EE432: Elements of Power System Engineering

Optimal reconfiguration of distribution net-work

Project Deliverable 1

Index Name ID Section

1
2

( Course Professor)

Contents
1. Abstract……………………………….
2. Introduction…………………………………………………….
Problem definition…………………………………………………
Criteria and constraints…………………………………………………………………….

Objectives………………………………………………………….

3. Available solutions……………………………………………………………………………………..
 Brain storming…………………………………………………………………………………………….
 Advantages ………………………………………………………………………………………………..
 Dis advantages…………………………………………………………………………………………….

4. Back ground and theory………………………………………

5. Effect of net-work reconfiguration on the voltage profile…………………………
6. Impact of reconfiguration on the joule losses…………………………………………….
7. Effect of generation insertion on the voltage profile………………………………….
8. Net-work reconfiguration using Genetic Algorism……………………………………..
9. Results and flow chart……………………………………………………………………………….
10.Conclusion………………………………………………… …………………………………………….
11.References……………………………………………………………………………………………….
12.Appendix……………………………………………………………………………………………………

Abstract:
In this project, we will study one of the most fundamental objects of power
system analysis which is reconfiguration of the power system net-work.
Distribution net-works can be connected using any type of the known net-work
configurations like radial, network, ring loop or mesh configuration. Radial
configuration has low cost and considered a simple connection which can be
duplicated, but the problem of this configuration, is during the fault; if a fault
happened at any transmission line or source, the all net-work will be out of
service. Network connection is considered a configuration which is more complex
than the radial, and it gives us the ability to connect two sources in parallel with
the substation to avoid the network to be out of service during faults. Ring
configuration is the configuration where we can find two sources are connected
every plant substation and protection components at the power inlet and outlet
which means if a fault occur, only the faulted line will be out of service not the all
network; but this configuration is very cost and complex.

In this report, we have a network of six nods which have a radial configuration
and we will use power flow analysis to make reconfiguration to the network to
find the configuration that gives the minimum losses. A MATLAB code will be
written for power flow analysis equations to compute the results of voltage and
power losses of the net-work.

Introduction
Problem definition:

Voltage stability of the electrical power system network is one of the fundamental
factors that affects the distribution system which must kept close to the rated
value as possible as we can. There are a lot of reasons of voltage variation like,
the variation of the load connected to the net-work, atmospheric phenomena or
any trip of the net-work caused by the protection equipment. Always the radial
configuration is used in the distribution system as it will be easy to find the faults
and deal with it quickly and also, give us the ability to management the network
power flow. When the net-work is radial configurated then we can add a source
and if a fault happened, we can isolate it and supply the load at the same time
and make reconfiguration to the net-work. So, we can make reconfiguration to
the power system distribution network for many reasons; to improve the quality
of its service, to enhance the reliability, to reduce the power losses otherwise,
using a generator insertion in the network to supply the loads is a process may
have some problems if the power that injected by the generator is much larger
than that demanded by the load. So, making reconfiguration to the network will
be the best solution.

Criteria and constraints:

 Reducing the power losses to the minimum value:

N br
Ri∗P2i + Q2i
Plosses=∑
i=1 V 2i

 The value of the voltage must kept within the specified range:
V i .mini ≤ V i ≤V i . max
 The value of the current must kept within the limited range:
I i ≤ I i . max
 The number of the nodes must kept constant to supply all the consumers.

Objectives:
 To studying net-work reconfiguration
 To learn how to make the calculations of the load flow analysis.
 To learn how to improve the voltage profile using reconfiguration.
 To learn how to reduce the losses of the power system using reconfiguration of
the network.

Available solutions:
Brain storming:
1. Inserted generators to the distribution system:
To decrease the real part of the losses is the main aim of this method which
works on voltage enhancement, achieving stability of the system for high
efficiency service of the distribution system.

2. Reconfiguration of the net-work:

Direct dealing with the feeders through the switches which is in the beginning
and in the end of the transmission lines to disconnect the line with high losses
leads to low losses for all the system.

Advantages and inconvenient:

1. The inserted generator has the advantage of helping to reduce the
power losses and the voltage increment for the connected bus but it
also may case an unbalanced voltage if the voltage has a large
increasing.
 2. The method of reconfiguring the power system lines can restore the
energy in a fast way and is not expensive also, it is easy to be performed.

Back ground and theory:

Gauss Sidle method for power flow analysis:

1. Building he matrix Y bus = [n×n];

1 −1
 Zij =P±jx y ij= Y ij =− y ij =
Z ij Z ij
−1
 Y 12=− y 12= Z
12

 Y ij =∑ y ij For example, Y 12= y 12 + y 13+ y14 + y15 + y 15

Y 11 Y 12 Y 13 Y 14 Y 15 Y 16
Y 21 Y 22 Y 23 Y 24 Y 25 Y 26
Y 31 Y 32 Y 33 Y 34 Y 35 Y 36
Y 41 Y 42 Y 43 Y 44 Y 45 Y 46
Y 51 Y 52 Y 51 Y 53 Y 54 Y 55
Y 61 Y 62 Y 63 Y 64 Y 65 Y 66

Step (2): assuming Initial values:

V=1v the unknown voltage.
δ=0 the unknown angle
(3): Apply G.S.M equation for every buss:

1 P −J Q
n
V i= ∗[ SCi ¿ sci − ∑ Y ik∗V k ]
Y ii Vi k=1
k≠ i

For first iteration of the voltage at buss (1),:

(1 ) 1
V2 = ∗¿
Y 22
 (1 ) 1
V3 = ∗¿
Y 33

(1 ) 1
V4 = ∗¿
Y 44

(1 ) 1
V5 = ∗¿
Y 55

(1 ) 1
V6 = ∗¿
Y 66

(4): Determine the power known and unknown values:

PSC =P g−P d Qsc =Q g−Qd

 For load bus,

 PSC =P g−P d=0−Pd = - Pd
 Qsc =Q g−Qd =0−Qd=−Qd

For generator buss:

PSC =P g−P d

Qsc =Q g−Qd

To calculate the reactive power:

(1 )
Q 5 =−I m ¿

The correction equation:

(1 ) |V i|∗V 1i
V =
|V (1)
i |
i . corr

Use the updated values for the second iteration:

(2 ) 1
V2 = ∗¿
Y 22
 (2 ) 1
V3 = ∗¿
Y 33

(2 ) 1
V4 = ∗¿
Y 44

(2 ) 1
V5 = ∗¿
Y 55

(2 ) 1
V6 = ∗¿
Y 66

Checking the condition to stop the iterations through the equation:

K+ 1 K
Vi −V I < ϵ ∘

Use the equation of the acceleration factor equation to accelerate the iteration:
k ( K−1)
V i =V I + ƛ∗¿

Where ( ƛ ): is the acceleration factor

1. . Investigate the importance reconfiguration of the net-work

from the following point of view:
 Node voltage profile:

Power flow analysis is one of the power system pre-installation tools. It is enables
us to evaluate the voltage, demanded and generated real reactive power at every
bus and the current at the transmission lines. It enables us to imagine the final
shape of the net-work and confine the number of sources, transmission lines, its
length, the protection equipment we need for the protection system and the
method of configuration we will use in the net-work. So, a configuration of the
net-work must be established to deliver the power to the consumers. But the
nature of the load that it isn’t fixed its always vary from time to time or from area
to area, on the other hand the distribution equipment has a rated fixed values
and the variation of the load may cause damage to the equipment due to
consuming more than the rated current or due to losses. So. The transmission
lines should be replaced to suit the new demand of the loads. Operating this
replacement is done using switches to disconnect a transmission line and connect
another one. The process of changing transmission lines has a main effect on the
total admittance of the network which results in changing the voltage at every
bus of the network, so, reconfiguration of the net-work can result in an
improvement of the voltage profile and reducing the power losses which rise the
efficiency of the network and enhance its reliability.

B. joule losses:

Using power flow analysis methods, we can recognize and determine the value of
losses in the equipment that forming the distribution network.

Classification of power losses in the network:

1. Non-technical losses.
2. Technical losses
 Fixed technical losses: The losses that occur due to corona losses,
leakage current losses, elements used in measurements or control
system.
 Variable technical losses: It is that power lost in the form of heat
because of the resistance of the transmission lines or the
inductive loads connected to the distribution net-work
To reduce the joule losses, we have some practical techniques like instilling
capacitor banks, installing transformers with rated power close to the required
demanded power of the network or connecting generators to the distribution
system for compensation of energy losses but all of these methods re very
expensive to be operated. Recently, reconfiguration of the net-work is method
with least cost which depends on exchanging the transmission lines with
maximum power losses and as a result, reconfiguration of the network reduces
the joule losses of the transmission lines.

c. Distributed generators insertion:

The importance of power flow analysis for the generation insertion in the
distribution net-work system lies in the long distance between the generation
center and the load we should deliver the energy to it. This long distance requires
long transmission lines so as to reach the load and that leads to more power lost
in the transmission line and large voltage drop. To solve this problem, we can
choose one of several solutions like; installing a capacitor or voltage regulator or
even reconductor the transmission lines in other words; recon figurate the
distribution network. But there is another solution which is inserting a generation
source connected to the load bus to solve the problem of power losses and to rise
the stability of the distribution system.

So, the effect of inserting a voltage source on the configuration of the net-work is
that the limitation of the length of the transmission lines, the ability of changing
the supplied voltage when the load changes quickly as the bus that will connect to
the generation source will operates as a voltage-controlled bus.
2. Propose a general algorithm to perform reconfiguration for a given
network

Too perform reconfiguration of the net-work, the process entails manually or

automatically altering the status of the switches of the feeder, such as opening
sectionalized switches and closing tie switches while keeping the radial
configuration and ensuring that all associated loads are supplied.

The method' principal goal is to select the status of the switch which give us the
least amount of system losses.

The strings of the genetic algorism are formed by fitness functions, tie switches,
and sectionalizing switches.

Applying the algorism on the following example:

Figure (1): Diagram of the distribution system.

For the system shown in figure (1); as we can see, the sectionalized transmission
lines are solid and the tied lines are dotted which are (20, 21, 22). The process of
reconfiguration can be performed when we change a sectionalized switch to be
opened or a tie switch to be closed. The branch number 21 has a tie switch, if we
close it, the system will has a loop connection but when we open a sectionalized
switch like branch 7th switch then, the radial configuration will be restored.

Power flow equations:

(Dist.) equations will be used for the network load flow analysis as follows:
2 2
Pi +Qi
Pi+1 =P i+ r i 2
−P Li +1
Vi

2 2
P i +Qi
Qi+ 1=Qi+ xi 2
−Q Li+ 1
Vi

2 2
P i +Q i
V i+ 1=V i −2 ¿*r i +Qi xi )+( r i + x i)
2 2
2
Vi

Where:
'
Pi=Pi + P Li
'
Qi=Qi +Q Li

'
i i L i
P P P = +
( 9
'
i i L i
Q Q Q = +
(
After that, we will write a coding scheme where we can write the open switches
(0) and the closed switch as (1). The system we has contains 13 closed switch and
3 open switch so we need 16 bits for 16 chromosome to establish the string.

the encoding, 16 bits has to be available for the chromosome string as shown in
the following table;

Table 1: Coding scheme of the power system

1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0

Steps of the genetic algorism preformation:

 At first, the information of the given distribution system should be

determined.
 Secondly, the status of every switch should be written by writing
the closed one as (1) and the open one as (0) to know how many
iteration will be calculated in the algorism.
 At this point, if the net-work is not loop configuration, we should
stop the algorism
 Find the initial population for optimal radial reconfiguration of the
net-work.
 Then, we calculate the fitness population.
 Cross over must be performed by choosing two chromosomes
from the population. For the cross over point a number must be
chosen randomly then the cross over operation should be
performed for the lift side of the bit.
 Then, we should apply the Mutual operation by choosing two
numbers randomly so as to mute the chromosome and change
the bits from (1) to (0) and from (0) to (1).
 If the radial connection doesn’t restored, the off spring should be
replaced.
 Then we need the population to be shorted and the string to be
off.
 Best size of individual population is chosen with the help of the
fitness equation.
 The maximum iteration can be achieved by repeating the steps
from 6 to 10.
 Maximum and minimum values of the system losses and the
fitness function can be displayed and also the minimum losses
status of the switch.
 Then, we stop the iteration.
Flow chart:
Results of configuration:

Results of the power losses of the net-work.

Reconfiguration of the net-work:

1. First configuration:
 UN acceptable value of the losses so, we will try to find another
reconfiguration which reduces the losses.
2. The second configuration :
Now, we will open the lines (1-5), (2-5), (3-5):

As shown in MATLAB results, the losses has been reduced to an acceptable

value of p=1.7216W

3. The third configuration:

 In this configuration we opened the feeders between buses (2,5)
(2,3), (4,5) but the value of the losses is high.

From the previous results, we notice that the best configuration which
gives the minimum losses is the second one which reduces the losses to
P=1.7216W.

Conclusion:
Our project interests in improving the distribution net-work, how to
achieve stability of the voltage profile and how to reduce the amount of the
power losses of the system. Achieving this goals rises the efficiency
transferring energy, enhancing the reliability of the system and good
service to the consumers. In this report, we have focus on the a widely
used way to achieve the goals we have mentioned; this method is line
reconfiguration of the distribution system which is one of several methods
that can be used to improve the voltage profile and reduce the power
losses of the system which can be an inserted generator or shunt
capacitor. The process of reconfiguration has some constraints which must
be considered for a successful process one of them is that we should keep
the voltage within the range of the minimum and the maximum values, and
also the current, moreover, we should make sure that all the loads are
supplied. Two types of reconfiguration topologies are used in distribution
system, loop or radial. Unstable distribution system leads us to change the
reconfiguration of the system from loop to radial. Radial configuration is a
connection were we have no loop in the system, the transmission line
straights through the service area and supplied from one source, We can
change the connection of the transmission lines so as to get minimum value
of the power losses. Changing the connection of the lines has a noticeable
effect on the voltage profile and the joule losses of the feeders. This effect
has been illustrated in the analysis of the given example of the network
using MATLAB tool where we have made several configuration on the net-
work to choose the optimal configuration which achieves the minimum
value of the losses
References:

 Acadian, M., Farcing, M. M., & Nezamabadi-pour, H. (2007, May).

Optimal Reconfiguration of Distribution System by PSO and GA using
graph theory. In International Conference on Applications of
Electrical Engineering (pp. 83-88).

 Shirmohammadi, D., & Hong, H. W. (1989). Reconfiguration of

electric distribution networks for resistive line losses reduction. IEEE
Transactions on Power Delivery, 4(2), 1492-1498.

 Bahia-El-Din, W. H., & Liu, M. T. (1992). Register-insertion: a protocol

for the next generation of ring local-area networks. Computer
networks and ISDN systems, 24(5), 349-366.

 Miguel, E., Cedars, J., Diaz-Dorado, E., & García-Dornelas, J. L. (2002).

An improved branch-exchange algorithm for large-scale distribution
network planning. IEEE Transactions on Power Systems, 17(4), 931-
936.

 Mutable, J., Strbac, G., Curcic, S., & Jenkins, N. (2000). Allocation of
losses in distribution systems with embedded generation. IEE
Proceedings-Generation, Transmission and Distribution, 147(1), 7-14.

Appendix: