A Novel Method for Loss Minimization in Distribution Networks

M. A. Kashem, V. Ganapathy, G. B. Jasmon, M. I. Buhari

Multimedia University
Selangor, Malaysia
Tel: (603) 83 125434, Fax: (603) 831 8 3029, Email: [email protected]

Abstract: Network reconfiguration for loss minimization is the Reduction in power losses is obtained by transferring
determination of switching-options that minimizes the power loads fiom the heavily loaded feeders to lightly loaded
losses for a particular set of loads on a distribution system. In this feeders by reconfiguring the network so that the radial
paper, a novel method is proposed by formulating an algorithm to structure of the distribution feeders can be modified fiom
reconfigure distribution networks for loss minimization. An
time to time. This is done in order to reschedule the loads
efficient technique is used to determine the switching
combinations, select the status of the switches, and find the best
more efficiently for minimizing the losses in the system.
combination of switches for minimum loss. In the first stage of Reconfiguration also allows smoothening out the peak
the proposed algorithm, a limited number of switching demands improves the voltage profile in the feeders and
combinations is generated and the best switching combination is increases the network reliability. A number of researches
determined. In the second stage, an extensive search is employed have been done on the reconfiguration of the network by
to find out any other switching combination that may give rise to closing/opening the tie and sectionalizing switches
minimum loss compared to the loss obtained in the first stage. respectively to minimize losses in a distribution system.
The proposed method has been tested on a 33-bus system, and the A branch exchange type heuristic algorithm has been
test results indicate that it is able to determine the appropriate
suggested by Civanlar et. al. [I], where, a simple formula
switching-options for optimal (or near optimal) configuration
with less computation. The results have been compared with
has been developed for determination of change in power
those of established methods reported earlier and a comparative loss due to a branch exchange. A different method has
study is presented. With the proposed method, for any input load been proposed by Baran and Wu [2] to identify the
conditions of the system, the optimum switching configuration branches to be exchanged using heuristic approach to
can automatically be identified within a reasonable computer minimize the search for selecting the switching options.
time and hence the method can be effectively employed for Merlin and Back [3] have used the branch and bound type
continuous reconfiguration for loss reduction. The method can be optimization technique to find the minimum loss
effectively used to plan and design power systems before actually
configuration. Following the method of reference [3],
implementing the distribution networks for locating the tie-
Shknohammdi and Hong[4] have developed a heuristic
switches and providing the minimum number of sectionalking
switches in the branches to reduce installation and switching
algorithm. In their method, the solution is obtained by frst
costs. closing all the switches, which are then opened one after
Keywords: Loss minimization, Network reconfiguration, another so as to establish the optimum flow pattern in the
Distribution automation, and Switching combinations. network. Goswami and Basu 151 also have proposed a
heuristic algorithm for feeder reconfiguration by solving
I. Introduction the KVL and KCL equations for the network with line
Distribution systems are normally configured radially for impedances replaced by their corresponding resistances
effective coordination of their protective systems. Most only. Broadwater et. al. [6] have applied the Civanlar's
distribution networks use sectionalizing-switches that are method [l] and Huddleston's quadratic loss function and
normally closed, and tie-switches that are normally multiple switching pair operation method 171 to solve the
opened. From time to time, modifying the radial structure minimum loss reconfiguration problem. Peponis et. al. [SI
of the feeders by changing the odoff status of the have used the switch-exchange and sequential switch
sectionalizing and tie switches to transfer Ioads fiom one opening methods for reconfiguration of the network for
feeder to another may significantly improve the operating loss reduction. The methods proposed by Rubin Taleski et.
conditions of the overall system. Feeders in a distribution al. [9] and Kashem et. al. [lo] are also based on branch-
system normally have a mixture of industrial, commercial, exchange technique for reconfiguration to minimize
residential and lighting loads. The peak load on the distribution losses.
substation transformers and feeders occur at different times In all the above methods, the environments under which
of the day, and therefore, the distribution system becomes the calculations are made do not correspond to the actual
heavily loaded at certain times of the day and lightly operating conditions of the system under consideration.
loaded at other times. They depend on some heuristics and approximations, as
the simultaneous switching of the feeder reconfiguration is
not considered due to combinatorial explosive nature of
Paper accepted for presentation at the International the switching options to be encountered. Methods [ll, 121
Conferenceon Electric Utility Deregulation and have been suggested to implement on-line control
Restructuring and Power Technologies2000,
City University, London, 4-7 April 2000.
strategies for power system planing and operations based
on continuous network reconfiguration for loss
0-7803-5902-X/00/%10.0002000 IEEE. minimization.

251
 In this paper, an efficient method has been proposed to switch-groups forming 3" different switching
identify the optimum switching configuration for loss combinations. Each switch position in a switching
minimization and a comparative study with other combination is found out by using the formula,
established methods made. At the beginning of the
proposed method, all the possible switching options
sci = [ SPij , ............. SP,, ,SPi, 3 (1)
obtained fiom the combination of the tie-switch and its two The switch positions of the j* switch in i* combination,
neighbors are selected and the infeasible combinations in SPij can be obtained as,
the selection are omitted. The minimum loss configuration
in the selected combinations is identified. Finally, an SPij = [(i - 1) I 3""] mod 3 (2)
extensive search is carried out by changing the switching where, i is the combination number (one of 3" ), j is the
status one at a time, by either moving to the left or to the switch number ( one of n) and SCi are the positions of the
right of the open branch in the configuration obtained, and various switches in the i th switching combinations.
the configuration with maximum loss reduction is
The switching positions SCi obtained above will be a 3-
determined.
digit number containing the combination of 0, 1 and 2,
where 0, 1 and 2 status represent the combination with tie-
switch open (zero-connect), the right neighbor open (left-
n. Switching Algorithm for Loss Reduction connect) the left neighbor open (right-connect)
In a radial distribution system to achieve a maximum respectively. For zero-connect, the configuration will
reduction in power loss, the aim is to identify the remain unchanged, but for left-connect or right-connect,
appropriate switching options. In most of the techniques, the configuration will change according to the left or right-
the losses are calculated by using load-flow studies for connect logic. Left and right-connect logic are developed
each configuration, and the minimum loss configuration is using the power flow in the adjoining branches of the tie-
found. In the proposed method, the distribution system is branch in the system. When a system is reconfigured by
considered with all its laterals simultaneously, instead of closing the tie-branch and opening either of the adjoining
determining the switching-options on loop by loop basis. branches, the power-flow in the open-branch which was
Usually, this type of technique involves very complicated flowing before reconfiguration, shifts to the newly
mathematical techniques and large computational time due connected tie-branch. If the left-connect is performed, then
to the combinatorial nature of the problem. However, the the power-flow of the right branch of the tie-branch will be
solutions obtained by these methods achieve global shifted to the tie-branch. Similarly, when the right-connect
optimum of loss minimization problem [9]. The proposed is performed, the power-flow of the left branch of the tie-
method develops a switching algorithm, which minimizes branch will be shifted to the tie-branch.
the computational time by performing the effective search
A distribution system after reconfiguration, may become
to the requisite switching combinations.
infeasible because there can be some loads which will be
1I.A Determination of Switching Status isolated from the source of power. In loss minimization
technique, there will be certain infeasible combinations
In general, any tie orland sectionalizing switches can be arising out of restructuring among the 3" different
closed orland opened to perform various network
switching combinations possible. When a branch is left
reconfigurations without creating any closed loop or unconnected due to the opening of the sectionalizing
leaving out any branch unconnected. Any reconfiguration
switches or a closed loop is created due to the closing of
which forms a closed loop, or leaving one or more
the tie-switches, infeasible combinations result. These
branches unconnected is classified as infeasible switching
infeasible combinations must be identified and eliminated
combination for network configuration [ 131. To avoid any fiom the total combinations. The infeasible combinations
infeasible switching combinations, the connectivity fiom
are determined by checking the connectivity of each
the source to all the nodes of the system are checked. If a
branch in the network. Connectivity of a node or branch
valid path exists, then the configuration is a feasible one,
means the existence of physical connections of a node or
otherwise it is infeasible. It is important to note that every
branch from the source. For each combination among all
tie switch or sectionalizing switch selected for closing or
the nodes, select each node one after another, check its
opening for reconfiguration is considered together with its
connectivity and make its connectivity active if it has the
two immediate neighboring switches. If a tie switch is
feasible connectivity. This procedure is repeated for all the
closed, then a closed loop will be formed. To avoid a nodes in the network.
closed loop and restore the radial structure, one of the
neighboring sectionalizing switches must be opened. The 1I.B Finding the Total Losses of a Network by Using
three possible switching combinations form a switch- Simplified Distflow
group. If a sectionalizing switch is opened during
Power flow equations for a radial distribution network
reconfiguration, this open branch and its adjoining
using real power, reactive power, voltages at the sending
branches again form a switch-group. In this way, all
and receiving ends of a branch proposed by Baran and Wu
branches can form switch groups of their own, recursively.
If there are n tie switches in a system, there will be n r2 I are,

252
 open lines or tie lines, depending upon whether the right
adjoining or the left adjoining branch of a tie-switch is
open.
The above equations are called Distflow equations, where, Once the input parameters of the base configuration or
Pi, Qi, V i are the real power, reactive power, and voltage at any other configuration of a radial system under
the sending end and Pi+l, Qi+l, and Vi+l are the receiving consideration, are fed to the algorithm, the algorithm
end quantities respectively. The quadratic terms in the automatically reassigns the branches and loads, for any
equations represent the losses on the branches and hence other switching configuration. Ln the input data, apart from
they are much smaller than the power terms Pi and Ql. specifying the branches, branch impedances, and the loads,
Hence, by dropping the second order terms the set of new we have to specify the tie-branches, the left and right
branch equations can be written as, neighboring branches to each of the tie branch and the
node numbers on both sides of the tie branches indicating
Pi+l=Pi - PLi+l (4a) the extent of search to be carried out onto the right and left
Q.r+l = QI - Q Li+l (4b) sides of the tie branches. The switching combinations are
formed and the status of each switch in a switch-group is
=
vi+12 v? - 2 ( riPi + XiQ) (4c) realized. The infeasible combinations are omitted from the
The solution for the simplified Distflow equations for a total combinations. These switching configurations are
radial network can be obtained as [2], formed according to the switching status found from the
various switching combinations. The power flows are
PI+, = c
k=i+2
PLk calculated for each branch of the switching configuration
under consideration, and the total power loss is estimated.
n

e;+,= C k=i+2
QLk
Likewise, the power loss in each of the remaining
configurations is calculated. The configuration that gives
the minimum power loss is identified. To further narrow
down the search towards the global minimum, a simplified
extension of the above procedure is carried out by
The power loss on a branch is calculated as, considering the connectivity of the combination found
ri(Pi2+ Qf) above. In a reconfigured system obtained from the base
Loss, = configuration, the open sectionalizing or tie switches are
Viz
considered as the tie switches for the next operation for
The total system loss is the sum of all the branch losses network reconfiguration. The switch-groups obtained from
given by, the combination found above are examined individually,
and the switching status of each switch-group is found. If
the connectivity of the switch-group remains in the zero
switch group (zero-connect) itself, then the switch group is
Unlike the other load-flow techniques that take a lot of unaltered. If the connectivity of a switch group is towards
iterations and computational time, in the simplified left (left-connect), the connection is made to move towards
DistfJow, power-flows are calculated in one cycle of right by opening the next neighbor of the right branch of
iteration. Therefore, it is suitable for on-line the current tie switch and closing the current tie switch
implementation as it reduces computational time. forming a new group. If the connectivity of the switch
group is towards right (right-connect), the connection is
1I.C Finding the Optimal Configuration for Loss made to move towards left by opening the next neighbor of
Minimization the left branch of the current tie-switch and closing the
The network reconfiguration for loss minimization is current tie-switch forming a new switch-group. For the
performed by opening and/or closing the sectionalizing- newly formed combination, the power loss is calculated
and tie-switches in such a way that the radiality of the and compared with the power loss in the combination
network is retained and at the same time power losses are estimated earlier. If it is less, then the current combination
minimized. In the proposed method, a systematic approach is considered as the best combination and the search is
is employed to perform automatically all types of further carried out in the same direction; otherwise, the
operations needed for (i) finding the total combinations of combination found earlier is considered as the best
the system, (ii) obtaining the feasible combinations, (iii) combination. The above procedure is applied to other
determining the switching status of the individual switch-groups in succession and the best combination is
switches, (iv) estimating the losses for all the switching found.
combinations, (v) identifying the minimum loss 111. Test System
configuration which might be a local or global minimum
from among the feasible switching combinations, and (vi) To demonstrate the utility of the proposed switching
finally making a search to find the global optimm for loss algorithm, a 33-bus test system is considered and the
minimization by moving to the left or right of each of the system details are given in Table I. It is assumed that there

253
 the proposed method can give optimal or near optimal
solution within a reasonable computer time and effort. It is
proven that in the case of test system, the proposed method
is found to be better because it is able to identify the
effective switching configuration for loss minimization.
No. of branches 32 Therefore, the proposed method can be applied to find the
No. of tie-line 5 optimal or near optimal solution for network of any size
. Svstem load (Active)
. 3715 kW and complexity, even though the computational time for
System load (Reactive) I 2300kVAr larger system is relatively higher than that of the heuristic
System loss for base configuration I 210.00 kW or approximate methods. However, optimal or near
1II.A Test Results optimal solution is guaranteed in the proposed method.
The proposed method is tested on the 33-bus test system Therefore, the proposed Switching Algorithm can be used
using Borland C++. The data pertaining to the base for real-time on-line implementation.
configuration of the system is fed to the algorithm. In the Table 111: Comparison with Other Methods using 33-bus Test Data
first stage of the proposed method, the search is limited to
the tie-lines and their two adjoining branches only which Methods Loss-minimum
generates the switching combinations (3” , where n is the investigated for solution obtained
33-bus system (Branch in-out) Simplified Distflow
number of tie-lines). Each of the switching combinations
will be converted to switching configuration using the Distflow
concept of zero-connect, left-connect and right-connect. Base 210.00 kW 223.50 kW
configuration
The configuration with minimum loss is identified from Proposed GSSA 33-7, 34-14, 35- 155.10 kW 154.40 kW
the feasible configurations. In the next stage, the search I 1,36-32,37-28
has been carried out by allowing the investigation to move Goswami & Basu 33-7, 34-9, 35- 157.50 kW 155.60 kW
to the left or right of each open branch of the configuration . 151 14,36-32
Shirmohammadi 33-7, 34-10, 35- 158.2 kW 156.30 kW
obtained above, depending on whether each switch-group & Hong [4] 14,36-32
is in right-connect or left-connect status. The test results BW & W U - 35-11,36-31, 37- 167.00 kW 163.00 kW
obtained by the proposed algorithm are given in Table 11. - MI [21 28
The overall loss minimum configuration for the test system B~UWI& WU - 33-6. 35-11. 36- 178.40 kW 170.40 kW
is obtained which gives the optimal or near optimal - M 2 121 31
solution for loss minimization.
B w & ~ WU - 33-6, 35-11, 36- 178.40 kW 170.40 kW
M3 P I

1II.C Discussions
33-bus test system The proposed algorithm is most suited to be used with
. Loss in the base configuration 210.00 kW SCADA (Supervisory Control and Data Acquisition
. Loss in the optimal or best- 155.10 kW Systems) and DAC (Distribution Automation and Control)
configuration as the algorithm requires only the inputs to be supplied and
Ootimal Solution I 33-7,34-14,35-11,36-
-
(branch in out) 32,37-28 any other subsequent manipulations such as partitioning,
Loss reduction 54.90 kW forming loops or sub-systems etc. are not necessary. In
’ Execution time 4.56 sec most of the methods available so far, some form of
(Pentium 200MHz ) I heuristics is employed and near optimal or optimal
solution is obtained. In these methods the implementation
1II.B Comparative Study of the loss minimization has not been fully automated. As
The 33-bus system results obtained by the proposed the loss minimum configuration has been obtained by
method are also compared with the results of the methods some heuristics and approximations, these methods do not
in [2], [4], and [5]. The solutions (switching-options) guarantee global optimum. Since the heuristic methods
obtained by the proposed method and the methods in [2], depend on the loads at a particular time, and as loads vary
[4], and [5] are examined for loss minimization by using with time, they are not reliable in load varying conditions.
simplified Distf3ow and Distflow solutions. The real power Whatever may be the combination of consumer types, the
losses for the base configuration and the loss minimum proposed method will offer reliable solutions for the loss
solutions of the 33-bus system obtained by the proposed minimization problems by properly reconfiguring the
method and the methods of [2], [4] and [5] are computed distribution systems.
and presented in Table 111. The proposed method has found The proposed method can be applied for proper planning
the appropriate switching-options that give the optimal and efficient operation for on-line systems, since any
configuration for loss minimization. The switching-options change in loads does not need any change to be made in
found by the proposed method are 33-7, 34-14, 35-11, 36- the algorithm and the optimum switching configuration
32, and 37-28. It is also found that the proposed method can be easily found without much effort. The problem of
gives the maximum loss reduction in the two different non-coincidence of peak loads and diversity of load
power-flow solutions considered. Test results show that categories are not a matter of concern in the proposed

254
system. Since simultaneous evaluation of power losses in IEEE Trans. on Power Delivery, Vo1.7, No.3, pp.1484-1491, July
all the branches are done for a set of switching 1992.
6. R. P. Broadwater, A. H. Khan, H. E. Shaalan and R. E. Lee, Time
combinations, the method is more reliable and gives better Varying Load Analysis to Reduce Distribution Losses through
results. Also, the method gives a lot of flexibility to limit Reconfiguration', IEEE Trans. on Power Delivery, Vo1.8, No.1,
or extend the search simply by specifying the branch pp.294-300, January 1993.
numbers to which the search is to be done on both ends of 7. C. T. Huddleston, R. P. Broadwater and A. Chandrasekaran,
'Reconfiguration Algorithm for Minimizing Losses in Radial
each tie-switch in the system. The proposed method can be Electric Distribution Systems', Electric Power System Research,
reliably employed for loss minimization as it has been Vol. 18, NO.1, pp.57-67, 1990.
found to be most effective and gives the best optimal 8. G. P. Peponis, M. P. Papadopoulos and N. D. Hatziargyriou,
solutions compared to the methods available for loss 'Distribution Network Reconfiguration to Minimize Resistive Line
Losses', lEEE Trans. on Power Delivery, Vol.10, No.3, pp.1338-
minimization by reconfiguring the radial distribution 1342, July 1995.
network. 9. R. Taleski and D. Rajicic, 'Distribution Network Reconfiguration
for Energy Loss Reduction', IEEE Trans. on Power Systems,
Vo1.12,No.l, pp.398406, February 1997.
IV.Conclusion 10. M. A. Kashem, M. Moghavvemi, A. Mohamed and G. B. Jasmon,
A switching algorithm is formulated for network 'Loss Reduction in Distribution Networks using New Network
reconfiguration based loss reduction, and a new solution Reconfiguration Algorithm', Electric Machines and Power Systems,
method is proposed in this paper. To reduce the Vo1.26, No.8, pp.815-829, October 1998.
11. B. C. G. Shin, 'Development of the Loss Minimization Function for
combinatorial explosive switching problems into a Real time Power System Operations: A New Tool', IEEE Trans. on
realizable one, an efficient technique has been introduced Power Systems, vo1.9, No.4, pp.2028-2034, November 1994.
for reconfiguration of distribution networks. In the first 12. V. Borozan and N. Rajakovic, 'Minimum Loss Distribution
stage, the tie-branches and its two adjoining branches are Network Configuration: Analyses and Management', Proc. of IEE
Conference, pp.6.18.1-6.18.5, 2-5 June 1997.
considered to generate switching combinations and the 13. M. A. Kashem, 'V. Ganapathy, G. B. Jasmon, 'Network
best combination among them is identified. In the second Reconfiguration for Enhancement of Voltage Stability in
stage, the search has been extended to find whether there is -
Distribution Networks', to appear in IEE Proceedings Generation,
any other switching option that may give minimum loss. Transmission and Distribution.
To improve the accuracy and to rule out the possibility of Biographies
leaving out any other better switching combination, the
search is done by considering switching options beyond M. A. Kashem was bom in Bangladesh in 1968. He received his B.Sc. in
the two neighboring branches of each of the tie-switches. Electrical & Electronic Engineering from Bangladesh University of
Engineering & Technology (BUET) in 1993. He has completed his
This is done by specifying the branch numbers to which Masters in Engineering Science from the University of Malaya, Malaysia
the search is extended along with each of the tie-switch at 1997. He worked in a power system consultancy firm for one year.
specifications in the input. The method has been compared Currently he is working as a Lecturer in the Faculty of Engineering,
with five other methods and a comparative study is Multimedia University, Malaysia and pursuing his Doctoral study in the
presented. From the test results and comparative study it same University. His research interests are distribution system
automation, power system planning, neural networks and computer
can be concluded that the proposed method can effectively networking and he has published more than seventeen papers in these
identify the appropriate switches for loss minimization areas.
with minimum computational effort. The technique can be V. Ganapathy was bom on In May 1941 in Salem, India He received his
Bachelor of Engineering and Masters or Engineering degrees in the years
effectively installed in real time on-line systems under 1964 and 1971 respectively from the University of Madras, India He got
widely varying load conditions, and this is suitable for his Ph.D. fiom the Indian Institute of Technology, Madras, India in the
power system planning and operations. The proposed year 1982. He has the experience of two years in industries and more than
method is suitable for continuous reconfiguration for loss thirty-four years in teaching profession. His research interests are in
neural networks, digital signal processing, power systems,
reduction. And it can also be used for planning and design microprocessors, computer graphics and robotics. He has published more
of power systems to reduce installation and switching than fifty papers in the above areas. He is currently with the Faculty of
costs. Information Technology, Multimedia University, Malaysia.
Ghauth Jasmon was bom in Malaysia in 1956. He graduated with
References: BSc(Eng) and PhD degrees in 1979 and 1982 respectively from the
University of London, UK. His research areas are in power systems
1. S . Civanlar, J.J. Grainger, H. Yin and S . S . H. Lee, 'Distribution security, stability, load management and voltage stability, and has
Feeder Reconfiguration for Loss Reduction', IEEE Trans. on Power published more than sixty papers in these areas. He is currently the Chief
Delivery, Vo1.3, No.3, pp.1217-1223, July 1988. Executive and President of Multimedia University, Malaysia He is a
2. M. E. Baran and F. F. Wu, 'Network Reconfiguration in Distribution Senior Member of the IEEE, Fellow of the IEE(UK) and Fellow of the
Systems for Loss Reduction and Load Balancing', IEEE Trans. on Institute of Engineers, Australia"
Power Delivery, Vo1.4, No.2, pp.1401-1407, April 1989. M.I. Seyed Mohamed Buhari was bom on 24th April 1974. He did his
3. A. Merlin, H. Back, 'Search for a Minimal-Loss Operating Spanning Bachelors Degree in Computer .Engineering at Mepco Schlenk
Tree Configuration for an Urban Power Distribution System', Proc. Engineering College, India and his Masters Degree in Computer Science
Of the 5* Power System Conference (F'SCC), pp.1-18, Cambridge, and Engineering from the PSG College of Technology, India. Currently
1975. he is pursuing his Doctoral Programme at the Multimedia University,
4. D. Shirmohammadi and H. W. Hong, 'Reconfiguration of Electric Malaysia. Till date, he has presented nine technical papers in
Distribution Networks for Resistive Line Losses Reduction', IEEE Intemational conferences. His areas of interest are Computer Networks,
Trans. on Power Delivery, vo1.4, No.2, pp.1492-1498, April 1989. Operating Systems and Parallel Processing.
5. S . K. Goswami and S . K. Basy 'A New Algorithm for the
Reconfiguration of Distribution Feeders for Loss Minimization',

255
Appendix: 33-bus Test System
-
26 25 0.2842 0.1447
-
Br.
-
Sen. Rec. Real
-- -
Reac. Resis- Reac-
27
28 28 120
20
70
1.0590
0.8042
0.9337
0.7006
no. node node load load tance tance 29 28 29 200 600 0.5075 0.2585
-
1
-
0
-
1
-- kw
100
kVAr
60
ohms
0.0922
ohms
0.0470
30
31
29
30
30 150
210
70
100
0.9744
0.3105
0.9630
0.3619
2 1 2 90 40 0.4930 0.251 1 32 32 60 40 0.3410 0.5302
3 2 3 120 80 0.3660 0.1 864 33* 20 2.0000 2.0000
4 3 4 60 30 0.3811 0.1941 34* 14 2.0000 2.0000
5 4 5 60 20 0.8190 0.7070 35* 11 21 2.0000 2.0000
6 5 6 200 100 0.1872 0.6188 36* 17 32 0.5000 0.5000
7
8
6
7
7
8
200
60
100
20
1.7114
1.0300
1.2351
0.7400
-
37* 24
* Tit lines
28 0.5000 0.5000

9 8 9 60 20 1.0440 0.7400 Subs:tation Voltage (kv) = 12

10 9 10 45 30 0.1966 0.0650
11 10 11 60 35 0.3744 0.1238
12 11 12 60 35 1.4680 1.1550
13 12 13 120 80 0.5416 0.7129
14 13 14 60 10 0.5910 0.5260
15 14 15 60 20 0.7463 0.5450
16 15 16 60 20 1.2890 1.7210
17 16 17 90 40 0.7320 0.5740
18 1 18 90 40 0.1640 0.1565
19 18 19 90 40 1.5042 1.3554
20 19 20 90 40 0.4095 0.4784
21 20 21 90 40 0.7089 0.9373
22 2 22 90 50 0.4512 0.3083
23 22 23 420 200 0.8980 0.7091
24 23 24 420 200 0.8960 0.701 1
25 5 25 -60 25 --
0.2030 0.1034

256