Recent progresses in bus-ducts design.pdf

The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/ / 0332-1649.htm www.emeraldinsight.com 0332-1649.htm

Recent progresses in bus-ducts design

Bus-ducts design

J. Faiz, H. Ehya and A.M. Takbash Center of Excellence on Applied Electromagnetic Systems, School of Electrical and Computer Engineering, College of Engineering, University of Tehran, Tehran, Iran, and

S. Shojaee, M. Hamidian and A. Ghorbani Generator Engineering and Manufacturing Co. (Pars), Mapna Group, Tehran, Tehran, Iran ) T P ( 6 1 0 2 y r a u n a J 7 0 1 5 : 9 0 t A z i a F d a w a J r o s s e f o r P y b d e d a o l n w o D

117 Received 26 February 2015 Revised 26 April 2015 Accepted 29 April 2015

Abstract Purpose – Electrical energy distribution systems must be low losses systems in order to enhance the system efficiency. Therefore, it is preferred to distribute electrical energy by bus-ducts in the place of cables over all energy levels and decrease the losses. The purpose of this paper is to focus on a comprehensive survey of various aspects of bus-ducts design including electromagnetic, mechanical and thermal. Advantages and disadvantages of different available design techniques are reviewed. Design/methodology/approach – Different works on various bus-based power transmission and distribution systems are reviewed. Generally these are done in three categories including systems modeling methods, heat transfer in the systems, short circuit and electromagnetic force. The attempt is made to provide geometrical geometrical and materials materials specifications specifications in order to present the analyzed system well. Findings – Different types of bus-ducts from used materials, voltage level and insulation types are reviewed. Bus-duct modeling techniques are introduced which can be easily applied for bus-ducts design. Electromagnetic field distribution, thermal pattern inside and outside of the bus-duct in normal and short circuit modes and finally mechanical considerations are dominant factors which must be taken into account in the bus-ducts bus-ducts design. This leads to an optimal design of bus-ducts which prolong the life span of the bus-ducts fixed in the installations. Originality/value – This This paper for the first time systematically reviews the latest state of arts in the design of bus-ducts for efficient electrical energy distribution. It summarizes a variety of design techniques applicable to bus-ducts design. Keywords Electromagnetics, Electromagnetics, Modelling, Modelling, Thermal Thermal modelling, modelling, Thermal Thermal analysis, Bus-ducts, Electromagneti Electromagneticc analysis, analysis, Modelling Modelling methods methods Paper type General review

1. Introduction Introduction Bus-ducts have wide applications in industrial complex and large residential buildings. Beside Beside good good safety safety of these these system systems, s, its most most import important ant advant advantage agess includ includee easy easy fixing, fixing, suitable suitable maintenan maintenance ce and repairs repairs (Hwang (Hwang et al., 1998). 1998). Generally, Generally, bus-duct bus-duct system consists of a number of insulated copper or aluminum conductors, normally surrounded by a earthed enclosures. Due to the the wide spread of electrical equipment equipment and import importanc ancee of their their manufa manufactu cturin ring g safety safety,, the bus-du bus-ducts cts are covere covered d by electr electrica icall insulation materials. This has a number of advantages; the most important advantage is that the bus-ducts containing different currents can be placed beside each other. This can reduce the size of electrical distribution system as well as leakage reactance of This paper was originally accepted for the special issue on “Numerical Field Calculation in Electrical Engineering (IGTE 2014)”, Vol. 34, Iss. 5. The authors wish to thank Mapna-Pars Generator Company in Iran for financial support of the bus-duct design project.

COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering Vol. 35 No. 1, 2016 pp. 117-136 © EmeraldGroupPublishi EmeraldGroupPublishing ng Limited Limited 0332-1649 DOI 10.1108/COMPEL-02-2015-0099

COMPEL 35,1

the the syst system em;; this this lead leadss to a lowe lowerr volt voltag agee drop drop (WET (WETOW OWN N BUSW BUSWAY AY Co., Co., 2010 2010). ). Enclosures around conductors prevent the bus-ducts from mechanical damages and dust. The high voltage (HV) bus-ducts are utilized in the following cases (Sarajcev, 2012): (1) (1) main main circ circui uit; t; (2) main transformer transformer connections connections;;

118

(3) auxiliary auxiliary transformer transformer connections connections;; (4) station station auxiliary auxiliary circuits; circuits; (5) generator generator connection connections; s; (6) excitation excitation transforme transformerr connect connections; ions; (7) current current transfor transformers mers assemb assembly ly structu structures; res;

) T P ( 6 1 0 2 y r a u n a J 7 0 1 5 : 9 0 t A z i a F d a w a J r o s s e f o r P y b d e d a o l n w o D

(8) wall throug through h seal and and expansi expansion on joint joint structures structures;; (9) removable removable expans expansion ion joint joint struct structures; ures; and and (10) PT-LA cabinet cabinet connection connections. s. Bus-du Bus-ducts cts design design depend dependss on the applic applicati ation on type type and manufa manufactu cturing ring compan company. y. The The comm common on feat featur uree of thes thesee syst system emss is the the exist existen ence ce of copp copper er or alum alumin inum um conductors with aluminum or galvanized steel enclosure. Isolated phase Bus-bars have enclos enclosure ure around around differ different ent bus-ba bus-bars rs and thus thus they they are isolat isolated ed from from each each other. other. The enclosures around different phases are linked at the end of the bus-ducts route. This prevents the axial induced currents in the bus-ducts. Generally, high current passes through the bus-bars, therefore, bus-bars and their enclosures are chosen cheap aluminum material (Niemoller, August 1968). On the other hand, bus-ducts and all belo belong nging ing must must be able able to supp supply ly elec electr tric ical al ener energy gy in the the foll follow owin ing g cond condit itio ions ns (Raja Ramanna Centre for Advanced Technology, 2015): (1) ambient ambient air temperatu temperature re 40°C; 40°C; (2) maximum maximum ambien ambientt air temperatu temperature re 48°C; 48°C; (3) relative relative humidity humidity not not exceeds exceeds 95 percent non-co non-conden ndensing; sing; and and (4) ingress ingress protection protection degree degree (acc. (acc. IEC 60529) 60529) at least least P65. P65. In therm thermal al and and hydr hydro o powe powerr plan plants ts,, high high curr curren entt buse busess link link gene generat rator orss to unit unit transformers. These bus-bars consist of a conductor inside an enclosure filled by the air under atmosphere atmosphere pressure. pressure. Such line has capacity to pass 10 kA at rated voltage of 36 kV in hydro power power plant and 20 kA in thermal thermal power plant. plant. These bus-ducts bus-ducts have been used since 70 decades. One of the common gases used in these systems was SF 6. Recently the mixture of 95 percent N 2 and SF6 has been used. Such power lines at voltage voltage higher higher than 245 kV and power between between 2,000 and 4,000 MVA, have the most al., 2010 advantage advantagess (Piatek (Piatek et al. 2010). ). Tran Transm smis issi sion on gas gas insu insula late ted d line liness (GIL (GIL)) as an economical method for large power transmission over short and long paths have been proposed. These lines can be used on the ground, buried underground and in the rail way tunnels. When these systems are used underground, a steel cover is included (Piatek et al., 2010). In a bus-duct system, different phases can have separate enclosures and this system is called non-segregated three-phase bus-duct. In this case, different bus-du bus-ducts cts may may be replac replaced ed in an enclos enclosure ure by separa separator torss and they they are called called segreg segregate ated d three-phase bus-duct (Piatek et al., 2010).

To justify the advantages of bus-ducts application over cables, two systems, GIL and over head line (OHL), for a 420 kV transmission line can be compared (Piatek et al., 2010). GIL system advantages include low losses, low capacitive load, longer electrical and thermal age, low impact on the environment and high reliability (Sarajec, 2011). Difference between the OHL and GIL losses is 64 MW. Cross-sections of bus-ducts conductors can be circular, rectangular (flat), hollow octagon and U -shape (Sarajcev and Goic, 2010). Figure 1 shows the segregated three-phase bus-duct with rectangular cross-section. In this paper, different works on various bus-based electric power transmission systems are reviewed. Generally, these are done in three categories including system modeling methods, heat transfer in these systems, and short circuit and electromagnetic force. It is tried to give the geometrical and materials specifications in order to specify the analyzed system well. ) T P ( 6 1 0 2 y r a u n a J 7 0 1 5 : 9 0 t A z i a F d a w a J r o s s e f o r P y b d e d a o l n w o D

Bus-ducts design

119

2. Bus-duct modeling methods Two general approaches are proposed in modeling bus systems such as bus-ducts. The first approach is analytical method based on mathematical equations and second approach is numerical method. Some references have been devoted to comparison of these two methods and their efficiencies. In Canova and Giaccone (2009), analytical and numerical methods in modeling a bus system have been compared. At this end, analytical equations for a system with arbitrary number of conductors has been obtained based on a current supply and/or voltage supply bus system, and then the proposed system has been modeled considering the supply type. The mentioned method has been used to examine the industrial bus system problems. In this modeling method, each conductor is divided into a number of thinner conductors and any of these conductors is taken equivalent with a wire. Cross-section of these wires has significant impact on the accuracy of the method. To achieve precise results using this method, the cross-section of the wires must not exceed the defined skin depth. The lumped parameter circuits of the system for two currents and voltages supply cases have been presented in (Canova and Giaccone, 2009). In the model, each branch indicates one of the wires; and inductance between any two-wires is modeled by a voltage-dependent source. The proposed bus system has double-conductor per phase and cross-section of each conductor is 10 × 100 mm. The space between the bus conductors of each phase is 37 mm. In this circuit, a set of currents satisfying the Kirchhoff ’s current law is applied to branches considering the supply type. It is shown in Canova and Giaccone (2009) that the current density in the conductors is not constant due to the skin and proximity effects. Referring to the structure of the separation partition

phase conductor shield

Figure 1. Segregated three-phase bus-duct with rectangular cross-section conductor

COMPEL 35,1

120


bus-ducts, only there is a force along horizontal axis. In addition, the current density along the line using the magnetic equivalent circuit and finite elements (FE) methods is very close. Other structures of bus systems have been also analyzed and their impacts upon system behavior have been considered by a factor defined as geometry factor vs electromagnetic force exerted on the conductors. Advantage of this analytical method is its simplicity and capability to extend to different bus systems. Various parameters are considered to improve analytical modeling method. For example, in Coneybeer et al. (1994), skin effect and proximity effect have been taken into account. In addition, electromagnetic coupling between conductors and enclosure in two high current bus-ducts with identical cross-section has been considered. Equivalent circuit (EC) method is one of the analytical modeling methods used in electrical devices. In Piatek et al. (2010), the EC method for a three-phase bus-duct has been introduced. In this case, it is assumed that the enclosure of bus-duct is earthed and then the impact of the return current on the induced magnetic fields in the enclosure is investigated (Figure 2). I e1, Ie2 and Ie3 are return currents and their amplitudes depend on many factors such as impedance of return phase current path, the related phase current and transmission line parameters, particularly the mutual inductance between the conductors and enclosures. In Figure 2, mutual inductances between phases 2 and 3 have been eliminated and Z g is the earth inductance. Since enclosures may be connected directly or through the earth, the return current path and its related phase current are not identical and consequently unique impedance cannot be introduced for phase system. To study the magnetic field of the circular cross-section of high current bus-ducts, an analytical model considering geometric mean distance has been introduced in Coneybeer et al. (1994). In this model skin effect, proximity effect, mutual impacts of conductors and enclosure effect have been taken into account in order to enhance the accuracy of the model. This model also is used in GIL systems. The most important problem in an analytical modeling method of bus-duct is determining correct current distribution in the system. The accuracy of remaining parameters and elements of the model depends on the current distribution. This method is based on filament method in

Z11

I1

L1

L2

Ie1

Ze1

I2

Z22

Ie2

Ze22

Z1e1 Z1e2

Z12

Z1e3

L3

Figure 2. Equivalent circuit of a high current threephase bus-duct with earthed enclosure

I1L

I3

Z33

Ie3

Ze33 Zg

Z13

which each conductor is divided into a set of smaller wires. There are the following assumptions in this method:


•

current passes along filaments;

•

electrical resistance and magnetic permeability are identical along filaments; and

•

all filaments are in series and this is the base for skin effect estimation.

Structure of a three-phase bus-duct system has been given in Imamura et al. (1998). The dimensions of conductor of each phase are: 5 mm thickness, and 50 mm width and required length. The space between the phases conductor is 70 mm. In Imamura et al. (1998), ac and dc magnetic fields due to a three-phase bus-duct system are estimated in order to model an optical transformer. Figure 3 shows the magnetic field distribution due to dc current in open conductor and conductor with fence. In this case, 424 A dc current pass through two lateral buses and 848 A pass through the middle conductor. As shown in Figure 3, in the first case the center of the magnetic loops in two lateral buses is displaced from the bus center due to the current passing the middle conductor where there is an enclosure around each bus and center of magnetic field loops coincide with the bus center. Magnetic field components for different types of dc and ac supplies in a bus system having two conductors with enclosure and with no enclosure close to the buses have been investigated in Imamura et al. (1998). Impact of particles and metal contents in the gas isolated bus-ducts upon the efficiency of these bus-ducts has been examined by Nageswara Rao (2012). Around 20 percent of the reported faults in this type of bus-ducts are caused by metal particles. Impact of the metal particles dimensions on their movements in different electrical fields in the bus-ducts has been investigated. The thinner particles more likely lead to electrical discharge. Also aluminum particles are more than copper particles under influence of voltage level.

Bus-ducts design

121

3. Thermal analysis It is essential to inspect the losses, thermal analysis and temperature of bus-ducts. Ohmic losses in bus-ducts are important from some aspects. First, Ohmic losses due to passing current through different conductors in various operating conditions must be considered. Second aspect in losses discussion is related to eddy current in the enclosure around the bus-duct. Magnetic field generated by bus-duct conductors current induces eddy currents in the enclosure; these eddy currents cause Ohmic losses in the bus-duct enclosure. Importance of Ohmic losses in heat dissipation and (a)

(b)

Notes: (a) Open conductor; (b) conductor with fence Source: Adapted from Imamura et al. (1998)

Figure 3. Magnetic field distribution due to dc current

COMPEL 35,1

122


temperatures rise in different parts of bus-ducts are very clear. In Hwang et al. (1998), heat dissipation in bus-duct has been examined. In addition to the mechanical and physical protection role of the external enclosure, it also plays an important role in the heat dissipation. These losses have been estimated using FEM in a bus-duct where cross-section (Figure 4), specifications, material properties and heat transfer coefficients have been given in Table I (Hwang et al., 1998). Losses density due to the eddy current in the upper layer of bus-duct enclosure is in y-direction, and in middle axis of upper layer is minimal. The reason is that considering the symmetry between phases, induced eddy current in this region is lower and consequently the Ohmic losses is smaller; while the eddy current losses in the lateral sheets close to the buses is maximal. Figure 5 shows the temperature distribution in the bus-duct cross-section. Temperature of the outer bus is only 0.5°C lower than that of the middle bus. The reason is that the heat dissipation close to the heat exchange surface is better and it has lower temperature. Figure 6 shows the temperature profile along y-axis for different types of insulations with different heat conduction. In this figure, k is the heat transfer coefficient. As expected using insulations with higher heat conduction results in lower temperature of bus. In addition, there is a good agreement between the test and predicted results. Mutual heat impacts of two adjacent buses have been examined in Coneybeer et al. (1994). Two bus-duct systems in the output of a medium voltage and very high current generator influence the heat exchange. In addition, magnetic field due to a bus-duct has negative impacts on the efficiency of adjacent bus-duct. The first step in such analysis is examining the current distribution in the conductors and inspecting the impacts of the conductors ’ current on the bus-duct enclosure (Canova and Giaccone, 2009). X

e c a f r u s p o T

Spacing

Busbar e c a f m r u m S 5 2 m o 1 t t o B

Side Surface

Y

Side Surface

Figure 4. Cross-section of proposed bus-duct

Table I. Dimensions, properties and heat transfer coefficients of modeled 2 kA and 50 Hz bus-duct

340mm

Source: Adapted from Imamura et al. (1998)

Bus-bar Insulation Bottom surface Side surface

Thermal conductivity (W°C−1 m−1 )

Electrical conductivity (S m−1 )

Electrical permeability ( μr )

386 0.15 16.286 0.52

5.8 × 107 0 1.03 × 107 0

1 1 200 1

Bus-ducts design

B: Steady-State Thermal whole system temperature Type: Temperature Unit: °C Time: 1 10/6/2013 5:53 AM

123 73.482 Max 72.698 71.914 71.13 70.345 ) T P ( 6 1 0 2 y r a u n a J 7 0 1 5 : 9 0 t A z i a F d a w a J r o s s e f o r P y b d e d a o l n w o D

69.561 68.777 67.993 67.208 66.424 Min

Figure 5. Temperature distribution in bus-duct cross-section

Source: Adapted from Hwang et al. (1998)

90 85 80

) C ° ( 75 e r u t 70 a r e p m 65 e T

K1=0.2 K2=0.25 K3=0.3

60 55 50

−150

−100

−50

0

Location along

50 Y - axis

100

150

(mm)

Source: Adapted from Hwang et al. (1998)

Heat dissipation in the air-isolated high current bus-ducts has been considered by Bachorec et al. (2004) in which two Ansys-Emag and Ansys-Flotran softwares have been coupled. In addition, three types of three-phase bus-ducts consisting of insulated three-phase bus-duct, no segregated three-phase bus-duct and segregated three-phase bus-duct have been investigated.

Figure 6. Temperature distribution along y-axis for different types of insulations with different heat transfer conduction

COMPEL 35,1

124


Figure 7 shows the schematic of insulated three-phase bus- duct. Simulations and tests have been carried out at rated current of 8.4 kA, frequency of 50 Hz and environment temperature of 16°C. Figure 8 shows the thermal analysis of the insulated three-phase bus-duct in which mutual impacts between phases from electromagnetic and thermal point of views are minimal because of insulated enclosure. Losses density as output of the electromagnetic analysis is the input of thermal analysis. Figure 9 shows 2D heat distribution in an insulated three-phase bus-duct.

Figure 7. Schematic of insulated three-phase bus-duct

Source: Courtesy of Pars Generator

1

NODAL SOLUTION

ANSYS 7.0

STEP=35 SUB=1 TEMP SMN=15.406 SMX=56.785

0.8m



Figure 8. Thermal analysis of an insulated three-phase bus-duct

15

17.5

20


22.5

25

27.5

30

1 AVG ELEMENT SOLUTION

Bus-ducts design

ANSYS 7.0

STEP=1 SUB=1 FREQ=50 PLOSSO2 (AVG) SMN=28.443 SMX=0.101E+08

125


0

10,000

20,000

30,000

40,000

Figure 9. Heat distribution of an insulated threephase bus-duct

50,000

Insulated three-phase duct: Joule heat per unit volume


Table II summarizes the heat losses of different parts of the above-mentioned bus-ducts based on 2D and 3D analysis (Isfahani et al., 2009). Parameters of two types of high current flat and symmetrical bus-ducts have been analytically estimated considering skin effect (Piatek et al., 2010). In three-phase flat bus-duct, each phase is placed in an insulated enclosure and it is used in HV or very HV. Figure 10 presents eddy current due to conductor itself and adjacent conductors in the conductor enclosure. Since the cross-section in such transmission lines is large, the skin effect must be included in calculation process. In Piatek et al. (2010), the estimated values for two above-mentioned types of bus-ducts have been summarized. Considering high current and medium voltage in the generator output to transformer input, a bus-duct with low Ohmic losses is used. In this case, the solution key for Ohmic losses estimation is the conductors current and induced current in the bus-duct enclosure (Imamura et al., 1998). The skin effect in the phase cross-section is clear in Figure 10; as shown the current tends to pass through the conductor surface. Figure 11 presents the effective induced current in the lower and lateral surfaces of the bus-duct enclosure. Current in the lower surface of the enclosure is very asymmetrical and the highest

Power losses/unit length (W/m) Right phase conductor Middle phase conductor Left phase conductor Right phase enclosure Middle phase enclosure Left phase enclosure Sum of Joule losses in enclosures Total sum of Joule losses

2D

3D

Difference (%)

199 199 199 229 251 222 702 1,298

215 211 222 218 258 237 713 1,361

7.5 5.7 10.4 5.0 2.7 6.3 1.5 4.6

Table II. Heat losses of different parts of bus-duct based on 2D and 3D analysis

COMPEL 35,1

126

current is generated close to the phase conductor. Such current is also induced in the upper surface of bus-duct enclosure. There is the similar situation in the lateral surface. Table III presents the currents in different parts of bus-duct and its Ohmic losses (Imamura et al., 1998). Thermal model of an air insulated HV bus system has been introduced by Del Vecchio (2003). Also heat dissipation coefficients in the boundary regions have been estimated analytically. Figure 12a shows a bus system and its eddy current. Figure 12b presents the heat dissipation of such system. Heat dissipation in this case is both (a)

(b)

Z

X


Hw

I2

θ

e

Y

Je2 Je1

µ0

γ 1

R1

R3 R4 d

) A ( t n e r r u C

(b)

−3

5

× 10

) A ( t n e r r u C

0

−5

10

20

30

40

50

60

70

80

90

100

v

−3

5

× 10

0

−5

10 20

Time (ms)

30

40

50

60

70

80

90

100

Time (ms)

Notes: (a) Lower face; (b) lateral of bus-duct enclosure

Part 1st phase 2nd phase 3rd phase Shield

Table III. Current and losses in different parts of bus-duct

R2

Je1

Je1

(a)

Figure 11. Effective induced current

X(γ ,θ)

I1

Figure 10. Eddy current in conductor enclosure due to magnetic field due to: (a) conductor itself and (b) adjacent conductor

Shield section current Bottom and top side Left-most side Right-most side Left separation partition Right separation partition

Current (A)

Power losses (W/m)

4,603.26 ⩽ 1.3 4,604.39 ⩽ 238.7 4,603.45 ⩽ 118.7 8.62 ⩽ 22.6

293.23 293.34 292.82 287.97

97.92 ⩽ 32.4 1,038.38o 180.4 1,120.38 ⩽ 59.3 989.20 o 130.9 1,009.73 o 18.8

(a)

Bus-ducts design

(b) q2rad

Current Q1

SF6

Eddy Current

q1rad

q2conv

q1conv

127 Tank

SF6

Conductor Tank ) T P ( 6 1 0 2 y r a u n a J 7 0 1 5 : 9 0 t A z i a F d a w a J r o s s e f o r P y b d e d a o l n w o D

Conductor

Flux Q1=q1rad + q1conv , Q1 + Q2 = q2rad +q2conv

Source: Adapted from Bachorec et al. (2004)

Figure 12. (a) A bus system and its eddy currents; (b) heat exchange of system

convection and radiating types. The power losses (heat) and estimated exchange coefficient have been given in Table III. According to Table IV, losses in the conductors are higher than that of the enclosure of bus system. Table V summarizes the measured, analytically estimated and numerically (FEM) estimated temperature of different parts of the system. As shown in Table III, losses in conductors are higher than that of bus system enclosure. According to Kim et al. (2005), the estimated temperature in different parts of the system such as conductors and tank using FEM, analytical method and measured are agreed well. It is also seen that the temperature in an enclosure of bus system is lower than that of conductors of heat dissipation. Eddy current, magnetic field and heat

Temperature (°C) Conductor Tank

Plate materials TS SS AL CU TS SS SL CU

Analytic method

FEM

Measured

43.75 24.31

41.44 21.82

42.6 20.3

Analytic Horizontal placement (H¼ 175, s ¼ 101.6) 214 236 34.1 27.2 Vertical placement (H ¼ 175, s ¼ 101.6) 111 119 12.9 27.2

Table IV. Comparison of calculated and measured temperature of system

FE 213 230 32.4 26.1 112 118 12.6 10.1

Table V. Estimated losses (W/m) in buses with different structures and enclosure material

COMPEL 35,1

128


losses have been considered by Ho et al. (2003) using 3D-FEM. This analysis has been done in an insulated 20 kV, 12.5 kA three-phase bus system and the results have been compared with the test results. Figure 13 shows magnetic flux distribution in conductors and enclosures at peak current of phase B. Additional losses due to the buses carrying high current in the transformer tank have been discussed by Tasic (2000). Impact of different structures of buses shearing using analytical and FEMs has been addressed. Figure 14 shows two horizontal and vertical structures. Table V summarizes the impacts of system structure type, bus system geometry and enclosure material upon the losses. These losses have been estimated in a 5 kA and 60 Hz threephase bus duct. According to Table V, losses in the horizontal structure are higher than that of vertical structure. The reason is that in the horizontal structure, conductors of the bus is closer to the enclosure and consequently more intensive magnetic field is induced in the tank leading to a higher eddy current. In addition, losses reduce in copper enclosure due to a lower electrical conductivity of copper compared to other materials.

(a)

(b) Z

Z Y

X

X

Figure 13. Magnetic flux distribution in (upper) conductors and (lower) enclosures at peak current of phase B

Y

Source: Courtesy of Pars Generator Co.

(a)

(b)

S

S

Figure 14. Two horizontal (a) and vertical (b) structures of buses

S

S

H plate

H d

Source: Adapted from Xu et al. (2007)

plate

d

In Sarajec (2011), thermal capacity of buses in steady-state and transient modes has been addressed using different models (Figure 15). Estimated results are obtained by:

T ðt ÞT 1 ¼ 1et = T 2 T 1


Bus-ducts design

(1)

t

Figure 15 are well agreed with the test results. The results are valid only for step change. Also in Coneybeer et al. (1994), a simple method has been introduced using algebraic equations instead of partial differential equations which studied the thermal analysis in non-steady case. Results of this modeling method have been compared with the measured values which confirm the model. In Del Vecchio (2003), thermal analysis of bus with industrial loads has been done using magneto-thermal method by applying analytical magneto-dynamic method. The results obtained by this method are in good agreement with the industrial data as such that the minimum and maximum temperatures in the analytical method are very close to the measured data. One of the advantages of this method is considering the magneto-dynamic coupled with electrical circuit equations.

129

4. Electromagnetic forces in bus-duct Force is exerted on the current carrying conductors in the magnetic field. This force is a deterministic factor in the electrical equipment design. These forces in bus-duct have been considered in normal conditions as well as short circuit fault. Oscillations and noises in a bus-bar system are very important problems which play vital role in the design of bus-ducts. In Tasic (2000), existing noises in normal conditions of LV and high current buses has been addressed by 3D-FEM. According to Table VI, amplitude of the electromagnetic force exerted on different parts depends directly on the system current amplitude. In addition, a proper analysis of the noise and acoustic oscillations on bus-duct has been done by Tasic (2000). Short circuit fault is dangerous and it must be investigated in bus-ducts particularly in indoor sites with short buses. In Hedia et al. (1999), short circuit electromagnetic forces in a three-phase bus system have been estimated by FEM. In this estimation, a sinusoidal current with the amplitude equal to the short circuit current is applied (Tasic, 2000). To determine meshes in FE analysis, different methods including neural network has been applied and impact of different meshes upon FE modeling has been investigated. Accuracy of the maximum electromagnetic forces per unit length of bus-duct calculated by FEM depends on the mesh type (LIG, graded and CDET) and number of nodes. Lower number of iteration belongs to LIG mesh type which is four but CDT (a)

(b) ht

hfree

hs

ht

hforced

hs hs

hs

Tubular

hs

ht

hs

hs

hb

Rectangular

Angular

Free convection situations

hs

ht

hs hs

Tubular

Rectangular

hb Angular

Forced convection situations

Figure 15. Surface areas used with Nusselt number correlations

COMPEL 35,1

130


mesh type takes 98 iterations. Obviously, more iteration needs longer CPU time and this is 638 s for CDT mesh type compared with 289 s for LIG mesh type. The maximum difference between the CDT and LIG types is around 10 percent. In Xu et al. (2007), a 3D non-linear model has been introduced for electromagnetic forces analysis due to short circuit in a three-phase gas insulated bus system. In this modeling, skin effects and non-linear permeance of materials have been considered. In Kim et al. (2005) and Tasic (2000), electromagnetic analysis in short circuit of buses has been carried out. In Triantafyllidis et al. (2003), the mentioned force with eddy current and magnetic field distribution have been analyzed using FEM in insulated phase bus (IPB); then a point having the highest impact, from electromagnetic force caused by short circuit, has been determined. Figure 16 shows IPB cross-section. Figure 17 shows the magnetic flux lines and flux density in IPB cross-section. When short circuit occurs, magnetic flux density increases. Also in Figure 17, the impact of bus-duct enclosure, which has no considerable effect on the internal magnetic field, upon the environment, is observed. As expected, short circuit causes the increase of the magnetic field density particularly around bus-duct conductor. Figure 18 presents the induced eddy current in bus-duct enclosure during the short circuit fault. The phase induced current in the enclosure has phase difference with the current passing the conductor due to the inductance of the enclosure. Also amplitude of this current is lower than the current passing the corresponding phase conductors. Figure 19 presents the force exerted on the conductors at the time of short circuit fault. Electromagnetic force exerted on the conductor and bus-duct enclosure is identical. Frequency of the electromagnetic force oscillations is equal to the supply current frequency (50 Hz). After approaching this force to its peak value it damps in 1 s.

Current (kA) 1 2

Table VI. 3 Variations of electromagnetic force 4 in different parts

Left plates (mN)

Right plates (mN)

Upper plates (mN)

Lower plates (mN)

Max. on bus-bar (N)

5.8 1.9 23.2 7.70 7.09 23.5 92.6 30.7

9.0 2.9 36.1 11.5 110.4 35.2 144.2 46.1

0.46 0.56 1.84 2.24 5.64 6.86 2.16 8.01

0.28 0.21 0.45 0.84 3.47 2.52 5.28 3.36

0.005

Phase A

Figure 16. Cross-section of a typical IPB

Phase B

Shell

Conductor

Source: Adapted from Del Vecchio (2003)

Phase C

0.018 0.056 0.073

Bus-ducts design

1 1 7 3 7 4 7 4 – 4 9 4 8 3 8 2 7 2 E 3 6 0 3 7 0 4 7 1 2 7 4 2 9 6 4 1 8 6 2 1 3 5 6 8 0 2 3 5 2 . 0 . 0 . 0 . 0 . 0 . 1 . 1 . 1 . 1 . 0 0 0 0 0 0 0 0 0 0

131


X Y

Z

X M

) b (

. o C r o t a r e n e G s r a P f o y s e t r u o C

) a (

: e c r u o S

Figure 17. (a) Magnetic flux lines and (b) flux density in IPB cross section

COMPEL 35,1

(a) × 105 5

0 −5

0

132


0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

(b)

5 ) × 10 A ( 5 s t n e r r u 0 C s l l e h −5 S 0

(c)

5

× 10

4 2

Figure 18. Induced eddy current in bus-duct enclosure during short circuit fault

0 −2 −4

0

Time (Sec)


(a) 1,000

X Component Y Component

0 ) −1,000 N ( s r o t c u d 2,000 n o C n 0 o s e c r −2,000 o F

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

(b)

(c) 500 0

Figure 19. Force exerted on conductors at time of short circuit fault

−500

Time (Sec)


Damping time differs for different phases. Figure 20 shows the electromagnetic force in the middle phase conductor at two phase angles of current. As shown in Figure 20, amplitude of the electromagnetic force varies by variations of the current phase angle, but these variations have no influence on the force amplitude. Figure 21 shows the electromagnetic force at the time of short circuit fault in the enclosure with two

Bus-ducts design

1,500 Current Phase Angle=PI/2 Current Phase Angle=PI/3

1,000

500 ) N ( r o t c 0 u d n o C n o −500 e c r o F

133

−1,000

−1,500


0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

Time (Sec)


Figure 20. Electromagnetic force in middle phase conductor at two phase angles of current

(a) Steel 416 Aluminum

1,000 0 ) −1,000 N ( s r o t c (b) u d 2,000 n o c n 0 o s e c r −2,000 o F

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

(c) 1,000 0 −1,000

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

Time (Sec)


different materials. Figure 21 indicates the maximum force for two materials is almost the same. In the steel, this force reaches its peak earlier and damps quicker and therefore it applies less stress on the supports. Aluminum enclosure makes it possible to cut the circuit by protection devices before approaching the maximum force (Isfahani et al., 2009). Impact of distance between phases on the electromagnetic force due to short circuit fault is shown in Figure 22 (Hassanpour and Vaezzadeh, 2008). Increasing distance between the phases has influence on the volume and affects seriously the electromagnetic force as such that the force decreases largely, but at the time fault occurring it has no impact on the maximum force. Effect of the enclosure material on the short circuit force has been shown in Figure 23 in which aluminum

Figure 21. Electromagnetic force at time of short circuit fault in enclosure with two different materials

COMPEL 35,1

134

alloy having larger resistivity leads to lower amplitude of the electromagnetic force. In Triantafyllidis et al. (2003), a method has been introduced to examine the operating conditions of a bus with several ac conductors. This method is based on the lumped parameters method assuming linear magnetic material. Table VII summarizes the bus short SC currents (Labridis and Dokopoulos, 1996). Asymmetrical current distribution in short circuit case applies large short circuit force (Chiampi et al., 1993). In Ho et al. (2003), magnetic forces in bus-bars with low current used indoor has been studied. The obtained forces using FEM has been 2,400 2,200


2,000 ) 1,800 N ( e c r 1,600 o F

Figure 22. Impact of distance between phases on electromagnetic force due to short circuit fault

1,400 1,200 1,000 1.15

1.2

1.25

1.3

1.35

1.4

1.45

1.5

1.55

1.6

Distance (m)

1,850 1,800 1,750 ) N ( e c r o F

1,700 1,650 1,600

Figure 23. Effect of enclosure material on short circuit force

1,550 1,500 25

26

27

28

29

30

31

Resistivity of Enclosure (nano ohm-m)

Table VII. Short circuit current for different conductors in different bus systems

Ph. no.

R1

S1

T1

R2

S2

T2

R3

S3

T3

3 3&N 3&PE 3 N, PE 2-RS 2-RT 2-ST

33.1 29.8 32.9 33.3 32.8 25 35

41 40 39.5 39.2 36.6 2.4 35

42.5 43 42.6 42.8 2.8 37.1 37.8

42.2 42.3 42.4 42.4 37.6 36.7 2.7

43.1 42.7 42.8 42.6 36.1 0.2 35.9

42.4 42.8 42.6 42.8 2.8 37.1 37.9

42 42.2 42.2 42.4 37.6 36.7 2.7

45.3 43.3 44.4 43.2 35.3 2.5 36.8

33.6 41.6 36.9 41.4 5.6 24.1 32

compared with DIN VDE 0103 and IEC 865 standards and larger than 50 percent difference has been observed. The reason for this difference is neglecting the proximity effect in the standard results. 5. Conclusion This paper reviewed different types of bus-ducts from used materials, voltage level and insulation types point of views. Bus-duct modeling methods were proposed and their resultant signals were investigated. Bus-duct design must be based on electromagnetic, thermal and mechanical considerations and all dominant factors upon these analyses were introduced in the paper. References ) T P ( 6 1 0 2 y r a u n a J 7 0 1 5 : 9 0 t A z i a F d a w a J r o s s e f o r P y b d e d a o l n w o D

Bachorec, T., Hosek, J. and Saska, M. (2004), “Heat transfer simulation and experimental verification of the high-voltage air-insulated bus ducts”, International ANSY Conference, Vol. 1, Pittsburg, PA, May 24-28 . Canova, A. and Giaccone, L. (2009), “Numerical and analytical modeling of bus bar systems”, IEEE Transactions on Power Delivery, Vol. 24 No. 3, pp. 1568-1578. Chiampi, M., Chiarabaglio, D. and Tartaglia, M. (1993), “A general approach for analyzing power bus-bar under A.C. conditions ”, IEEE Transactions on Magnetics , Vol. 29 No. 6, pp. 2473-2475. Coneybeer, R.T., Black, W.Z. and Bush, R.A. (1994), “Steady state and transient ampacity of bus bar”, IEEE Transactions on Power Delivery , Vol. 9 No. 4, pp. 1822-1829. Del Vecchio, R.M. (2003), “Eddy current losses in a conducting plate due to a collection of bus bars carrying currents of different magnitudes and phases ”, IEEE Transactions on Magnetics , Vol. 39 No. 1, pp. 549-552. Hassanpour, A. and Vaezzadeh, S. (2008), “Transient finite element analysis of short circuit electromagnetic forces in isolated phase buses”, Electromagnetics , Vol. 28 No. 6, pp. 590-600. Hedia, H., Henrotte, F., Meys, B., Dular, P. and Legra, W. (1999), “Arrangement of phase and heating constraints in a bus bar”, IEEE Transactions on Magnetics, Vol. 35 No. 3, pp. 1274-1277. Ho, S.L., Li, Y., Lo, E.W.C. and Lin, X. (2003), “Analysis of three-dimensional eddy current field and thermal problems in an isolated phase bus”, IEEE Transactions on Magnetics , Vol. 39 No. 3, pp. 1515-1518. Hwang, C., Chang, J.J. and Jiang, Y.H. (1998), “ Analysis of electromagnetic and thermal fields for a bus duct system ”, Electric Power System Research , Vol. 45 No. 1, pp. 39-45. Imamura, M., Nakahara, M., Yamaguchi, T. and Tamura, S. (1998), “Analysis of magnetic fields due to three phase bus bar currents for the design of an optical current transformer ”, IEEE Transactions on Magnetics , Vol. 34 No. 4, pp. 2274-2279. Isfahani, A.H., Vaez-Zadeh, S. and Khodabakhsh, A.N. (2009), “Calculation of maximum short circuit electromagnetic forces in the IPB using time stepping finite element method”, Electrical Review, Vol. 85 No. 7, pp. 31-35. Kim, J.K., Hahn, S.C., Park, K.Y., Kim, H.K. and Oh, Y.H. (2005), “Temperature rise prediction on EHV GIS bus bar by coupled magneto-thermal finite element method ”, IEEE Transactions on Magnetics, Vol. 41 No. 5, pp. 1636-1639. Labridis, D.P. and Dokopoulos, P.S. (1996), “Electromagnetic forces in three phase rigid bus-bars with rectangular cross sections”, IEEE Transactions on Power Delivery , Vol. 11 No. 2, pp. 793-800.

Bus-ducts design

135

COMPEL 35,1

Nageswara Rao, R. (2012), “Effect of particle size on particle movement in a single phase gas insulated bus duct”, International Journal of Engineering Research and Applications , Vol. 2 No. 3, pp. 544-549. Niemoller, A.B. (1968), “Isolated phase bus enclosure currents ”, IEEE Transactions on Power Apparatus and Systems, Vols Pas-S-7 No. 8, pp. 1714-1718. Piatek, Z., Kusiak, D. and Szczegielniak, T. (2010), “ Electromagnetic field and impedances of high current bus ducts ”, International Symposium of Modern Electric Power Systems (MEPS), Wroclaw, September 20-22 .

136

Raja Ramanna Centre for Advanced Technology (2015), “Technical specifications for sandwich bus duct for power conditioning system ”, Techno-commercial bid (Part I), Raja Ramanna Centre for Advanced Technology, Indore, March 27. Sarajec, P. (2011), “Numerical analysis of the magnetic field of high-current bus ducts and GIL systems”, Energies , Vol. 4, pp. 2196-2211. ) T P ( 6 1 0 2 y r a u n a J 7 0 1 5 : 9 0 t A z i a F d a w a J r o s s e f o r P y b d e d a o l n w o D

Sarajcev, P. (2012), “ Analysis of electromagnetic influences concerning two adjacent parallel high current bus ducts ”, Electric Power Component and Systems , Vol. 40, pp. 829-844. Sarajcev, P. and Goic, R. (2010), “Power loss computation in high current generator bus ducts of rectangular cross section”, Electric Power Component and Systems , Vol. 38, pp. 1469-1485. Tasic, D. (2000), “A procedure for analysis of non stationary heating states of acsr conductors ”, NIS, Electrics and Energetics, Vol. 13 No. 1, pp. 83-94. Triantafyllidis, D.G., Dokopoulos, P.S. and Labridis, D.P. (2003), “ Parametric short circuit force analysis of three phase bus bars- a fully automated finite element approach”, IEEE Transactions on Power Delivery , Vol. 18 No. 2, pp. 531-537. WETOWN BUSWAY Co. (2010), “Medium and high voltage bus-way system”, technical note, catalogue serial number: WTMV, v2010, available at: http://wetownbusway.com Xu, S., Jin, X. and Pang, F. (2007), “Analysis of vibration and acoustic radiation characteristics of bus-bar bridge system under electromagnetic force”, Electric Power Component and System, Vol. 35, pp. 1317-1330.

Further reading Ho, S.L., Li, Y., Lo, E.W.C., Cheng, K.W.E. and Wong, K.F. (2007), “Calculation of eddy current, fluid and thermal fields in an air insulated bus-duct system”, IEEE Transactions on Magnetics, Vol. 43 No. 4, pp. 1433-1436.

Corresponding author Professor J. Faiz can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: [email protected]

View publication stats

Recent progresses in bus-ducts design.pdf

Recommend Documents