1417 Pipeline Surge Analysis Studies Atmos_PSIG

PSIG 1417

Pipeline Surge Analysis Studies Garry Hanmer - ATMOS International Limited Susan Bachman and Gregory Lind – Enterprise Products

Copyright 2014, Pipeline Simulation Interest Group This paper was prepared for presentation at the PSIG Annual Meeting held in Baltimore, Maryland, 6 May – 9 May 2014. This paper was selected for presentation by the PSIG Board of Directors following review of information contained in an abstract submitted by the author(s). The material, as presented, does not necessarily reflect any position of the Pipeline Simulation Interest Group, its officers, or members. Papers presented at PSIG meetings are subject to publication review by Editorial Committees of the Pipeline Simulation Interest Group. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of PSIG is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, Pipeline Simulation Interest Group, P.O. Box 22625, Houston, TX 77227, U.S.A., fax 01-713-586-5955.

Abstract Pipeline Surge Analysis Studies require the hydraulic simulation of pressures and flows in fluids caused by the transient operations of pumps and valves. Pressure surges can cause significant damage to pipelines producing pipeline leaks, cracked pump casings, contamination and environmental damage. Without adequate surge protection this will result in significant downtime in process plants and distribution systems, and the reduced life expectancy of the pipeline. This paper discusses pipeline surge analysis and looks to address the challenge of efficiently reviewing entire pipeline networks for pressure surges in order to comply with the Department of Transport regulatory requirements. To reduce the effort required in conducting pipeline surge analysis studies for pipeline design, operational changes, and product changes, a surge analysis program has been developed to automate the procedure from the scheduling of simulation scenarios to the creation of the surge analysis report.

Why Surge Analysis? The consequences of pipeline failure may be catastrophic, and strict regulatory requirements are in place to ensure the safe operations of pipelines. The UK Pipeline safety regulations state that: “The pipeline operator shall ensure that no fluid is conveyed in a pipeline unless the Safe Operating Limits (SOL) of the pipeline have been established and that a pipeline is not operated beyond its SOL.” (Ref 2) This pipeline safety regulation can be found (with words to the same effect) globally in numerous pipeline safety regulations. A few examples are: •

• •

U.S. Department of Transportation's Pipeline and Hazardous Materials Safety Administration British Standard Code of Practice for Pipelines European Harmonised Standard Petroleum and Natural Gas Industries – Pipeline Transportation Systems

The recognised standards often allow short excursions of pressure above Maximum Allowable Operating Pressure (MAOP). The pipeline may therefore, for limited periods, see pressures above MAOP and still be in code. The SOL, above which the pipeline is not allowed to run under any circumstances, is therefore higher than MAOP. The SOL for the maximum pressure is typically 10% to 15% larger than the MAOP limit, depending upon

2

Garry Hanmer, Susan Bachman, Gregory Lind

local pipeline safety regulations. Pipeline failures are rare occurrences, but have the potential to cause extensive property damage, loss of life and shut down of pipeline facilities for extended periods of time. The causes of pipeline failure can be internal or external. Internal causes include events such as pressure surge and material defects. External causes include events such as earthquakes and third party intervention. Pressure surge in a pipeline system can produce pressures in excess of the allowable maximum and minimum pipeline pressures. High pressures can damage pumps, valves and other pipeline objects, along with the potential to rupture the pipeline. Low pressures can lead to pipeline collapse, and cavitation. Vapour cavity closures can produce high shock pressures. Transient pressure surges occur in pipelines because of sudden changes in the fluid flow velocity. These changes in flow velocity can occur due to operations such as valve movements, pump power failures, and pump start-up. Pressure surges can occur in all fluid pipeline systems resulting in pipeline fatigue and pipeline failure. Pressure surge may be avoidable through sufficient pipeline assessments and protection. The consequences of a pipeline failure can be tragic. On Saturday 1 June 1974, a 20 inch pipeline in Flixborough, UK ruptured. Within one minute of the rupture, approximately 40 tonnes of cyclohexane leaked from the pipeline, forming a vapour cloud 200 metres (650 feet) in diameter. This vapour cloud ignited resulting in injuries to thirty-six workers, and fatalities to twenty-eight workers. Offsite a further fifty-three injuries were reported and varying levels of damage to property were registered up to three miles away. No one

PSIG 1417

escaped from the control room, where all eighteen personnel were killed (Ref 1). The rupture occurred due to an over pressurization of the pipeline. An assessment of the technical failings concluded that a plant modification had occurred without a full assessment of the potential consequences, where only a limited amount of calculations were conducted on the integrity of the pipeline. No pressure testing was carried out on the modification to the installed pipework. Detailed pipeline assessments may be timeconsuming, requiring large amounts of calculations to be computed for each pipeline section. This is due to the large variety of different operating conditions which must be analyzed within the surge analysis study, where each valve closure, pump trip and fluid change must be considered. Pipeline simulation packages can be used to run pressure surge analysis on pipeline sections to provide accurate transient surge predictions and analysis. A surge analysis application has been developed to directly interact with the pipeline simulation tool in order to schedule dynamic simulation scenarios, run transient simulations, conduct initial analysis of the simulation results and create a surge analysis report. This whole procedure is automated into a single step through the utilization of the surge analysis program, allowing the surge analysis study to be conducted in a much timelier manner, making the surge analysis study a much more efficient process.

What Causes Surge? Pressure surge refers to the pressure produced by a change in velocity of the moving fluid that results from events such as the shutting down of a pumping station or pumping unit, unstable controls, oscillations in tank levels, vapour pocket collapse, the closure of a valve, or any other sudden blockage

PSIG 1417

Pipeline Surge Analysis Studies

of the moving fluid. These pressure surges may occur in all fluid pipeline systems and can result in pipeline fatigue and pipeline failure. The effects of valve movement, pump trip, and pump start-up will be considered below. Valve Movements One of the most common causes of pipeline pressure surge is the movement of a valve. Any valve movement causes pressure waves to propagate through the pipeline system. The magnitude of this pressure wave depends on several factors. These factors include the type of valve, how quickly the valve is moved, the hydraulic properties of the system, and the elastic properties and restraint of the pipeline. A sudden valve closure, such as an emergency shutdown valve closure with a short valve closure time creates the same behaviour as a slamming check valve. Check valves can cause large pressure surges if the flow back through them can occur before the valve closure is complete. This may occur following a pump failure where the pump discharge pressure drops rapidly resulting in the check valve closure. Most modern check valves do not slam. A spring or weight are commonly used to close the valve at the instant when the flow ceases, while other check valves close slowly, regulated by a damping mechanism, to bring the flow to rest gradually. Even for these check valves there may be some elastic energy in the system which will cause a pressure surge. Therefore it is important to ensure that the valve either closes quickly before a reverse flow can become large or closes slowly over a time interval that is considerably greater than the critical time of closure as given by Equation 4 below. Otherwise a high pressure could occur at the time of the valve closure.

3

Pump Trip Pipelines with a large static lift and where the pipeline elevation profile immediately downstream of the pumps rises rapidly can observe severe pressure surge following an event such as a pump power failure. When a pump stops, the pressure drop propagates down the pipeline. This pressure drop may result in cavitation and cavity closure shocks. A flow reversal may occur, resulting in over pressurisation of the system if the transient is not controlled correctly. This over pressurisation tends to occur in the vicinity of the pumps. The magnitude of a pressure surge depends upon the fluid compressibility, its density, and the magnitude of the change in the flow velocity. The magnitude of the pressure surge caused by a sudden change in fluid velocity can be approximately expressed by Joukowsky's Law (Equation 1), which is based on the total conservation of kinetic energy of motion into pressure head. Pump Start-up When a pump starts up a positive pressure surge is created in the downstream pipeline section. The magnitude of this pressure increment depends upon how quickly the velocity is increased when the check valve is forced open and the fluid in the line begins to move. Pump start-up pressure surge typically occurs when the pipeline is not completely primed for the startup, such as following the tripping of a pump in the event of a power failure. The fluid may come to rest in the pipeline following the possible formation of vapour cavities. When pumping is resumed these vapour cavities collapse resulting in transient pressures developing. (Ref 3)

What are the Effects of Surge? Pressure surges within pipeline systems can lead to severe damage and the potential failure of pipeline equipment. Pressure surge damage can occur due to pressures in excess of the maximum allowable pipeline pressures, resulting in damage to pumps, valves and the pipe. Damage may also occur due to pressures below the minimum allowable operating

4


pressures, where low pressures can result in cavitation and pipeline collapse. Failures can occur under a number of different situations, including: • • •

Failure of static components through fatigue, erosion or corrosion Failure of dynamic components leading to high fatigue loads on other components Failure of the piping system due to extreme pressures or temperatures

Transients in pipeline systems can cause the local absolute pressure of the fluid to approach its vapour pressure. At the vapour pressure of the fluid, gases begin to come out of the solution. If a pipeline transient results in a drop in pressure which is severe enough to cause the pressure to reach the vapour pressure, then the fluid boils (cavitates) forming pockets of un-dissolved gases and vapour (column separation). When the local absolute pressure increases, the cavitation bubbles collapse rapidly and violently. During the collapse of the cavitation bubbles, the vapour bubbles will transform themselves back into a liquid state. There is a large volume change during this transformation, where the collapsing bubbles release a large amount of energy resulting in pipeline damage. Damage due to cavitation can include, material fatigue, component damage and cavitation erosion.

PSIG 1417

level of importance. A surge mitigation device or procedure which does not perform when required may result in catastrophic consequences. The following devices and procedures will be analyzed in the following sections; valve movements, relief valves, surge tanks, increased pipeline diameter and increased pipeline wall thickness. Valve Movement The impact of valve movements varies significantly between different types of valves. Gate valves for example must be nearly closed before it generates enough head loss to decrease the velocity by a significant amount. The sudden closure of a valve causes an increase in pressure head to occur at the upstream location and a decrease in pressure head ( Δ H ) in the downstream location, which propagate at a speed (c). Using the linear momentum equation the change in pressure head can be calculated:

ΔH = −

c * ΔV ⎛ V ⎞ ⎜1 + ⎟ g ⎝ c⎠ (Equation 1)

In most rigid pipe situations V/c is less than 0.01 and Equation 1 is therefore often reduced to:

ΔH = −

c ΔV g (Equation 2)

When the vapour pressure of a homogenous fluid such as water is reached, the entire fluid begins to change phase, resulting in the formation of large vapour cavities. For non-homogenous fluids such as a hydrocarbon, only the light ends (such as condensates) with low specific volume are affected. ‘Figure 9- Fluid Vapour Pressures’ shows the vapour pressures of Ethane, Iso-Butane, Propane, Iso-Pentane, Pentane and Hexane.

How to Mitigate Surge? There are many devices and procedures which may be used in order to mitigate unacceptable levels of pressure. Surge mitigation is a safety critical requirement and should be treated with the highest

Where: Δ H = Pr essure Head Change Δ V = Velocity Change c = Wavespeed g = Gravity

From this equation it can be seen that a change in velocity results in a change in pressure head. The calculation depends upon the wave speed as calculated in (Equation 3). There are several equations which can be used to calculate the speed of the pressure wave. These can

PSIG 1417


vary depending upon fluid types and pipe properties. For an approximation of thin-walled pipes the wave propagation speeds can be calculated using:

⎛ ⎡ 1 D ⎤⎞ c = ⎜⎜ ρ ⎢ + φ ⎥ ⎟⎟ K Ee ⎣ ⎦⎠ ⎝

−

1 2

(Equation 3) Where:

c = Wavespeed ρ = Density K = Bulk Modulus D = Pipe Diameter E = Modulus of Elasticity e = Wall Thickness φ = Re stra int Factor The wave will subsequently reflect from upstream and downstream and may then result in excessively high or low pressures on either side of the pumps. The wave reflection time can then be calculated using the following equation:

T=

2L c

5

the pressure immediately upstream of a valve. A linear valve curve was used for the purposes of generating this data. The figure shows valve closure times of one second (red), five seconds (blue), ten seconds (pink), thirty seconds (green), and sixty seconds (orange). The increasing valve closure time in each case shows a decreasing peak pressure. The simulated fluid velocities for each of these valve closure times within ‘Figure 3- Upstream Velocity When the Valve’ shows the velocity is reduced at lower rates with the increasing valve closure time, resulting in the lower pressure. Figure 3 shows valve closure times of one second (red), five seconds (blue), ten seconds (pink), thirty seconds (green), and sixty seconds (orange). Relief Valve Relief valves allow for fast acting pressure relief within the pipeline. Relief valves open when a predefined pressure is exceeded and range from inexpensive and simple spring loaded discs to more expensive and complicated systems designed to operate within milliseconds. The purpose of relief valves is to provide an escape for the flowing fluid so that a sudden change in velocity and consequent change in pressure do not occur.

Thus according to (Equation 4), the configuration given in ‘Figure 1- Single Pipe’ would show conditions at the supply occurring L/c seconds after the conditions at the demand.

‘Figure 4- The Effect of a Relief Valve’ compares the simulated upstream pressure of a one second valve closure on two similar networks. The only difference between these two networks is the presence of a relief valve in one network immediately upstream to the closing valve. The red trend shows the pressures without the relief valve, and the green trend shows the pressures with the relief valve. It can be seen that the chosen relief valve has a significant effect on lowering the peak pressure at the upstream location of the closing valve.

The effects of the pressure head change can be demonstrated by varying the valve closure time. ‘Figure 2- Upstream Pressure when the Valve is closed within Different Time Periods’ shows the simulated effects of different valve closure times on

Surge Tanks Surge tanks can be used to mitigate both high and low pressures. They may act as temporary storage devices for excess liquid which has been diverted from the main pipeline flow. This diversion allows

(Equation 4) Where: T = Wave Reflection Time L = Pipe Length c = Wavespeed

6


for a more gradual change in velocity in the pipeline and a reduction in the magnitude of transient pressure waves. Surge tanks may also be used to supply liquid to the pipeline to prevent excessive deceleration and low pressures. Surge tanks can be used to act as damping devices on a pipeline where velocities go back and forth frequently. Increased Pipeline Diameter Increased pipeline diameters may be included in the pipeline design to reduce potential surge pressures. This is done by reducing the fluid velocity resulting in a reduced change in momentum to bring the fluid to rest.

Increased pipeline diameters may also have a negative impact due to the reduction of the frictional damping of the pressure fluctuations. The following formula can be utilized to calculate the pipeline design pressure:

P=

2* S *t * F * L * J *T D (Equation 5)

Where: P = Design Pressure S = Specified Minimum Yield Strength t = Wall Thickness D = External Pipe Diameter F = Design Factor (Table1) L = Location Factor (Table1) J = Logitudinal Joint Factor (Table 2) T = Temperature Derating Factor (Table 3)

The calculated design pressure observed when varying external pipe diameter can be seen in ‘Figure 5- External Diameter Vs. Design Pressure’. This figure utilizes the equation above and shows that as the external diameter is increased, the design pressure decreases. An increase to the pipe wall thickness may be used to balance the reduced

PSIG 1417

design pressure produced by the increased diameter. Note that ‘Figure 5- External Diameter Vs. Design Pressure’ assumes that all other factors within the equation remain constant, for example as the pipeline diameter is increased, the pipe wall thickness remains constant. Pipeline Wall Thickness A stronger pipeline may be necessary where other surge mitigation techniques are not possible. Stronger pipelines may be achieved through increased wall thickness for example.

Increased pipeline wall thickness is a more costly method in terms of initial capital, although it does not require the same level of further maintenance and testing as other mitigation methods require. (Equation 5) may be used to show the effects of an increased pipeline design pressure in relation to pipeline wall thickness. ‘Figure 6- Pipe Wall Thickness Vs. Design Pressure’ shows the calculated effects on the design pressure to a variation in pipeline wall thickness. Note that this figure assumes that all other factors remain constant and the only variable is the pipe wall thickness.

How to Perform Surge Analysis? When analyzing the pipeline for pressure surge scenarios, it is important to ensure that all of the potential hazards and threats to the pipeline which result in pressure surge are addressed. Each pipeline will be unique in this respect; however the following is a sample of scenarios which should be considered: 1. What if power fails to the motors driving the pumps? 2. What if the power fails to the motors driving the pumps? 3. What if the pump delivery valve closes in a given number of seconds? 4. What if one pump trips and another keeps running?

PSIG 1417


5. What if a pump is restarted within a given number of seconds after being tripped? 6. What if a control or emergency shut-down valve is closed rapidly? 7. What if an operator opens/closes a valve too quickly? 8. What if a given pipeline component malfunctions? 9. What if the demand on the system is increased? Should there be any changes to the pipeline design or operating conditions; the pressure surge analysis should be conducted again on all parts of the pipeline system. This pressure surge analysis should include all pipeline and operating condition modifications and should monitor the effects of these modifications on all parts of the network and not just the modified section of the pipeline. This analysis should ensure that the design data is correctly input and attempts should be made to ensure that the design data is reliable. Unreliable design data can have a significant impact on the reliability of the analysis. This design data includes items such as, flow rates, component operating characteristics, material specification and fluid properties. An Example Surge Analysis The procedure for a manual surge analysis study utilizing some of the equations discussed above will be followed for a simple pipeline. A valve closure at the outlet of a short crude oil pipeline will be studied. There is a safe operating limit on the pipeline equal to a head of 1000 meters (3280.83 feet) for the crude oil which will be transported in the pipeline. Table 4 details the pipeline properties which will be used for the surge analysis study calculations.

In order to calculate the Joukowsky’s head, the wave propagation speed and the fluid deceleration must first be calculated. The wave propagation speed can be calculated as follows:

7

⎛ ⎡ 1 D ⎤⎞ c = ⎜⎜ ρ ⎢ + φ ⎥ ⎟⎟ ⎝ ⎣ K Ee ⎦ ⎠

−

1 2

⎛ 914.4 ⎡ 1 ⎤⎞ c = ⎜⎜ 840 ⎢ + *1⎥ ⎟⎟ ⎣1.3E9 2.0E11 *12.7 ⎦ ⎠ ⎝

−

1 2

c = 1026.76 m / s = 3368.64 ft / s In order to calculate the fluid deceleration it is assumed that the fluid velocity will decrease linearly over the valve closure time. This assumption may result in inaccuracies within the calculated values. The magnitude of the inaccuracies will depend upon the actual valve type being used on the pipeline.

ΔV = −

q π * r2 (Equation 6)

Where: ΔV = Fluid Velocity q = Flow Rate r = Pipe Radius

This gives: ΔV = −

5 π * 0.4445 2

ΔV = − 8.06 m / s = − 26.44 ft / s

Joukowsky’s law may now be used to approximate the head increase for the given valve closure:

8


ΔH = −

c ΔV g

ΔH = −

PSIG 1417

c dV g dt (Equation 7)

ΔH =

1026.76 * 8.06 9.81

ΔH = 843.52 m = 2767.45 ft Before this value can be compared to the pipeline design pressure, the initial pipeline head at the location of the valve closure must be added to the Joukowsky’s head. This will give the maximum head value as:

Max Head = Initial Head + Joukowsky Head Max Head = 597.02 + 843.52 Max Head = 1440.54 m = 4726.17 ft It can be seen that the calculated head is above the design head for the required fluid and there is now a requirement for a surge mitigation procedure or device. This can be done be controlling the valve movements, such as the manipulation of the valve closure time. Joukowsky’s law assumes that the valve closes instantaneously. A modification of this law can be made for short pipes to include the valve closure time, where the wave reflection time given in (Equation 4) is sufficiently less that the valve closure time. The 500 meter pipeline given in Table 4 was used to demonstrate this equation, and compared against simulated data. An approximation for valve closure times of 1 second, 2 seconds, 3 seconds, 4 seconds and 5 seconds can be given by the equation:

Rearranging this equation to give the required valve closure time; an approximation of 2.78 seconds ensures that the pipeline pressure does not rise to unacceptable levels following the valve closure with a safety factor of 10%. This equation assumes a linear fluid deceleration for the entire valve closure time, which is an unrealistic assumption. For a gate valve for example, only the last 2-5% of the valve closure motion is critical for determining the maximum pressure and different valve types will produce different results. The length of the pipeline section immediately upstream of the closing valve (to the next upstream constraint or boundary condition) will also have a considerable effect on the maximum head, which is also not taken into account. As shown above, although it is possible to perform simple surge analysis based on various equations, the use of an accurate hydraulic simulation software is essential for pipeline design, expansion study and operations planning. A few benefits of hydraulic simulation analysis are: •

•

•

Accurate assessment of pressure surges caused by different operating scenarios, based on the exact pipeline material, dimensions, fluid properties, equipment type and location Effective transient simulations and analysis of emergency situations such as power failure, and equipment malfunction Timely analysis and automatic report of a large quantity of surge scenarios

How to Analysis?

Automate

Surge

To carry out surge analysis, the entire pipeline network must be included for the effects of pressure surge, including any branches. Hydraulic

PSIG 1417


simulations are an effective way of conducting this analysis, where the simulation may be used to analyze the entire pipeline network as a whole or various subsections of the network. A hydraulic simulation software that can accurately model the effects of the pressure surge should be used. The following elements should be considered when selecting a hydraulic simulation software for surge analysis: •

• • • • •

Variable knot spacing allows for accurate analysis of the pressure surge, minimizing interpolation and errors Variable time steps allow for accurate analysis of the pressure wave propagation Reverse velocity modelling allows dynamic check valve modelling Pump modelling allows accurate spin –up and spin-down times Control logic for accurately simulating pipeline control systems during surge events Valve characteristic inputs allows for accurate valve closure modelling

Surge Analysis Tool While a hydraulic simulation software is effective at identifying pressure surges, the effort required in conducting surge analysis studies can be extensive and time consuming. This is due to the large variety of operating conditions that must be analyzed. This issue is only amplified for large pipeline networks and poses a significant challenge.

To reduce the effort required in conducting pipeline surge analysis studies for pipeline design, operational changes, and product changes, a surge analysis program has been developed to automate the procedure from the scheduling of simulation scenarios to the creation of the surge analysis report.

9

The surge analysis program has been exclusively designed for the purpose of surge analysis, simplifying and automating the process of providing the necessary submissions to the regulatory authorities. The surge analysis program will analyze all selected parts of the pipeline sequentially without any further input from the user. This can be an extremely efficient method for analyzing very large pipelines and networks, allowing the users to conduct other tasks while the calculations are being conducted by the software. The user will receive a report in their selected format when the analysis is complete. This report outlines each surge scenario with the resulting pressure and produces tables and trends of the data. The automatically generated surge analysis report is provided in a format that the regulatory authorities and surge analysis engineers can easily scrutinise. ‘Figure 7- Surge Analysis Sample Pipeline Network’ shows a sample pipeline configuration for a pipeline with four intermediate pumping stations. ‘Figure 8 – Surge Analysis Report’ shows the automatically generated report from the surge analysis tool for the sample pipeline configuration, where examples of the tables and trends are displayed. The automatically generated report shows a selection of information within a tabulated format for each pressure surge scenario. This information includes: • • • • • •

Software version for traceability Project name and description Date report is generated Name of closed valve \ tripped pump Indication of MAOP \ SOP violation Time \ Location \ Pressure of MAOP \ SOP violation

10


•

Pump trip times

The report also generates a selection of trends for each pressure surge scenario including: • •

PSIG 1417

automatically rerun the schedule of scenarios with the applied pressure surge mitigation procedures and devices incorporated. An updated report will then be generated for the modified scenarios.

Pressure trend at MAOP \ SOP violation Pressure trend upstream of valve closure \ pump trip Maximum piezometric pressure profile for each scenario Minimum piezometric pressure profile for each scenario Scenario flow trends

A pipeline configuration report may also be generated from the hydraulic simulation software to be included within the pressure surge analysis report. This configuration report will provide detailed information regarding the pipeline configuration, such as pipe lengths, diameters, wall thickness, pump and valve performance curve data and the model boundary conditions for each model item.

This information is automatically generated without any further interaction with the pressure surge analysis engineer. Pressure trends at the location of the MAOP \ SOP violations are automatically generated at the point of the violation. There is no requirement for a reporting point to be pre-defined within the simulation at the location of the violation. This prevents a duplication of the surge analysis study, where an initial analysis is conducted to determine the location of the violations and a second study is conducted to interrogate the pressure trends at the location of any violations.

A surge analysis study could take several days, weeks or even months to conduct depending upon the size and complexity of the pipeline. Running each individual pressure surge scenario, interrogating the simulation results and generating a surge analysis report are all routine tasks which can be automated. The automation of these tasks often allows the surge analysis study to be conducted within a few hours. This can reduce costs and enhance opportunities significantly. The enhanced opportunities could be from the additional revenue generated by completing the surge analysis study timely for an increase in pipeline capacity earlier, or the ability to provide temporary capacity increases to fulfil a one-off demand which may not have been possible if the surge analysis study were to take several days or weeks to conduct. The automated interrogation of results also allows for the removal of human errors from the procedure, thus increasing the accuracy of the analysis. The selection of an incorrect location of the peak pressure, for example, could have catastrophic consequences.

• • •

In the event of a failure within any of the surge analysis scenarios this information can be used to determine which surge mitigation procedure or device to incorporate. The minimum and maximum piezometric pressure profiles as determined by the hydraulic analysis results may be used with (Equation 5) for example to calculate the required wall thickness for a given design pressure. Increases within the pipeline wall thickness are normally not a practical solution for an existing operational pipeline, due to the costs associated with the modification. Existing operational pipelines are usually ‘looped’ as a way to increase capacity for example. Increased pipeline diameter is therefore typically only considered during the pipeline design phase. The pressure surge analysis study may then

Pipeline Logic Control During a pressure surge a flow reversal may occur, which could result in the over pressurization of the system if the transient is not controlled correctly. Pipeline pressure surge can be controlled providing the procedures are fully understood and planned in advance. Planned procedures may assist with minimizing the impact of a pressure surge. This requires that a previous analysis of the system

PSIG 1417


11

including the pipeline controls has been conducted. In the event of a pipeline operator closing a pipeline valve too quickly for example, the pipeline control logic may be replicated and where necessary modified to safely respond to the valve closure, minimizing the pressure surge and reducing the likelihood of cavitation.

mile) downstream. ‘Figure 11: Case 1 Pipeline Modification Surge Analysis’ shows the pipeline configuration.

Advanced pipeline control logic may be included in the hydraulic simulation to achieve the most accurate representation of the pipeline system response. This control logic includes feedback control elements such as PID control blocks. ‘Figure 10: Control Diagram Editor’ shows a PID control block embedded within the hydraulic simulation software, where feedback control of the pipeline system is being used to automate the response of pipeline components to the effects of a pressure surge. Utilizing control elements within a hydraulic simulation allows the simulation to replicate the physical control system on the pipeline network.

Steady state analysis of the pipeline indicates that the pipeline discharge pressure at the main pump station is required to be 5100 kPa (740 psi) and a discharge pressure at the intermediate pump station is required to be 4500 kPa (652.67 psi). A dynamic surge analysis study can now be conducted on the pipeline. Three pressure surge events will be considered:

Pipeline pressure surge analysis may therefore be conducted on the pipeline control system in addition to the pipeline. This allows the pressure surge analysis study to monitor the effects of the control system and indicate much safer methods for operating the pipeline network.

Case Studies Two surge analysis case studies will be considered in this section. Case 1 considers a capacity expansion of an existing low sulphur diesel pipeline, and case 2 considers a new pipeline design study. Case 1 Consider a pipeline modification study for an existing low sulphur diesel pipeline with 100 km (62.14 mile) in length and 16 inches in diameter. The pipeline requires an increase in the existing flow rate. This increases the volumetric flow rate from 0.15 m3/s (5.3 ft3/s) to 0.2 m3/s (7.06 ft3/s). The pipeline has a booster pump and a mainline pump located at the pipeline inlet and an intermediate pumping facility located 50 km (31.07

There is a maximum allowable operating pressure of 6550 kPa (950 psi) and a minimum pressure at the pipeline outlet of 450 kPa (65 psi).

1. Pump trip at the pipeline inlet 2. Pump trip at the intermediate pumping station 3. Sudden valve closure at the pipeline outlet The steps demonstrated within the ‘An Example Surge Analysis’ section above could take several days to perform for this modification. A hydraulic simulation utilizing the surge analysis program can perform this operation in a number of minutes. This allows the surge analysis study to be much more cost effective and allows the results to be obtained much sooner. The pipeline operator would therefore benefit from the cost savings of conducting the studies and would be able to utilize the increased pipeline capacity much sooner, generating additional revenue. ‘Figure 12: Case 1 Pipeline Maximum Pressures Violation’ shows an automatically generated trend from the surge analysis tool for the peak pressure location on the pipeline. An image of this trend is appended to the surge analysis report which is produced by the surge analysis programme. The blue trend shows the pressure over time at the location of the peak pressure during the simulation run. The red and orange trends show the MAOP and SOP limits respectively. It can be seen that the peak pressure exceeds the MAOP limit when the valve is closed suddenly at the outlet.

12


In order to mitigate the pressure surge an increase in the valve closure time could be made. Each of the simulation scenarios can be rerun automatically with the applied change. ‘Figure 13: Case 1 Pipeline Maximum and Minimum Pressures’ is a pipeline profile trend from the report produced by the surge analysis programme following the applied change. This trend shows the maximum (red trend) and minimum (green trend) pressures for each point along the selected pipeline section, for the entire duration of the hydraulic simulation. It can be seen that the increased valve closure time has the desired effect on lowering the peak pressure below the MAOP setpoint. This peak pressure could be lowered further through the use of the advanced control logic within the hydraulic simulation. This could be done by tripping the upstream pumping facilities following the valve closure. This logic control could then be implemented on the physical pipeline. Case 2 Consider a pipeline design study for a new crude oil pipeline with 60 km (37.28 mile) in length and 16 inches in diameter. The pipeline requires a pumping station to be located at the main inlet, and has a tiein point at 20 km downstream. There are 8 intermediate block valve stations along the pipeline. ‘Figure 14: Case 2 Pipeline Design Surge Analysis’ shows the pipeline configuration.

There is a maximum flow rate at the pumping station of 0.15m3/s (5.3ft/s) and 0.04m3/s (1.41ft/s) at the tie-in point, and a discharge pressure requirement at the pipeline inlet of 10100 kPa (1450 psi). There is a maximum allowable operating pressure of 10200 kPa (1480 psi). Steady state analysis of the pipeline indicates that the pipeline pressures and flows are sufficient to meet the production targets. A dynamic surge analysis study can now be conducted on the pipeline. Eleven pressure surge events will be considered: 1. Pump trip at the pipeline inlet

PSIG 1417

2. Sudden valve closure at the 8 intermediate block valve facilities 3. Sudden valve closure at the pipeline inlet and outlet Usually this analysis could take several days to weeks to conduct. The surge analysis tool can be used to conduct the required pressure surge analysis in a much more efficient manner, reducing the analysis time to several hours. The one step solution provided by the surge analysis programme prepares a detailed analysis of the pipeline pressures and flows for the required pipeline section. The automatically generated report shows that the required valve closure time produces peak pressures of 12100 kPa (1750 psi). This pressure is above the MAOP threshold of 10200 kPa (1480 psi), and also above the SOP threshold. In order to mitigate the pressure surge an increase in the valve closure time could be made, however this may be limited by the emergency shutdown requirements on the pipeline. A second option is for a surge relief tank to be located immediately downstream of the pumping station with a relief pressure of 10100 kPa (1465 psi), and an additional tank located at the first and second block valve stations. ‘Figure 15: Case 2 Pipeline Design Surge Analysis with Surge Relief’ shows the modified configuration with the pressure relief valves and surge relief tanks installed. ‘Figure 16: Case 2 - The Effect of a Relief Valve on Pressure’ shows the effects of the surge relief valves in reducing the pressure surge to below the MAOP limit. The green trend shows the pressure at the pipeline high pressure location without the relief valve, and the blue trend shows the pressure at the same location with the relief valve. The orange trend shows the MAOP setpoint. It can be seen that the chosen relief valve has the desired effect on lowering the peak pressure below the MAOP setpoint. As with ‘Case 1’, the peak pressure could also be lowered further in ‘Case 2’ through the use of the

PSIG 1417


advanced control logic within the hydraulic simulation. This could be done by tripping the upstream pumping facilities following the valve closure. This logic control could then be implemented on the physical pipeline.

Conclusions Surge analysis studies are an essential part of pipeline design and operations planning. Hydraulic analysis of the pipeline network allows for potential pressure surge risks to be identified and for surge mitigation measures to be designed and implemented. In order to avoid pipeline damage resulting from pressure surge, it is necessary to determine if there is likely to be a pressure surge. This may be determined through the use of pipeline simulation tools, where valve closures and pump station shutdowns may be simulated and the resulting fluid behaviour analyzed.

13

This paper has demonstrated that the whole surge analysis procedure can be automated into a single step, reducing the time from several day or weeks to a few hours. In addition, it improves the accuracy of the analysis by removing human errors from the process. The reduced time in completing such surge analysis can help increase revenues by running pipelines at higher capacities sooner and reduce the cost of such analysis.

Author Biography Garry Hanmer is a Principal Project Engineer at ATMOS International in Manchester, United Kingdom. Garry Hanmer has over 8 years’ experience in the pipeline industry with an emphasis on pipeline hydraulic simulation. He also has experience in development and delivery of pipeline operations and integrity management software systems. Garry Hanmer has a Master of Engineering in Aeronautical Engineering (MEng Hons) from the University of Salford, UK.

References To prevent an increase in fatigue damage to the pipeline, devices and procedures such as valve opening and closing times, pressure relief valves, surge tanks, increased pipeline diameter and increased pipeline wall thickness can be used as surge mitigation measures. Efforts however should be made to ensure that these measures agree with emergency procedures. To reduce the effort required in conducting pipeline surge analysis studies, routine procedures may be automated utilizing a surge analysis tool to schedule and run simulation scenarios, interrogate the simulation results and generate a surge analysis report.

1. Flixborough (Nypro UK) Explosion 1st June 1974. http://www.hse.gov.uk (http://www.hse.gov.uk/comah/sragtech/caseflix boroug74.htm) 2. Pipeline Pressure Limits - Pipelines Safety Regulations 1996 http://www.hse.gov.uk (http://www.hse.gov.uk/pipelines/resources/pipe linepressure.htm) 3. Entrapped Air in Pipelines, Martin C.S., 1976 4. Hydraulics of Pipeline Systems. Bruce E. Larock, Roland W Jeppson, Gary Z Watters. 5. Pipeline Design and Construction – A Practical Approach. M. Mohitpour, H. Golshan, A. Murray. 6. Fluid Transients in Pipeline Systems. A.R.D Thorley

FIGURES

Figure 1- Single Pipe Example for Water Hammer Analysis

Figure 2- Upstream Pressure when the Valve is closed within Different Time Periods

Figure 3- Upstream Velocity When the Valve is Closed within Different Time Periods

PSIG 1417


Figure 4- The Effect of a Relief Valve on Pressure

Figure 5- External Diameter Vs. Design Pressure

15

16


Figure 6- Pipe Wall Thickness Vs. Design Pressure

Figure 7- Surge Analysis Sample Pipeline Network

PSIG 1417

PSIG 1417


Figure 8- Surge Analysis Report

Figure 9- Fluid Vapour Pressures

17

18


Figure 10: Control Diagram Editor

Figure 11: Case 1 Pipeline Modification Surge Analysis

PSIG 1417

PSIG 1417


Figure 12: Case 1 Pipeline Maximum Pressures Violation

Figure 13: Case 1 Pipeline Maximum and Minimum Pressures

19

20


Figure 14: Case 2 Pipeline Design Surge Analysis

Figure 15: Case 2 Pipeline Design Surge Analysis with Surge Relief

Figure 16: Case 2 - The Effect of a Relief Valve on Pressure

PSIG 1417

PSIG 1417


21

TABLES Application Gas (nonsour) General & cased crossings Roads Railways Stations Gas (sour service) General & cased crossings Roads Railways Stations High Vapour Pressure Liquid General & Cased crossings Roads Railways Stations Low vapour pressure liquid All but uncased RR crossings Uncased railroad crossings

Design Factor * Location (Gas and Liquid)) Class 1 Class 2

Class 3

Class 4

0.8 0.6 0.5 0.5

0.72 0.5 0.5 0.5

0.56 0.5 0.5 0.5

0.44 0.4 0.4 0.4

0.72 0.6 0.5 0.5

0.6 0.5 0.5 0.5

0.5 0.5 0.5 0.5

0.4 0.4 0.4 0.4

0.64 0.64 0.5 0.64

0.64 0.64 0.5 0.64

0.64 0.64 0.5 0.64

0.8 0.8

0.8 0.5

0.8 0.5

Table 1 – Design Factors for CSA Z662-07 Pipe Type Seamless Electric Welded Submerged arc welded Furnace butt welded

CSA Z662-07 1.0 1.0 1.0 0.6

Table 2 – Longitudinal Joint Factors for CSA Z662-07 Temperature (C) > 120 150 180 200 130

CSA Z662-07 1.0 0.97 0.93 0.91 0.87

Table 3 – Temperature Derating Factors for CSA Z662-07 Pipe Diameter Pipe Wall Thickness Pipe Length Pipe Modulus of Elasticity Pipe Inlet pressure Pipe Steady State Flow Restraint Factor Fluid Density

SI 0.9144 0.0127 500 1300000000 5101325 5 1 840

m M M Pa Pa m3/s Kg/m3

Table 4 – Pipeline Properties

Imperial 36 0.5 1640.41 188549.06 739.88 176.57 1 52.43

in in ft psi psi ft3/s lb/ft3

1417 Pipeline Surge Analysis Studies Atmos_PSIG

Recommend Documents