in daylight.
6.
All personnel considered by the OIM to be non-essential must be taken off the rig before the start of the test and remain off until after the end of the test.
7.
During the testing period, all off duty personnel shall be restricted to the accommodation area and their movements will be controlled by the OIM.
254
8.
At the production of first hydrocarbons to surface, essential personnel will all wear SCBA and be masked up. Masks will be worn until the level of H2S being produced has been established at the choke or at the separator.
9.
In stream H2S levels will initially be monitored every 10min for changes, initially, and thereafter at periods agreed by the OIM, Production Test Supervisor and H 2S technician.
10.
When H2S is present in the flow stream, the well will be shut-in if the wind speed is less than 5 knots. In any event, it is the responsibility of the OIM to decide if the wind speed or direction presents a hazard which requires the suspension of testing.
11.
Testing personnel must wear SCBA and mask up prior to operating or performing work on equipment or systems which have contained H 2S, e.g. changing chokes, operating flowhead valves, using bubble hoses, taking separator samples, etc.
12.
No open tanks will be used for collecting flow products. Surge tanks and separators will be equipped with vent/overflow lines which discharge at the flare.
13.
Background levels of H 2S will occur from various sources such as flare residue, valves, flanges, couplings etc. This level must be monitored for increases so that preventative actions can be taken.
14.
The installation must be monitored for the presence of sulphur dioxide (SO2) using portable monitors.
15.
When the test tool retrieval gets to within five stands of tubing from the first test tool, i.e. the reverse circulating valve, all rig floor personnel will wear SCBA and be masked up until the testing string has been broken down, sample chambers have been emptied and purged and slip joints stroked.
255
DESCRIPTION
QUANTITY
Equipment for the constant of H 2S concentration consisting of: Unit with control panel and tool indicator of the H2S concentration with range capacity of 0-50 ppm and two levels (10 ppm – 20 ppm) Sensor with short response time as per market availability Portable detectors to measure H 2S in the atmosphere (either manual or electronic) Colorimetric vials for H 2S 10 vials package: 1-200ppm 10 vials package: 50-500ppm 10 vials package: 100-2000ppm Colorimetric vials for SO 2 10 vials package: 1-200ppm 10 vials package: 20-200 ppm 10 minutes Automatic Positive Pressure Escape Breathing Apparatus
Dependent on rig type Dependent on rig type 3 manual 2 electronic 10 5 5 5 5 30 for land rig 120% of the rig and supply vessel personnel for off-shore rig 15
30-45 mins Self Contained Breathing Apparatus for land rig Extra cylinders for Breathing Apparatus for land 45 rig 30-45 mins Self Contained Breathing Apparatus 120% of emergency team for off-shore rig Extra cylinders for Breathing Apparatus for off- 3 spare bottles per breathing shore rig apparatus AMBU type reanimator 2 Battery operated portable explosimeter 2 Wind sleeve 2 Two tone alarm hooter 2 Walkie talkie completed with batteries and 6 battery loader Electric lamp (explosion proof type) 6 Alarm flashlight 2 red 2 yellow Gas garret Train test Kit or Hatch Kit 1 (in sour area) H2S Scavenger for mud 30 kg/m3 of mud Fan 3
Tab. 18.2: Constant H 2S Detection for Land Rigs
256
Tab. 18.2: Constant H 2S detection for land rigs
257