(AUTONOMOUS)

SreeSainath Nagar, A.Rangampet - 517 102

VISION
To become the Nation’s premiere centre of excellence in electrical engineering through
teaching, training, research and innovation to create competent engineering
professionals with values and ethics.

MISSION
 To foster intellectual curiosity, pursuit and dissemination of knowledge.
 To explore students’ potential through academic freedom and integrity.
 To promote technical mastery and nurture skilled professionals to face competition
in ever increasing complex world.

QUALITY POLICY
Sree Vidyanikethan Engineering College strives to establish a system of Quality
Assurance to continuously address, monitor and evaluate the quality of education offered
to students, thus promoting effective teaching processes for the benefit of students and
making the College a Centre of Excellence for Engineering and Technological studies.

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING

Vision
To become the Nation’s premiere centre of excellence in electrical engineering through
teaching, training, research and innovation to create competent engineering
professionals with values and ethics.
.

Mission
 Department of Electrical Engineering strives to create human resources in
Electrical Engineering to contribute to the nation development and improve the
quality of life.
 Imparting Knowledge through implementing modern curriculum, academic
flexibility and learner centric teaching methods in Electrical Engineering
 Inspiring students for aptitude to research and innovation by exposing them to
industry and societal needs to creating solutions for contemporary problems
 Honing technical and soft skills for enhanced learning outcomes and employability
of students with diverse background through comprehensive training
methodologies
 Inculcate values and ethics among students for a holistic engineering professional
practice.

M. Tech. (Electrical Power Systems)

Program Educational Objectives

Within a few years of graduation, graduates will

1. Have enrolled in doctoral studies or engage in research activities of societal
importance.
2. Assume key positions in research divisions, industry and academia.
3. Advance professionally through continuing education, ethics and values.
Program Outcomes
After Successful completion of the Programme, post graduates will be able to

1. Acquire in-depth knowledge of specific discipline or professional area, including

wider and global perspective, with an ability to discriminate, evaluate, analyse
and synthesis existing and new knowledge, and integration of the same for
enhancement of knowledge. (Scholarship of Knowledge)
2. Analyze complex engineering problems critically, apply independent judgement
for synthesizing information to make intellectual and/or creative advances for
conducting research in a wider theoretical, practical and policy context. (Critical
Thinking)
3. Think laterally and originally, conceptualise and solve engineering problems,
evaluate a wide range of potential solutions for those problems and arrive at
feasible, optimal solutions after considering public health and safety, cultural,
societal and environmental factors in the core areas of expertise. (Problem
Solving)
4. Extract information pertinent to unfamiliar problems through literature survey
and experiments, apply appropriate research methodologies, techniques and
tools, design, conduct experiments, analyse and interpret data, demonstrate
higher order skill and view things in a broader perspective, contribute
individually/in group(s) to the development of scientific/technological knowledge
in one or more domains of engineering. (Research Skill)
5. Create, select, learn and apply appropriate techniques, resources, and modern
engineering and IT tools, including prediction and modelling, to complex
engineering activities with an understanding of the limitations. (Usage of
modern tools)
6. Possess knowledge and understanding of group dynamics, recognise
opportunities and contribute positively to collaborative-multidisciplinary scientific
research, demonstrate a capacity for self-management and teamwork, decision-
making based on open-mindedness, objectivity and rational analysis in order to
achieve common goals and further the learning of themselves as well as others.
(Collaborative and Multidisciplinary work)
7. Demonstrate knowledge and understanding of engineering and management
principles and apply the same to one‟s own work, as a member and leader in a
team, manage projects efficiently in respective disciplines and multidisciplinary
environments after considerisation of economical and financial factors. (Project
Management and Finance)
8. Communicate with the engineering community, and with society at large,
regarding complex engineering activities confidently and effectively, such as,
being able to comprehend and write effective reports and design documentation
by adhering to appropriate standards, make effective presentations, and give
and receive clear instructions. (Communication)
9. Recognise the need for, and have the preparation and ability to engage in life-
long learning independently, with a high level of enthusiasm and commitment to
improve knowledge and competence continuously. (Life-long Learning)
10. Acquire professional and intellectual integrity, professional code of conduct,
ethics of research and scholarship, consideration of the impact of research
outcomes on professional practices and an understanding of responsibility to
contribute to the community for sustainable development of society. (Ethical
Practices and Social Responsibility)
11. Observe and examine critically the outcomes of one‟s actions and make
corrective measures subsequently, and learn from mistakes without depending
on external feedback. (Independent and Reflective Learning)

Program Specific Outcomes

After Successful completion of the Programme, post graduates will

1. Demonstrate specialized knowledge in Electrical power systems, its operation

and control with an ability to combine existing and recent practices.
2. Analyze and solve complex problems to obtain optimal solution in power system
operation and control to meet the needs of industry and society.
3. Demonstrate research competence in power system to design innovative
products and provide services in the field of electrical power systems and
related areas.
4. Apply modern tools, techniques and resources to provide solutions to complex
engineering.
 A Project Report on

ANALYSIS OF COST BY MITIGATING THE

POWER QUALITY PROBLEMS AT
DISTRIBUTION SYSTEM USING
OPTIMIZATION TECHNIQUE
submitted to
JAWAHARLAL NEHRU TECNOLOGICAL UNIVERSITY ANANTAPUR,
ANANTHAPURAMU
in partial fulfillment of requirements for the award of the degree of
MASTER OF TECHNOLOGY
in
ELECTRICAL POWER SYSTEMS
by
MUCHUGUTTA RAMALINGA REDDY
(18121D0710)
Under the esteemed guidance of
Mr. U. KAMAL KUMAR, M.Tech., (Ph.D.)
ASSISTANT PROFESSOR,
DEPARTMENT OF EEE

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING

(An autonomous institution affiliated to JNTUA, Anantapuramu)

Sree Sainath Nagar, Tirupati – 517102.
2018 – 2020
 (An Autonomous institution affiliated to JNTUA, Anantapuramu)
Sree Sainath Nagar, Tirupati-517 102.

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING

CERTIFICATE

This is to certify that the report on the Project work entitled “ANALYSIS
OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTRIBUTION SYSTEM USING OPTIMIZEATION
TECHNIQUE” submitted by MUCHUGUTTA RAMALINGA
REDDY (18121D0710) to Department of EEE, Sree Vidyanikethan
Engineering College, an autonomous institution affiliated to Jawaharlal
Nehru Technological University Anantapur, Anantapuramu in partial
fulfillment of the requirements for award of degree of Master of
Technology in Electrical Power Systems, is a bonafide work carried out
and submitted by him under my supervision.

Supervisor: HOD:
Mr. U. KAMAL KUMAR, M.Tech., (Ph.D.) Dr. M.S. SUJATHA, M.Tech., Ph.D.
Assistant Professor Professor
Department of EEE Department of EEE
Sree Vidyanikethan Engineering College Sree Vidyanikethan Engineering College
A. Rangampet. A. Rangampet.

EXAMINER – I EXAMINER – II

Place: Tirupati

Date:
M. RAMALINGA REDDY
Roll No.: 181121D0710
M.Tech. Student
Department of Electrical and Electronics Engineering
Sree Vidyanikethan Engineering College, Tirupati

Declaration

I hereby declare that the project report entitled “ANALYSIS OF COST BY

MITIGATING THE POWER QUALITY PROBLEMS AT DISTRIBUTION
SYSTEM USING OPTIMIZATION TECHNIQUE” has been submitted by me under
the guidance and supervision of Mr. U. KAMAL KUMAR, Assistant Professor,
Department of Electrical and Electronics Engineering, Sree Vidyanikethan
Engineering College, Tirupati. This work / report has not been submitted either for the
award of any degree or any other similar title.

Place: Tirupati (M. RAMALINGA REDDY)

Date:
 ACKNOWLEDGEMENTS

I would like to express my profound sense of gratitude to

Mr. U. KAMAL KUMAR, M.Tech., (Ph.D.) Assistant Professor, Department of EEE
for his valuable guidance and constructive suggestions made at various levels of the
work.

I am thankful to Dr. M.S. SUJATHA, M.Tech., Ph.D., Professor and HOD,

Department of Electrical and Electronics Engineering, for her guidance, constant
encouragement and support.

I express my gratitude to beloved Principal Dr. P.C. KRISHNAMACHARY, M.Tech.,

Ph.D., and the Management of SVEC for providing all the facilities in completing
project successfully. My heartfelt thanks to all my Teachers in the department of EEE
of Sree Vidyanikethan Engineering College for their moral support and good wishes.

Finally, I would like to express my sincere thanks to my parents, friends, one and all
those who guided, inspired and helped me in completion of my project work.

M. RAMALINGA REDDY
 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

ABSTRACT
In recent days the distribution network affected with load uncertainties, which
ever adhere the power quality problems. The quality problems are isolates the healthy
network into unbalanced and collapse the devices in the different factors like, damages,
power factor variations and changes in the tariffs rates at end user side etc. Mostly due
to sudden load variations the voltage dips and long duration interruption will cause to
changes in cost at utility and end user. This work proposes the analysis of cost under
the power quality problems. Therefore in this scenario the quality of supply may
enhanced using distributed generations (DGs). The another approach is to sizing and
placement of DG, so far an optimized technique is proposed. The impacts of DG
placement at distributed networks are also analyzed. This proposed work towards the
detailed analysis of tariff at uncertainties and provide the best solution to mitigate the
power quality problems using optimized technique with the help of
MATLAB/SIMULINK software.

Keywords: Power quality, Distribution generation (DG), Photovoltaic (PV)

Generation.

DEPT. OF EEE Page v

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

CONTENTS
Page
S.No. Title
No.
Acknowledgments iv
Abstract v
List of Tables viii
List of Figures x
List of Abbreviations xi

Chapter 1 Introduction

1.1 Introduction 1
1.2 Literature survey 2
1.3 Problem Identification 3

1.4 Objectives 4

Chapter 2 Mathematical Analysis of Radial Distribution System

2.1 Introduction 5
2.2 Load Flow Methods for Calculation of Distribution System 5
2.3 Backward forward sweep method 6
2.3.1 Algorithm for BFS method 6
2.3.2 Test System 7
2.3.3 Formulas used for BFS method 8
2.3.4 Flowchart BFS method based on branch current 10
2.4 Tariff Calculation 11
2.5 Test system load flow calculation 11

Chapter 3 Power Quality Problems And Distribution Generation

3.1 Power Quality 13

3.1.1 Definitions 13
3.1.2 Types of power quality problems 13
3.2 Distribution Generation 17
3.2.1 Definition 17

DEPT. OF EEE Page vi

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

3.2.2 Types of DGs 18

3.2.3 Various distributed generation capacities 19
3.2.4 Benefits of using DG 19
3.3 DG Rating Constraints 20
3.4 Real Power and Bus Voltage Constraints 20
Distribution System Load Models And Proposed
Chapter 4
Algorithm

4.1 Load Modeling in Distribution System 21

4.2 Particle Swarm Optimization 22
4.2.1 Simulation procedure of PSO 26

Chapter 5 Results

5.1 Load flow calculation without DG 27

5.2 Load flow calculation with DG 30
5.3 Tariff calculation 34
5.4 Comparison of Results 37
5.4.1 Load flow results comparison 37
5.4.2 Load Modeling results comparison 37
5.4.30 Power Quality problems results comparison 38

Chapter 6 Conclusion and Future Scope of Work

6.1 Conclusion 39
6.2 Future scope of work 39
References 42

DEPT. OF EEE Page vii

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

LIST OF TABLES
Figure Page
Title
No. No.
Table 2.1 : Test system data 7
Table 2.2 : Load flow calculation of IEEE 15 Bus system 11
Table 2.3 : Tariff calculation of IEEE 15 bus system 12
Table 3.1 : Classification of power quality problems as per IEEE 13
standards

Table 3.2 : Types of DG and their models 18

Table 4.1 : Common Values for the Exponents for Different Static 22
Loads
Table 5.1 : Load flow calculation of IEEE 15 bus radial system 27
Table 5.2 : Load Flow Calculation of Constant power Load Model 28
Table 5.3 : Load Flow Calculation of Constant current Load Model 28
Table 5.4 : Load Flow Calculation of Constant impedance Load Model 29
Table 5.5 : load flow calculation of IEEE 15 bus system with power 30
quality problems
Table 5.6 : Load flow calculation of IEEE 15 bus system using PSO 31
Table 5.7 : Load Flow Calculation of Constant Power Load Model 31
using PSO

Table 5.8 : Load Flow Calculation of Constant current Load Model 32

using PSO

Table 5.9 : Load Flow Calculation of Constant impedance Load Model 33

using PSO

Table 5.10 : Load flow calculation of IEEE 15 bus system with power 34
quality problems using PSO
Table 5.11 : Tariff calculation of IEEE 15 bus system 34
Table 5.12 : Tariff calculation of IEEE 15 Bus system with power 35
quality problems
Table 5.13 : Tariff calculation of IEEE 15 bus system using PSO 36
Table 5.14 : Tariff calculation of IEEE 15 Bus system with power 36
quality problems using PSO
Table 5.15 : Load flow results comparison 37

DEPT. OF EEE Page viii

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

Figure Page
Title
No. No.
Table 5.16 : Constant power load model comparison 37
Table 5.17 : Constant current load model comparison 38
Table 5.18 : Constant impedance load model comparison 38
Table 5.19 : Power Quality problems results comparison 38

DEPT. OF EEE Page ix

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

LIST OF FIGURES
Figure Page
Title
No. No.
Fig. 2.1 : IEEE 15 – Bus Single line diagram 7
Fig. 2.2 : Flowchart of BSF Method based on branch current 10
Fig. 3.1 : Impulsive transient 14
Fig. 3.2 : Oscillatory transient 15
Fig. 3.3 : Voltage sag 15
Fig. 3.4 : Voltage swells 15
Fig. 3.5 : Interruption 16
Fig. 3.6 : Classification of DG 19
Fig. 4.1 : Birds food searching with PSO 23
Fig. 4.2 : Concept of a searching point by PSO 24
Fig. 4.3 : Flowchart of PSO algorithm 24
Fig. 4.4 : PSO Procedure flowchart 26

DEPT. OF EEE Page x

ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

LIST OF ABBREVIATIONS
Abbreviation Description
BFS Backward Forward sweep
PQ Power quality
LDI Long duration interruptions
IEEE Institute of Electrical and Electronics Engineers
IEC International Electro technical commission
KCL Kirchhoff's current law
KVL Kirchhoff's voltage law
MVA Mega volte amps
SDG Synchronous Distributed Generators
DG Distributed Generators
PV Photo voltaic
ES Energy storage
SCADA Supervisory control and data acquisition
DMS Distribution management system
AMI Advanced metering infrastructure
GA Genetic Algorithm
PSO Particle Swarm Optimization
ANN Artificial Neural Network
IEEE Institute of Electrical and Electronics Engineers
IET The Institution of Engineering and Technology
MCS Monte Carlo simulations
MLMC Multilevel Monte Carlo simulation
MOPSO Multi-objective particle swarm optimization
EGA Enhanced genetic algorithm
VCA voltage control algorithm
PLF Probabilistic load flow
VRT Variance reduction techniques

DEPT. OF EEE Page xi

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

CHAPTER 1

INTRODUCTION

1.1 INTRODUCTION:
Nowadays, rapid growing of technology like SCADA, DMS, AMI etc., are
reduces the complexity at distribution level for analysis but still power system is
facing new issues such as sensitive loads, environmental effects and economic
concerns which require special consideration. The main issue facing at end user level
is tariff variation, at the time of billing the load should vary with respect to
consumption, but it is not happens in few conditions. Due to internal disturbances the
consumption and energy units are varies regularly which may affect the tariff. They
are many reasons behind the variations.

 Conductor/Cables Failures
 Switches failures
 Power Quality Problems etc.,
As said the above some reasons are affected the tariff, so that many of external
failures are easily identifies rather than the internal, such that the power quality
problems is the one of major problem which varies the tariff at consumer level.

It requires the sufficient method to analyze power quality problems under No-
load and loading conditions. So far many optimized methods are analyzed the quality
problems at end user, like

 Genetic Algorithm(GA)
 Particle Swarm Optimization(PSO)
 Fuzzy logic
 Artificial Neural Network (ANN) etc.
This proposed work is analysis of power quality problems in distribution
network with respect to cost variations. By the proposed work minimizing power loss
and improve voltage profile using distribution generation units.

DEPT. OF EEE Page 1

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

1.2 LITERATURE SURVEY:

The major problem of distribution system is mitigation of power quality
problems using DGs. So, many authors can do research on placement and sizing of
DGs to reduce power loss and improve voltage profile. No one can concentration on
cost calculation.

 Ahmadreza Eslami, Mohammad Esmail Hamedani Golshan [1] - In this

the author describes the placement of synchronous distributed generators
(SDGs) to mitigate the voltage dips and long duration interruption. The main
objective of this paper is to reduce the cost of the planning SDGs compare to
network upgrade cost and power lose cost.
 Priyanka Gangwar, Sri Niwas Singh, Saikat Chakrabarti[2] -In this paper
authors describe the selection of DG and capacitor & sitting of DG and
capacitor to reduce the power quality problems in multi-phase distribution
system are balanced and unbalanced conditions. The multi-objectives in this
paper is stability margin improvement, energy loss minimization, power loss
minimization voltage deviation and etc.
 Huayi Wn, Ping Dong [3] -This paper is focus on handling the distribution
network reconfiguration. Authors proposed PPSO method to provide a new
way to solve network reconfiguration problems like power loss reduction and
load demand using DGs. The wind and PV type DG are used.
 Bashir Bakhshideh Zad, Jacques Lobry, François Vallee [4] - In this paper
author is proposed voltage regulation problem of the medium-voltage
distribution systems. Voltage regulation problem is occurring due to the
uncertain models of the system components.
 Alberto Escalera, Barry Hayes, Milan Prodanovic [5] - In this paper author
is proposed Analytical method to improve the reliability of supply. Author is
proposed how the impact of DG and energy storage system on distribution
system reliability.
 Mehrdad Tarafdar Hagh, Payman Amiyan, Sadjad Galvani, Naser
Valizadeh [6] - In this paper author is proposed load flow calculation in the
presence of wind generations by using particle swarm optimization (PSO)

DEPT. OF EEE Page 2

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
algorithm. The main objective of this is providing simple and fast method for
calculating load flows.
 Yang Fu, Huanhuan Liu, Xiangjing Su, Yang Mi, ShuxinTian [7] - In this
paper author is proposed to placing roof top photovoltaics in distribution
systems. Probabilistic load flow (PLF) is introduced to overcome the
uncertainties in power systems.
 Andres Arguello, Jose D. Lara , Jose David Rojas, Gustavo Valverde [8]
- In this paper author is proposed the impact of rooftop photovoltaic systems
in large-scale distribution networks. The proposed method captures the effects
of customers with high probability of installing photovoltaic systems.
 S. Nazmul Huda, Rastko Zivanovic [9] - In this paper author is proposed the
reliability of distribution system including weather conditions. The main
objective of this paper is computation speed is increased and maintains
accuracy level constant.
 Mike Brian Ndawula, Ignacio Hernando-Gil [10] - In this paper author is
proposed the reliability assessment of a urban distribution network and Two
smart grid scenarios considering controllability of PV and energy storage (ES)
are designed to improve network performance.
From the above survey there is some existing methods as shown below
 Genetic Algorithm (GA)
 Particle Swarm Optimization (PSO)
 Fuzzy logic
 Artificial Neural Network (ANN)
 Water Cycle Algorithm (WCA)
 Whale Optimization Algorithm (WOA) etc.

1.3 PROBLEM IDENTIFICATION:

From the literature few problems are identifies as follows

 The distribution load flows in real time shouldn’t be analyzed for cost at
consumer end.
 The power quality problems should be estimated under load uncertainties with
respect to economic aspects.

DEPT. OF EEE Page 3

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
 The mitigation of power quality problems using DGs, capacitor banks etc., are
not analyzed their benefits towards tariff.
 Most of the optimized techniques shouldn’t opted to analyze the performance
of distribution system by considering the cost under load uncertainties.

1.4 OBJECT OF THE WORK:

 Objective.1: Estimating the load flows and cost at each load using distribution
load flows.
 Objective.2: Identify and mitigate the power quality issues in the distribution
system under economic aspects.
 Objective.3: Optimized solution towards analysis of power quality issues
under economic aspects.

DEPT. OF EEE Page 4

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

CHAPTER 2

MATHEMATICAL ANALYSIS OF RADIAL

DISTRIBUTION SYSTEM

2.1 INTRODUCTION:

The planning of optimal distribution system is important for the growth and
effective usage of distribution system network. At distribution side load demand
continuously increase and future expansion of the distribution network depends on the
load flow study. Some of the feature of the distribution system at ill condition is
shown below
 High R/X ratios
 Unbalanced and Multi phase operation
 Radial and weakly meshed networks
 Load unbalance
There are so many methods to calculation of load flows. In radial distribution
system load flow calculation is easy by using backward forward sweep method.
Kirchhoff’s laws are used to calculate load flows in backward forward sweep method.

2.2 LOAD FLOW METHODS FOR CALCULATION OF

DISTRIBUTION SYSTEM:
There are several methods to calculate load flows in distribution system.
1. Primitive impedance based Distribution load flow method
2. Current injection based load flow method
3. Ladder network based load flow method
4. Backward-forward sweep method
• Based on branch current
• Based on branch power
• Based on branch impedance

DEPT. OF EEE Page 5

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

2.3 BACKWARD - FORWARD SWEEP METHOD:

The backward – forward sweep method is an iterative method for load flow
calculation and at each iteration two stages are performed. The first stage of
calculation of branch currents through the branches starting from at end branch to root
branch in backwarddirection.The second stage of calculation bus voltage magnitude
and phase angle of each node starting from root node to end node in forward
direction.

The backward – forward sweep method is derivative free approach and it have
the simple circuit laws.

Backward sweep method:

The main aim of backward sweep method is calculate branch currents. The
section currents are calculated from the last node to the source bus by the application
of Kirchhoff's current law (KCL). The branch currents are updated in each section by
considering the previous iteration voltages at each node.

Forward sweep method:

The main aim of forward sweep method is calculate voltage drop and update
the power flow. In this section bus voltages are calculated from source node to the end
node by the application of the Kirchhoff's voltage law (KVL). During the forward
sweep power in each branch is constant. The previous iteration voltage magnitude
values are compared at each bus in present iteration.

2.3.1 Algorithm for BFS Method:

Step 1: Read distribution system line data, load data, base MVA and base KVA, no.
of buses, no. of lines.
Step 2: Assume initial node voltage 1 pu
Step 3: Set iteration count 1 and also set the no. of iterations
Step 4: Calculate load currents except slack bus
Step 5: Calculate branch currents in backward process by adding branch currents
from end node to source node (apply KCL)
Step 6: Calculate node voltage magnitudes and phase angles in forward process
(apply KVL)

DEPT. OF EEE Page 6

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
Step 7: Calculate power flows
Step 8: Calculate real power loss and reactive power loss
Step 9: Increment the iteration count by 1
Step 10: If (count > max iteration) go to step 4 otherwise go to next step.
Step 11: Stop

2.3.2 Test System:

Figure 2.1: IEEE 15 – Bus Single line diagram

Fig.2.1 shows the single line diagram of 15-bus distribution system line diagram of
radial distribution system with sending end nodes, receiving end nodes, and branch as
impedance. Each node has a load as apparent power, we can take real and reactive
power by calculating the load given in Table 2.1. And also find voltage, current,
power losses, power flow can be calculated in load flows.

Table 2.1: Test system data

S.No. Fb Tb R(Ω) X(Ω) KVA
1 1 2 1.35309 1.32349 63
2 2 3 1.17024 1.14464 100
3 3 4 0.84111 0.82271 200
4 4 5 1.52348 1.02760 63

DEPT. OF EEE Page 7

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
5 2 6 2.55727 1.72490 200
6 6 7 1.08820 0.73400 200
7 6 8 1.25143 0.84410 100
8 2 9 2.01317 1.35790 100
9 9 10 1.68671 1.13770 63
10 3 11 1.79553 1.21110 200
11 11 12 2.44845 1.65150 100
12 12 13 2.01317 1.35790 63
13 4 14 2.23081 1.50470 100
14 4 15 1.19702 0.80740 200

2.3.3 Formulas used for BFS Method:

The following formulas are used for power loss calculations in backward and
forward sweep method

1. Calculation of load currents at each bus:

( ) ( ) (2.1)

for j = 1,2,3……n
Where,
IL(j) = Load current
Sj= Apparent power
Vj = Bus voltage
Pj = Active power
Qj = Reactive power
n = no. of buses
2. Calculation of branch currents:
Ibr(i,j) = IL(i) + IL(j) (2.2)

for i = Sending end bus(1,2,3……n-1)

for j = Receiving end bus (1,2,3……n)

Where,
Ibr(i,j) = Branch current

DEPT. OF EEE Page 8

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
3. Voltage drops in each branch:
∆V(i,j) = Z(i,j)*Ibr (2.3)
for i = 1,2,3…..n-1
for j = 1,2,3……n
Where,
∆V(i,j) = Change in voltage
Z(i,j) = Branch impedance
4. Voltages at receiving bus:
V(j) = V(j) - ∆V(i,j) (2.4)
for j = 1,2,3……n
5. Calculate power flow :
S(i,j) = Vj*Ibr(i,j) (2.5)

for i = Sending end bus(1,2,3……n-1)

for j = Receiving end bus (1,2,3……n)

6. Real and Reactive power loss:

Ploss(i,j) = Ibr(i,j)2*R(i,j) (2.6)
Qloss(i,j) = Ibr(i,j)2*X(i,j) (2.7)
for i = 1,2,3……n-1
for j = 1,2,3……n
Where,
R(i,j) = Resistance
X(i,j) = Reactance
7. Total power loss:
Ptl = ∑j=1…nPloss(i,j) (2.8)
Qtl = ∑j=1…nQloss(i,j) (2.9)

Where,
Ptl = Total active power loss
Qtl = Total reactive power loss

DEPT. OF EEE Page 9

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

2.3.4 Flowchart for BFS Method:

Start

Read line data and load data

Set flat voltage (1 pu) at all the nodes

Compute effective branch currents of all

Branches using backward Propagation

Update node voltage and phase angles using forward

Propagation

No

Is load flow is
converged?

Yes

Compute branch power loss, total power loss and print

the Results

End

Fig. 2.2: Flowchart of BSF Method based on branch current

DEPT. OF EEE Page 10

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

2.4 TARIFF CALCULATION:

 1 unit = 1 kwh

 Energy in kilowatt hours = Power in watts x time in hours

 One day energy in kwh = Energy in kilowatt hours x 24 hours

 One month energy in kwh = One day energy in kwh x 30 days

 Monthly cost = One month energy in kwh x unit price + Basic charge

2.5 TEST SYSTEM LOAD FLOW CALCULATION:

Backward/Forward Sweep method is tested on IEEE 15-bus radial distribution system

using MATLAB 2014b. IEEE 15 bus radial Distribution system has 15 nodes and 14
branches. The base voltage of the system is 11 KV and the base MVA is 100. Here
load power factor is taken 0.8 then calculate the active power and reactive power at
each load by using table 2.1. The test system load flow calculation of each and every
branch is shown in below table 2.2.
Table 2.2: Load flow calculation of IEEE 15 Bus system

Br Se Re V (p.u) PL (kW) QL(kVAR) Pflow (kW) Qflow (kVAR)

1 1 2 0.9715 37.687 36.862 1417.7 1082.2

2 2 3 0.9570 11.282 11.035 819.7 629.7

3 3 4 0.9513 2.441 2.388 446.7 344.1

4 4 5 0.9503 0.055 0.037 49.9 38.4

5 2 6 0.9582 5.768 3.890 398.7 302.8

6 6 7 0.9559 0.393 0.265 159.3 120.9

7 6 8 0.9569 0.112 0.076 79.6 60.5

8 2 9 0.9681 0.472 0.318 129.8 98.8

9 9 10 0.9670 0.059 0.039 50.1 38.2

10 3 11 0.9502 2.175 1.467 288.7 221.3

DEPT. OF EEE Page 11

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

11 11 12 0.9460 0.601 0.405 129.5 99.2

12 12 13 0.9446 0.074 0.049 50 38.3

13 4 14 0.9490 0.204 0.138 79.2 61

14 4 15 0.9488 0.439 0.296 158.5 122

Table 2.3: Tariff calculation of IEEE 15 bus system

Bus no Power (kw) at Power in one month Cost in Rupees

1 hour (kwh) (1 unit = Rs 5)
1 1783.5 1284100 6420600

2 1033.7 744200 3721200

3 563.8 406000 2029800

4 63 45400 226800

5 500.6 360400 1802200

6 200 14400 720000

7 100 72000 360000

8 163.1 117400 587100

9 63 45400 226800

10 363.8 261900 1309700

11 163.1 117400 587100

12 63 45400 226800

13 100 72000 360000

14 200 144000 720000

The active power and reactive power loss of the test system is 61.76 kw (4.35%) and
57.27 kvar (5.29%). The low voltage bus of test system is 13 and the voltage value is
0.9446. Table 2.3 shows the tariff calculation at each load.

DEPT. OF EEE Page 12

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

CHAPTER 3

POWER QUALITY PROBLEMS AND

DISTRIBUTION GENERATION

3.1 POWER QUALITY:

3.1.1 Definition:

There are different definitions for power quality

 According to Utility, power quality is reliability.
 According to load aspect, it is defined as the power supplied for satisfactory
performance of all equipment i.e., all sensitive equipment.
 IEEE: Power quality is defined as, “the concept of powering and grounding
Sensitive equipment in a matter that is suitable to the operation of that
equipment”.
 IEC: Power quality defined as, “Characteristics of the Electricity at a given
point on the Electrical System, evaluated against a set of reference technical
parameters”.

3.1.2 Types of power quality problems:

Different power quality problems and its characteristics shown in below table,

Table 3.1: Classification of power quality problems as per IEEE standards

Categories Typical characteristic’s

Over voltage transients
Impulsive : Nano second : 5 ns rice time for less than 50 ns
Micro second : 1 ns rise time for 50 ns – 1 ms
Microsecond : 0.1 ms rise time for > 1 ms
Oscillatory : Low. freq. : >5kHz for 0.3-50ms at 0-4 pu
Med. Freq. : 5-500kHz for 20µs at 0-8 pu
High freq. : 0.5-5MHz for 5µs at 0-4 pu
Short duration voltage
variations
Interruption : Momentary : <0.1 pu for 0.5 cycles – 3 s
Temporary : <0.1 pu for 3 s – 1 min
Sag : Instantaneous : 0.1 – 0.9 pu for 0.5 – 30 cycles
Momentary : 0.1 – 0.9 pu for 30 cycles – 3 s

DEPT. OF EEE Page 13

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
Temporary : 0.1 – 0.9 pu for 3 s – 1 min

Swell : Instantaneous : 1.1 – 1.8 pu for 0.5 – 30 cycles

Momentary : 1.1 – 0.4 pu for 30 cycles – 3 s
Temporary : 1.1 – 1.2 pu for 3 s – 1 min
Long duration voltage
variations
Interruption : Sustained : 0.0 pu for > 1 min
Under voltages : 0.8 – 0.9 pu for > 1 min
Over voltages : 1.1 – 1.2 pu for > 1 min
Voltage waveform distortions
DC offset : 0 - 0.1%
Harmonics : 0 – 100th H with 0-20% magnitude
Inter harmonics : 0 – 6 kHz with 0 – 2% magnitude
Notching :
Voltage fluctuations
Intermittent : < 25 Hz with 0.1 – 7% magnitude

I. Transients:

It is an event that is unwanted and momentary in nature. It is the sudden

change in one steady state operating condition to another.
Transients can be classified into two categories:
a. Impulsive Transient: An impulsive transients a sudden non–power frequency
change in the steady-state condition of voltage, current, or both that is unidirectional
in polarity (either positive or Negative).
Source: Lightning

Fig 3.1: Impulsive transient

b. Oscillatory Transient: An oscillatory transient is a sudden, non–power frequency
change in the steady-state condition of voltage, current, or both, that includes both
positive and negative polarity values.

DEPT. OF EEE Page 14

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
Sources: Back-to-back capacitor switching, Transformer energization.

Fig 3.2: Oscillatory transient

II. Short - Duration Voltage Variations:
When the rms value of voltage deviates for duration less than 1 minute, it is
termed as short duration voltage variation.
It may be categorized into following types:
a. Sags (dips): A sag is decrease in rms voltage or current between 0.1 and 0.9 pu at
the power frequency for durations from 0.5 cycle to 1 min.
Sources: System faults, starting of large motors.

Fig 3.3: Voltage sag

b. Swells: A swell is defined as an increase to between 1.1 and 1.8 pu in rms voltage
or current at the power frequency for durations from 0.5 cycle to 1 min.
Sources: Energizing large capacitor bank.

Fig 3.4: Voltage swells

DEPT. OF EEE Page 15

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
c. Interruption: An interruption occurs when the supply voltage or load current
decreases to less than 0.1 pu for a period of time not exceeding 1 min.
Sources: Faults, equipment failures, and control malfunctions

Fig 3.5: Interruption

III. Long - Duration Voltage Variations:

When the rms value of voltage deviates for duration more than 1 minute, it is
termed as long duration voltage variation.
Sources: Load variations, System switching operation.
It may be categorized into following types
a. Over Voltage: An overvoltage is an increase in the rms ac voltage greater than 110
percent at the power frequency for duration longer than 1 min.
Sources: Load switching, incorrect tap settings on transformers
b. Under Voltage: An under voltages a decrease in the rms ac voltage to less than 90
percent at the power frequency for a duration longer than 1 min.
Sources: A load switching on or a capacitor bank switching off.
c. Sustained Interruptions: When the supply voltage becomes zero for a period of
time in excess of 1 min, the long-duration voltage variation is considered a sustained
interruption.
IV. Voltage Imbalance:
Voltage imbalance (also called voltage unbalance) is defined as the maximum
deviation from the average of the three-phase voltages or currents, divided by the
average of the three-phase voltages or currents, expressed in percent.
V. Waveform Distortion:
Waveform distortion is defined as a steady-state deviation from an ideal sine
wave of power frequency.

DEPT. OF EEE Page 16

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
There are five primary types of waveform distortion:
a. DC offset: The presence of a DC voltage or current in an AC power system is
termed dc offset.
b. Harmonics: Harmonics are sinusoidal voltages or currents having frequencies that
are integer multiples of the supply frequency (fundamental frequency).
Sources: Non-linear loads
c. Inter Harmonics: Voltages or currents having frequency components that are not
integer multiples of the frequency are called inter Harmonics.
Sources: Static frequency converter, cycloconverters, induction furnaces, and arcing
devices. Power line carrier signals can also be considered as inter Harmonics.
d. Notching: Notching is a periodic voltage disturbance caused by the normal
operation of power electronic Devices when current is commutated from one phase to
another.
e. Noise: Noise is the unwanted electrical signals with broadband spectral content
lower than 200 kHz.
Sources: Power electronic devices, control circuits, arcing equipment.
VI. Voltage Fluctuation:
Voltage fluctuations are systematic variations of the voltage envelope or a
series of random Voltage changes, the magnitude of which does not normally exceed
the voltage range 0.9 to 1.1 pu.
In order to analyze how an LDI affects the equipment, it is important to know
how long a customer is going to be without electricity after the fault. This duration
consists of fault locating time, repair time, etc. It is referred to as the cumulative
probability curve of power supply interruption duration.

3.2 DISTRIBUTION GENERATION:

3.2.1 Definition:

The IEEE defines distributed generation as the generation of electricity by

facilities that are sufficiently smaller than central generating plants so as to allow
interconnection at nearly any point in a power system. Electrical power systems are
developing towards more decentralized operation from centralized bulk power
systems. These decentralized systems consist of distributed generators (DGs), which

DEPT. OF EEE Page 17

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
offer many economic benefits, such as a reduction in fuel cost, power loss, and CO2
emission; technical benefits such improvement in power quality, voltage profile, and
reliability.

3.2.2 Types of DG’s:

The types of distributed generations are mainly divided in to usage of fuel.

The fuel used in DGs classified in to two types, they are renewable energy DG and
non-renewable energy DG. The examples of renewable energy DG fuel usage are
wind, solar and biomass etc. The examples of renewable energy DG fuel diesel
generator, micro turbine, gas turbine etc.

Type I DG: This DGs provides constant real power model. The examples of these
DGs are solar and fuel cells.

Type II DG: These DGs supporting both real and reactive power to the distribution
system. The examples are gas turbines, micro turbine and biomass plants. These DGs
provide constant power factor.

Type III DG: These DGs are providing real power to the system and absorb the
reactive power to the system. The Examples are double fed induction generator and
wind farm.

Type IV DG: These DGs are providing only reactive power to the network. The
examples are synchronous compensators and synchronous generators.

Table 3.2: Types of DG and their models

Type Model of DG Example Real power of Reactive power of DG

of DG DG
Type 1 Real Power Solar Photo PDGi 0
Constant voltaic cells
Type 2 Power Factor Gas turbine PDGi = SDGi cosϕ QDGi = SDGi sinϕ
Constant
Type 3 Reactive Power Wind farm PDGi QDGi = -(0.5+0.04(PDGi)2)
Variable
Type 4 Reactive Power Synchronous 0 QDGi
Constant compensator
s

DEPT. OF EEE Page 18

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
Based on the sources used for generation at or near the load centers, they are
classified as renewable and non-renewable and the classification is clearly shown with
the help of below Figure

Fig. 3.6: Classification of DG

3.2.3 Various distributed generation capacities:

Different types of Distribution system capacities shown below

 Micro DGs – 1 W to < 5 MW

 Small DGs – 5 kW to < 5 MW
 Medium DGs – 5 MW to < 50 MW
 Large DGs – 50 MW to < 300MW

3.2.4 Advantages of using DGs:

The advantages of using DG units in radial distribution system are shown below:

a. Technique al benefits:
 System losses are decreases
 Voltage profile improved
 In transformer thermal stresses are reduced
 DG unit take less time to construct and develop
 In peak hour time DG satisfy the customer demand

DEPT. OF EEE Page 19

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
b. Economical benefits:
 Cost of distribution system is reduce
 The fuel price of distribution network by placing of DG is less and rick of
service less
c. Environmental benefits:
 Compare to other power plants like coal, oil, or gas the emission of green
house gases is reduced.

3.3 DG Rating Constraints:

The DG unit ratings are must in between minimum and maximum levels. The
limitations are given as follows,

Where,
= Capacity of DG
= DG Maximum capacity
= DG Minimum capacity

3.4 Real Power and Bus Voltage Constraints:

The active power flow maximum limits of distribution feeder show in below,

where,

= Active power flow at feeder

= Active power flow maximum at feeder

The each bus voltage magnitude ranges are given below,

This voltage boundary of each bus between 0.95 pu to 1.05 pu

Where,
= b bus voltage
= b bus minimum voltage
= b bus maximum voltage

DEPT. OF EEE Page 20

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

CHAPTER 4

DISTRIBUTION SYSTEM LOAD MODELS

AND PROPOSED ALGORITHM

4.1 Load Modeling in Distribution System:

Distribution load flow studies have been intensely impacted by Load

modelling. Load flow results and meeting ability are greatly affected by load
characteristics.
Load models are mainly classified into two types i.e., static load models and dynamic
load models.
1. Static load model: Static load model shows the frequency and voltage magnitude
at any moment of time as the function of active and reactive power at same instant.
Static load components like resistive and light loads are static load models and
dynamic nature of load components like motor driven loads are dynamic load models.
2. Dynamic load model: Dynamic load model expresses the active and reactive
powers at any instant of time as functions of the bus voltage magnitude and frequency
at past instants of time and, usually, including the present instant. These types of
models are represented by using differential equations.
The static load models are considered to calculate load flows in distribution
system. Here mainly concentrated loads are constant power, constant current,
constant impedance. The calculation of different load models by using equations 4.1
and 4.2. The equation 4.1 calculates active power and equation 4.2 calculate reactive
power at loads.

( ) (4.1)

( ) (4.2)

Where
np & nq = load exponents
V = load bus voltage
V0 = load nominal voltage
= Active and powers at the nominal voltages

DEPT. OF EEE Page 21

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
= Reactive and powers at the nominal voltages
The different loads can be represented different load exponents. The load
exponents of constant power, current & impedance models by placing np & nq values
are 0, 1, or 2 respectively. The different types of load components can be represented
by using other exponents. The table 4.1 represents different static loads and its
exponents values.
Table 4.1: Common Values for the Exponents for Different Static Loads
S.No Load components Np Nq
1 Constant Power 0.00 0.00
2 Constant current 1.00 1.00
3 Constant impedance 2.00 2.00
4 Incandescent light 1.55 0.00
5 Fluorescent light 0.96 7.38
6 Air conditioner 0.20 2.30
7 Dryer 2.04 3.27
8 Freezer 0.77 2.50
9 Heater 2.00 0.00
10 Pumps, fans etc 0.08 1.60
11 Computer, T.V. etc 2.00 5.20

4.2 Particle Swarm Optimization:

Particle swarm optimization is an optimization technique. This is inspired by
the nature. Based on Swarms intelligence and movement this optimization technique
operates. It was developed in 1995 by James Kennedy (social psychologist) and
Russell Eberhart (electrical engineer). The group of particles to searching food is the
system model. The group of creatures has this relative behavior, for example, bee
swarm, fish school and bird flock.
Figure 4.1 shows bird flock hunter that is a bird suspect to a particle. In victim
searching, the bird group flying one direction and leader is always near to the food
that is the shortest distance and all other birds are follow leader this is the best fitness.
The food searching of birds have some velocity and direction. Particles velocity and
position is helps to best fitness.

DEPT. OF EEE Page 22

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

Fig. 4.1: Birds food searching with PSO

PSO consists of a group (swarm) of individuals (particles) moving in the
search space looking for the best solution. The particle flying experience is comes by
updating the current position using velocity vector. Select particle with the least good
value is Pbest of all particles, and choose the value is Pbest and the current overall
good value is Gbest with a random weighted acceleration at each time step as shown
in Fig. 4.2. Each particle goes to change the position using the below information with
the PSO algorithm flowchart as shown in Fig. 4.3:
 Update positions
 Update velocities
 The distance between the current position and the pbest
 The distance between the current position and the gbest
The Particle’s velocity and position by using equations (4.3) and (4.4)

( ) ( ) [4.3]
[4.4]
where,
c1 & c2 = Weighting factors,
r1 & r2 = Random numbers between 0 and 1,
w = Weighting function,
= Current velocity of particle i at iteration k,
= Modified velocity of particle i,
= Current position of particle i at iteration k,
= Modified position of particle i,
Pbest = Personal best of particle i,
Gbest = Global best of the group.

DEPT. OF EEE Page 23

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

Figure 4.2: Concept of a searching point by PSO

start

Swarm initialization

Compute the individual performance

Locate best individual and collective positions

Calculate the velocity for each particle

Compute the position for each particle

No
Criterion of
termination

Yes

End

Figure 4.3: Flowchart of PSO algorithm

The benefits of PSO algorithm compare to other algorithms as follows:
 The operation of this algorithm is not complex and here using basic logic
operations and simple mathematical formulas.

DEPT. OF EEE Page 24

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
 This algorithm generates the objective function random.
 To start iteration process it does not need best initial solution.
However, the Disadvantages of PSO algorithm as follow:
 Adjust the more number of parameters is required.
 To improve this algorithm required programming skills are need.

4.2.1 Simulation procedure of PSO:

The Simulation procedure of PSO algorithm for sizing and placement of DG
to minimizing the power loss and improve voltage profile at each bus as shown in
following steps:
Step 1: Read distribution system line data, load data, base MVA and base KVA, no.
of buses, no. of lines and voltage limits.
Step 2: By using backward forward sweep method to calculate load flows of radial
distribution system.
Step 3: Generate initial population, positions and velocities of each particles
randomly. Choose iteration count k = 0.
Step 4: Calculate each particle bus voltages are in limits and also calculate the total
power loss.
Step 5: Each particle balances its individual best objective value. If the objective
value is less than the Pbest, set this value as current Pbest, and corresponding particle
position is record.
Step 6: Select particle with the least good value is Pbest of all particles, and choose
the value is Pbest and the current overall good value is Gbest.
Step 7: The position and velocity of each particle is updated.
Step 8: If the iteration goes to the maximum limit, go to Step 9. Else, set iteration
count k = k+1, and go to Step 4.
Step 9: Print the results to the objective problem. The locations and sizing of
distribution generation best positions are included and the fitness value is showing the
minimum power loss.
The objective function of PSO algorithm is power loss reduction. The
optimization problem is sets the PSO algorithm. The PSO algorithm is give good
solutions by finite steps. This is programmed by using MATLAB software. The
following figure shows the flowchart of PSO procedure.

DEPT. OF EEE Page 25

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

Start

Input system data

Calculate the original loss using Bw-Fw sweep

Initialize particle population

Calculate the total loss

Record Pbest ,Gbest

Update particle position and velocity

No Check the
stopping
criterion
Yes

Particle Swarm Print out location and size of DG

Optimization

End
Figure 4.4: PSO Procedure flowchart

DEPT. OF EEE Page 26

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

CHAPTER 5

SIMULATION RESULTS
The proposed methodology was tested on the 11kv and 100MVA level of the
IEEE 15 bus radial distribution system. The 15 bus system has 14 sections with the
total load 1401.6 Kw and 1051.2 Kvar. The total real and reactive power loss of the
system is 61.76 Kwand 57.27 Kvar. For the PSO parameters, population size = 20,
Maximum iterations = 50.

5.1 LOAD FLOW CALCULATION WITHOUT DG:

The load flow calculation of IEEE 15 bus radial distribution system as shown
in table 5.1. The table represents branch number, sending end and receiving end bus
numbers, voltage in per unit, active and reactive power loss, active and reactive power
flows. The total active and reactive power loss of 15 bus system is 61.76 Kw and
57.27 Kvar. The weak bus of system is 13th bus and its voltage value 0.9446.
Table 5.1: Load flow calculation of IEEE 15 bus radial system

Br Se Re V (p.u) LP (kW) LQ (kVAR) Pflow (kW) Qflow (kVAR)

1 1 2 0.9715 37.687 36.862 1417.7 1082.2

2 2 3 0.9570 11.282 11.035 819.7 629.7
3 3 4 0.9513 2.441 2.388 446.7 344.1
4 4 5 0.9503 0.055 0.037 49.9 38.4
5 2 6 0.9582 5.768 3.890 398.7 302.8
6 6 7 0.9559 0.393 0.265 159.3 120.9
7 6 8 0.9569 0.112 0.076 79.6 60.5
8 2 9 0.9681 0.472 0.318 129.8 98.8
9 9 10 0.9670 0.059 0.039 50.1 38.2
10 3 11 0.9502 2.175 1.467 288.7 221.3
11 11 12 0.9460 0.601 0.405 129.5 99.2
12 12 13 0.9446 0.074 0.049 50 38.3
13 4 14 0.9490 0.204 0.138 79.2 61
14 4 15 0.9488 0.439 0.296 158.5 122

DEPT. OF EEE Page 27

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
The load flow calculation of constant power load model is shown in table 5.2.
The total active and reactive power loss of 15 bus system is 61.76 Kw and 57.27
Kvar. The lower voltage bus is 13th bus and its voltage is 0.9446.
Table 5.2: Load Flow Calculation of Constant power Load Model

Br Se Re V (p.u) LP (kW) LQ (kVAR) Pflow (kW) Qflow (kVAR)

1 1 2 0.9715 37.687 36.862 1417.7 1082.2

2 2 3 0.9570 11.282 11.035 819.7 629.7
3 3 4 0.9513 2.441 2.388 446.7 344.1
4 4 5 0.9503 0.055 0.037 49.9 38.4
5 2 6 0.9582 5.768 3.890 398.7 302.8
6 6 7 0.9559 0.393 0.265 159.3 120.9
7 6 8 0.9569 0.112 0.076 79.6 60.5
8 2 9 0.9681 0.472 0.318 129.8 98.8
9 9 10 0.9670 0.059 0.039 50.1 38.2
10 3 11 0.9502 2.175 1.467 288.7 221.3
11 11 12 0.9460 0.601 0.405 129.5 99.2
12 12 13 0.9446 0.074 0.049 50 38.3
13 4 14 0.9490 0.204 0.138 79.2 61
14 4 15 0.9488 0.439 0.296 158.5 122

The load flow calculation of constant current load model is shown in table 5.3.
The total active and reactive power loss of IEEE 15 bus radial distribution system is
55.87 Kw and 51.80 Kvar. The lower voltage bus is 13th bus and its voltage is 0.9475.
By comparing constant power load model the power losses are decreased and bus
voltages are improved.
Table 5.3: Load Flow Calculation of Constant current Load Model
BR SE RE V (p.u) LP (Kw) LQ (Kvar) Pflow (Kw) Qflow (Kvar)
1 1 2 0.9729 34.1647 33.4173 1352.1 1031.2
2 2 3 0.9592 10.1273 9.9058 778.7 597.5
3 3 4 0.9537 2.1918 2.1438 424.5 326.5
4 4 5 0.9528 0.0497 0.0335 47.5 36.5

DEPT. OF EEE Page 28

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

5 2 6 0.9602 5.2614 3.5489 381.6 289.6

6 6 7 0.95880 0.3582 0.2416 152.3 115.6
7 6 8 0.9589 0.1030 0.0695 76.2 57.8
8 2 9 0.9696 0.4407 0.2972 125.6 95.5
9 9 10 0.9686 0.0552 0.0372 48.5 36.9
10 3 11 0.9527 1.9446 1.3116 273.8 209.7
11 11 12 0.9487 0.5345 0.3605 122.5 93.7
12 12 13 0.9475 0.0656 0.0443 47.3 36.2
13 4 14 0.9515 0.1834 0.1237 75.2 57.8
14 4 15 0.9514 0.3936 0.2655 150.4 115.7

The load flow calculation of constant impedance load model is shown in table
5.4. The total active and reactive power loss of IEEE 15 bus radial distribution system
is 50.86 Kw and 46.86 Kvar. The lower voltage bus is 13th bus and its voltage is
0.9501. By comparing above two models i.e. constant power and current load model
the power losses are decreased and bus voltages are improved.
Table 5.4: Load Flow Calculation of Constant impedance Load Model
BR SE RE V (P.U) LP (Kw) LQ (Kvar) Pflow (Kw) Qflow (Kvar)
1 1 2 0.9742 30.9830 30.3052 1289.7 982.8
2 2 3 0.9612 9.0934 8.8945 739.7 566.9
3 3 4 0.9560 1.9678 1.9248 403.4 308.8
4 4 5 0.9551 0.0447 0.0301 45.1 34.6
5 2 6 0.9620 4.8000 3.2376 365.3 227.0
6 6 7 0.9600 0.3260 0.2199 145.6 110.4
7 6 8 0.9609 0.0939 0.0634 73.0 55.3
8 2 9 0.9710 0.4115 0.2776 121.6 92.4
9 9 10 0.9700 0.0514 0.0347 46.9 35.6
10 3 11 0.9550 1.7387 1.1728 259.7 198.6
11 11 12 0.9513 0.4752 0.3205 115.8 88.5
12 12 13 0.9501 0.0582 0.0393 44.7 34.1

DEPT. OF EEE Page 29

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

13 4 14 0.9539 0.1643 0.1108 71.4 54.8

14 4 15 0.9538 0.3525 0.2378 142.8 109.6

The load flow calculation of IEEE 15 bus system with power quality problems
is shown in table 5.5. The total active and reactive power loss is 168520 Kw and
127910 Kvar. The lower voltage bus is 10th bus and its voltage is 0.001. By
comparing normal load flow the power losses are increased and bus voltages are
decreased. The voltage sag occurs at 6th bus and long voltage interruption occurs at
10th bus.
Table 5.5: load flow calculation of IEEE 15 bus system with power quality problems

Br Se Re V (p.u) LP (kW) LQ (kVAR) Pflow (kW) Qflow (kVAR)

1 1 2 0.9715 46737.24 45887 50207 37933

2 2 3 0.9570 11.103 10.860 823 625
3 3 4 0.9513 2.431 2.378 449 341
4 4 5 0.9503 0.055 0.037 50 38
5 2 6 0.8000 6.723 4.743 334 252
6 6 7 0.9559 0.393 0.265 160 120
7 6 8 0.9569 0.112 0.075 80 60
8 2 9 0.9681 66252 44688 48746 36826
9 9 10 0.0010 55327 37318 50 38
10 3 11 0.9502 2.162 1.458 290 220
11 11 12 0.9460 0.600 0.404 130 99
12 12 13 0.9446 0.073 0.049 50 38
13 4 14 0.9490 0.204 0.137 80 61
14 4 15 0.9488 0.438 0.296 159 121

5.2 LOAD FLOW CALCULATION WITH DG:

The load flow calculation with DG of IEEE 15 bus radial distribution system
as shown in table 5.6. The total active and reactive power loss of 15 bus system is
44.59 Kw and 48.03 Kvar. The weak bus of system is 13th bus and its voltage

DEPT. OF EEE Page 30

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
value0.9591. By comparing table 5.1 the power losses are reduced and voltage at each
bus is improved.
Table 5.6: Load flow calculation of IEEE 15 bus system using PSO

Br Se Re V (p.u) LP (kW) LQ (kVAR) Pflow (kW) Qflow (kVAR)

1 1 2 0.9791 27.21 26.61 1415.8 1035.1

2 2 3 0.9683 8.19 8.01 822.4 613.4
3 3 4 0.9641 1.79 1.75 449.2 340.5
4 4 5 0.9633 0.04 0.02 50.2 38.1
5 2 6 0.9694 4.06 2.73 399.5 257.7
6 6 7 0.9677 0.27 0.18 159.6 120.5
7 6 8 0.9685 0.08 0.05 79.9 47.1
8 2 9 0.9765 0.35 0.23 130.2 98.2
9 9 10 0.9757 0.04 0.02 50.3 37.9
10 3 11 0.9633 1.59 1.07 289.9 219.4
11 11 12 0.9601 0.44 0.29 130.0 98.4
12 12 13 0.9591 0.05 0.03 50.2 38.0
13 4 14 0.9623 0.15 0.10 79.7 60.4
14 4 15 0.9622 0.32 0.21 159.4 120.8

The load flow calculation of constant power load model with DG as shown in
table 5.7. The total active and reactive power loss of 15 bus system is 56.14 Kw and
57.33 Kvar. The lower voltage bus is 13th bus and its voltage is 0.9914. By comparing
table 5.2 the power losses are reduced and voltage at each bus is improved.
Table 5.7: Load Flow Calculation of Constant Power Load Model using PSO
BR SE RE V (p.u) LP (Kw) LQ (Kvar) Pflow (Kw) Qflow (Kvar)
1 1 2 0.9962 34.3231 33.5723 1422.3 1011.6
2 2 3 0.9947 10.1808 9.9581 837.6 583.1
3 3 4 0.9941 2.2033 2.1551 461.0 317.4
4 4 5 0.9907 0.0500 0.2800 51.4 35.3
5 2 6 0.9924 5.2832 3.5636 410.2 279.1

DEPT. OF EEE Page 31

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

6 6 7 0.9777 0.3597 4.3537 161.0 111.0

7 6 8 0.9853 0.1034 1.0030 81.0 56.1
8 2 9 0.9953 0.4420 0.2982 132.3 93.8
9 9 10 0.9950 0.0553 0.0373 51.2 36.2
10 3 11 0.9929 1.9553 1.3189 297.2 203.9
11 11 12 0.9918 0.5376 0.3626 133.7 90.9
12 12 13 0.9914 0.0660 0.0445 51.7 35.1
13 4 14 0.9935 0.1844 0.1244 81.9 56.2
14 4 15 0.9935 0.3957 0.2669 163.9 112.4

The load flow calculation of constant current load model with DG as shown in
table 5.8. The total active and reactive power loss of IEEE 15 bus radial distribution
system is 51.26 Kw and 52.47 Kvar. The lower voltage bus is 13th bus and its voltage
is 0.9918. By comparing constant power load model the power losses are decreased
and bus voltages are improved.
Table 5.8: Load Flow Calculation of Constant current Load Model using PSO
BR SE RE V (p.u) LP (Kw) LQ (Kvar) Pflow (Kw) Qflow (Kvar)
1 1 2 0.9964 31.4087 30.7216 1359.7 969.4
2 2 3 0.9949 9.1924 8.9913 795.0 555.6
3 3 4 0.9943 1.9899 1.9464 437.5 302.6
4 4 5 0.9911 0.0454 0.2545 49.0 33.8
5 2 6 0.9927 4.9114 3.3128 392.5 269.8
6 6 7 0.9784 0.3386 4.0979 156.1 107.9
7 6 8 0.9859 0.0961 0.9316 78.1 54.1
8 2 9 0.9954 0.4140 0.2793 128.0 90.9
9 9 10 0.9951 0.0517 0.0349 49.5 35.0
10 3 11 0.9931 1.7578 1.1857 281.4 194.0
11 11 12 0.9921 0.4807 0.3242 126.2 86.3
12 12 13 0.9918 0.0589 0.0398 48.8 33.2
13 4 14 0.9937 0.1661 0.1120 77.7 53.5

DEPT. OF EEE Page 32

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

14 4 15 0.9937 0.3563 0.2403 155.3 107.0

The load flow calculation of constant impedance load model with DG is

shown in table 5.9. The total active and reactive power loss of IEEE 15 bus radial
distribution system is 50.86 Kw and 46.86 Kvar. The lower voltage bus is 13th bus
and its voltage is 0.9501. By comparing above two models i.e. constant power and
current load model the power losses are decreased and bus voltages are improved.
Table 5.9: Load Flow Calculation of Constant impedance Load Model using PSO
BR SE RE V (p.u) LP (Kw) LQ (Kvar) Pflow (Kw) Qflow (Kvar)
1 1 2 0.9965 28.7463 28.1175 1300.0 982.9
2 2 3 0.9951 8.3001 8.1185 754.6 529.4
3 3 4 0.9946 1.7971 1.7578 415.3 288.5
4 4 5 0.9914 0.0413 0.2314 46.7 32.3
5 2 6 0.9929 4.5660 3.0798 38.0 260.8
6 6 7 0.9791 0.3187 3.8571 151.5 104.9
7 6 8 0.9864 0.0892 0.8652 75.2 52.3
8 2 9 0.9956 0.3878 0.2616 123.8 88.22
9 9 10 0.9953 0.0484 0.0326 47.8 33.99
10 3 11 0.9934 1.5803 1.0659 266.5 184.5
11 11 12 0.9924 0.4297 0.2899 119.2 81.9
12 12 13 0.9921 0.0526 0.0355 46.0 31.5
13 4 14 0.9940 0.1496 0.1009 73.6 50.9
14 4 15 0.9939 0.3208 0.2164 147.2 101.8

The load flow calculation of IEEE 15 bus system with power quality problems
and DG is shown in table 5.10. The total active and reactive power loss is 50432 Kw
and 37821 Kvar. The lower voltage bus is 10th bus and its voltage is 0.001. By
comparing table 5.5 the power losses are decreased and bus voltages are improved.

DEPT. OF EEE Page 33

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
Table 5.10: Load flow calculation of IEEE 15 bus system with power quality
problems using PSO

Br Se Re V (p.u) LP (kW) LQ (kVAR) Pflow (kW) Qflow (kVAR)

1 1 2 0.9790 35419 34643 50745 38027

2 2 3 0.9683 8.2 8.0 825 610
3 3 4 0.9640 1.8 1.8 451 338
4 4 5 0.9633 0.45 0.25 50 38
5 2 6 0.8000 4.7 3.2 358 250
6 6 7 0.9676 0.3 0.2 160 120
7 6 8 0.9683 0.1 0.1 80 60
8 2 9 0.9765 50015 33735 49294 36971
9 9 10 0.0010 41769 28173 50 38
10 3 11 0.9632 1.6 1.1 291 218
11 11 12 0.9601 0.4 0.3 130 98
12 12 13 0.9591 0.1 0.1 50 38
13 4 14 0.9623 0.2 0.1 80 60
14 4 15 0.9622 0.3 0.2 160 120

5.3 TARIFF CALCULATION:

The tariff calculation of each and every branch in IEEE 15 bus radial bus
system is shown in below table 5.11.
Table 5.11: Tariff calculation of IEEE 15 bus system
Bus no Power (kw) at 1 Power in one month Cost in Rupees
hour (kwh) (1 unit = Rs 5)
1 1783.5 1284100 6420600
2 1033.7 744200 3721200
3 563.8 406000 2029800
4 63 45400 226800
5 500.6 360400 1802200
6 200 14400 720000
7 100 72000 360000

DEPT. OF EEE Page 34

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

8 163.1 117400 587100

9 63 45400 226800
10 363.8 261900 1309700
11 163.1 117400 587100
12 63 45400 226800
13 100 72000 360000
14 200 144000 720000

The tariff calculation of each and every branch in IEEE 15 bus radial bus
system with power quality problems is shown in below table 5.12.
Table 5.12: Tariff calculation of IEEE 15 Bus system with power quality problems
Bus no Power (kw) at 1 Power in one month Cost in Rupees / Month
hour (Kwh) (1 unit = Rs 5)
1 63412 4565700 228280000
2 1026 739000 3690000
3 564 406000 2030000
4 63 45000 230000
5 437 305000 1570000
6 200 144000 720000
7 100 72000 360000
8 61618 44365000 221820000
9 63 45000 230000
10 364 262000 1310000
11 163 117000 590000
12 63 45000 230000
13 100 72000 360000
14 200 144000 720000

The tariff calculation with DG of each and every branch in IEEE 15 bus radial
bus system is shown in below table 5.13.

DEPT. OF EEE Page 35

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
Table 5.13: Tariff calculation of IEEE 15 bus system using PSO
Bus no Power (kw) at 1 Power in one month Cost in Rupees
hour (kwh) (1 unit = Rs 5)
1 1753.7 1262700 6313500
2 1025.9 738700 3693300
3 563.6 405800 2029000
4 63 45400 226800
5 485.4 349500 1747500
6 192.3 138500 692400
7 100 72000 360000
8 163.1 117400 587000
9 63 45400 226800
10 363.6 261800 1308900
11 163.1 117400 587000
12 62.8 45216 226080
13 100 72000 360000
14 200 144000 720000

The tariff calculation of each and every branch in IEEE 15 bus radial bus
system with power quality problems and DG is shown in below table 5.14.
Table 5.14: Tariff calculation of IEEE 15 Bus system with power quality problems
using PSO
Bus no Power (kw) at 1 Power in one Cost in Rupees / Month
hour month (Kwh) (1 unit = Rs 5)
1 63413 4565800 228290000
2 1026 739000 3690000
3 558 402000 2010000
4 63 45000 230000
5 438 316000 1580000
6 196 141000 710000
7 100 72000 360000

DEPT. OF EEE Page 36

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

8 61618 44365000 221820000

9 63 45000 230000
10 364 262000 1310000
11 163 117000 590000
12 63 45000 230000
13 100 72000 360000
14 200 144000 720000

5.4 COMPARISON OF RESULTS:

5.4.1 Load flow results comparison:
The normal load flow calculation of IEEE 15 bus system with and without DG
is shown in table 5.15. The active and reactive power losses are decreased and voltage
profile is improved after connecting DG.
Table 5.15: Load flow results comparison

Without DG With DG
Total Active Losses 61.76 KW 44.59 KW

Total Reactive Losses 57.27 KVAR 41.37 KW

Vmin 0.9446 0.9591

Weak bus 13 13

5.4.2 Load Modeling results comparison:

The active and reactive power losses and bus voltage of Constant power load
model comparison with and without DG is shown in table 5.16. The active and
reactive power losses are decreased and voltage profile is improved after connecting
DG.
Table 5.16: Constant power load model comparison

Constant power model Without DG With DG

Total Active Losses 61.76 KW 56.14 KW
Total Reactive Losses 57.27 KVAR 57.33 KW
Vmin 0.9446 0.9914
The active and reactive power losses and bus voltage of Constant current load
model comparison with and without DG is shown in table 5.17. The active and

DEPT. OF EEE Page 37

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE
reactive power losses are decreased and voltage profile is improved after connecting
DG.
Table 5.17: Constant current load model comparison

Constant current model Without DG With DG

Total Active Losses 55.87 KW 51.26 KW
Total Reactive Losses 51.80 KVAR 52.47 KW
Vmin 0.9475 0.9918

The active and reactive power losses and bus voltage of Constant impedance
load model comparison with and without DG is shown in table 5.18. The active and
reactive power losses are decreased and voltage profile is improved after connecting
DG.
Table 5.18: Constant impedance load model comparison

Constant impedance Without DG With DG

model
Total Active Losses 50.86 KW 46.82 KW
Total Reactive Losses 46.86 KVAR 48.03 KW
Vmin 0.9501 0.9921
5.4.3 Power Quality problems results comparison:

The active and reactive power losses and bus voltage of power quality
problems comparison with and without DG is shown in table 5.19. The active and
reactive power losses are decreased and voltage profile is improved after connecting
DG.
Table 5.19: Power Quality problems results comparison

Without DG With DG
Total Active Losses 168520 KW 50432 KW
Total Reactive Losses 127910 KVAR 37621 KW
Vmin 0.001 0.001
Weak bus 10 10

DEPT. OF EEE Page 38

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMS
AT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

CHAPTER 6
CONCLUSION AND FUTURE SCOPE OF
WORK

6.1 CONCLUSION:

A particle swarm optimization for placement and sizing of DG is used

minimizing the total power losses under voltage unbalance and power quality
problems conditions. This optimized method is fast and accurate in determining the
sizes and location of DG with economic aspects. And it is tested on IEEE 15 bus
distribution system. By this analysis end user can easily avoids the power losses and
save the money, similarly the system performance enhanced.

6.2 FUTURE SCOPE OF WORK:

From future scope point of view, following points may too be considered to
apply this proposed work:

 Placement and Sizing of multiple DG’s and their ratings for improvement of
voltage profile should be developed using PSO.

 The cost analysis approach suggests to do further research for optimal

Placement and Sizing of DG by using advanced optimization techniques.

DEPT. OF EEE Page 39

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY
PROBLEMSAT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

FINANCE AND PROJECT MANAGEMENT

Hardware
Since it is a simulation based project a commercial PC with the following
configuration is used.

Table 7.1: System Configuration and price

Unit type Unit model Price (Rs.)

CPU Intel Core i5
Clock 2.67 GHz
42,500.00
RAM 4GB
HDD 500GB

Software
The software used for this project is MATLAB R 2011a and 2018a with
additional Toolboxes. The price of each of these Toolboxes along with the total cost is
shown in Table 7.2

Table 7.2: Software configuration and price

Module type Price (Rs.)

MATLAB 2011a, 2018a version 1109558/-
(Purchased by college including tool boxes)
Power system and Power Electronics Toolbox

Total 1109558/-

DEPT. OF EEE Page 40

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY
PROBLEMSAT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

TIME MANAGEMENT AND SOCIETAL &ENVIRONMENTAL

IMPACT
Time Management
The total duration of the project work is 24 weeks with 36 hours spent each
week. This entire duration can be divided in to two phases with each phase taking
eighteen weeks. In phase 1, presentation 1, complete literature survey is done. In
phase 1, presentation 2, the problem in existing method is identified. Mathematical
model of the proposed method is developed and identified the software version that is
required to simulate the proposed algorithm i.e., Matlab R2014a.

In phase 2, presentation1,gone through some books to acquaint myself to the

Matlab software and some important tool boxes that are available in the Matlab. The
Simulink model for the proposed method is simulated the code using Matlab R2014a
and alsoin phase 2, the limitations of the proposed algorithm in the base paper are
found. Identified the new method by having a study on various papers to overcome
the problems in base paper algorithm. Developed the Mathematical model of the new
method. The Simulink model is developed for the new method and simulated the
project successfully.

Societal and Environmental Impact

By this PSO technique it is easy to analyze the system performance. The
distribution end users can easily identify the power losses and tariff variations. By
implementing the proper DG size at an exact location helps to improve the
distribution system performance and save the money of the customers.

DEPT. OF EEE Page 41

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY
PROBLEMSAT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

REFERENCES
[1] Ahmadreza Eslami, Mohammad Esmail Hamedani Golshan, ―Monte-Carlo
based approach to consider the cost of voltage dip and long duration
interruption in optimal planning of SDGs‖, IET Gener. Transm. Distrib., Vol.
12 Iss. 8, pp. 1856-1865, 2018.
[2] Priyanka Gangwar, Sri Niwas Singh, Saikat Chakrabarti, ―Multi-objective
planning model for multi-phase distribution system under uncertainty
considering reconfiguration‖, IET Renew. Power Gener., Vol. 13 Iss. 12, pp.
2070-2083, 2019.
[3] Huayi Wu, Ping Dong, ―PPSO method for distribution network
reconfiguration considering the stochastic uncertainty of wind turbine
photovoltaic and load‖, This is an open access article published by the IET
under the Creative Commons J. Eng., Vol. 2017, Iss. 13, pp 2032-2036, 2017.
[4] Bashir Bakhshideh Zad, Jacques Lobry, François Vallee, ―Impacts of the
model uncertainty on the voltage regulation problem of medium-voltage
distribution systems‖, IET Gener. Transm. Distrib., Vol. 12 Iss. 10, pp. 2359-
2368, 2018.
[5] Alberto Escalera, Barry Hayes, Milan Prodanovic, ―Analytical method to
assess the impact of distributed generation and energy storage on reliability of
supply‖, IET CIRED, Open Access Proc. J., Vol. 2017, Iss. 1, pp. 2092–2096,
2017.
[6] Mehrdad Tarafdar Hagh, Payman Amiyan, Sadjad Galvani, Naser Valizadeh,
―Probabilistic load flow using the particle swarm optimization clustering
method‖, IET Gener. Transm. Distrib., Vol. 12 Iss. 3, pp. 780-789, 2018.
[7] Yang Fu, Huanhuan Liu, Xiangjing Su, Yang Mi, Shuxin Tian, ―Probabilistic
direct load flow algorithm for unbalanced distribution networks considering
uncertainties of PV and load‖, IET Renew. Power Gener., Vol. 13 Iss. 11, pp.
1968-1980, 2019.
[8] Andres Arguello, Jose D. Lara, Jose David Rojas, Gustavo Valverde, ―Impact
of Rooftop PV Integration in Distribution Systems Considering
Socioeconomic Factors‖ IEEE Systems Journal, Vol. 12, No. 4, December
2018.

DEPT. OF EEE Page 42

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY
PROBLEMSAT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

[9] A. S. Nazmul Huda, Rastko Zivanovic, ―An Efficient Method for Distribution
System Reliability Evaluation Incorporating Weather Dependent Factors‖,
IEEE conference paper 2019.
[10] Mike Brian Ndawula, Ignacio Hernando-Gil, ―Evaluation of Customer-
oriented Power Supply Risk with Distributed PV-Storage Energy Systems‖,
IEEE conference paper 2019.
[11] D Das, D P Kothari, A Kalam, ―Simple and efficient method for load flow
solution of radial distribution networks‖, Electrical Power & Energy Systems,
Vol. 17, No. 5, pp. 335-346, 1995.
[12] K. Nagaraju, S. Sivanagaraju, T. Ramana & P. V. Prasad, ―A Novel Load Flow
Method for Radial Distribution Systems for Realistic Loads‖, Electric Power
Components and Systems, Vol 39, no. 2 pp. 128–141, 2011.
[13] V. V. S. N. Murty, B. Ravi Teja, and Ashwani Kumar, ―A Contribution to Load
Flow in Radial Distribution System and Comparison of Different Load Flow
Methods‖, IEEE Conference paper, pp. 6-11, jan 2014.
[14] Gundugallu Peddanna, Y. Siva Rama Kishore, ―Power Loss Allocation of
Balanced Radial Distribution Systems‖, International Journal of Science and
Research (IJSR), Vol. 4, Iss 9, September 2015.
[15] Gurpreet Kaur, Harmeet Singh Gill, ―Power Flow Analysis of Radial
Distribution System using Backward/Forward Sweep Method‖, International
Journal for Research in Applied Science & Engineering Technology
(IJRASET), Vol. 5, Iss 7, July 2017.
[16] R. C. Dugan, M. F. McGranaghan, S. Santosa, H. W. Beaty, ―Electrical Power
Systems Quality‖, 3rd edition, McGraw-Hill Education, July 2017.
[17] Eklas Hossain, Mehmet Rida Tur, Sanjeevikumar Padmanaban , Selim Ay,
Imtiaj Khan, ―Analysis and Mitigation of Power Quality Issues in Distributed
Generation Systems Using Custom Power Devices‖, IEEE journal, Vol. 6,
March 2018.
[18] Juan C. Cebrian, Nelson Kagan, Jovica V. Milanovic, ―Probabilistic Estimation
of Distribution Network Performance With Respect to Voltage Sags and
Interruptions Considering Network Protection Setting—Part II: Economic
Assessment”, IEEE Transactions on power delivery, Vol. 33, Iss 1, Feb 2018.

DEPT. OF EEE Page 43

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY
PROBLEMSAT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

[19] Daniel Ogheneovo Johnson, Kabiru Alani Hassan, ―Issues of Power Quality in
Electrical Systems‖, International Journal of Energy and Power Engineering,
Vol. 5, Iss 4, pp. 148-154, Jan 2016.
[20] Oladepo Olatunde and Awofolaju Tolulope Tola, ―Power Quality
Improvement in Electrical Distribution Network‖. American Journal of
Engineering Research (AJER), Vol. 5, Iss 12, pp. 224-227, 2016.
[21] Karar Mahmoud, Naoto and Abdella Ahmed, ―Optimal Distributed Generation
Allocation in Distribution Systems for Loss Minimization‖, IEEE
Transactions On Power Systems,2015.
[22] D.B. Prakasha, C. Lakshminarayana, ―Multiple DG Placements in Distribution
System for Power Loss Reduction Using PSO Algorithm‖, Elsevier, pp. 785 –
792, 2016.
[23] Mounika Lakshmi Prasanna K, Amit Jain and James Ranjith Kumar, ―Particle
Swarm Optimization Application for Optimal Location of Multiple Distributed
Generators in Power Distribution Network‖, J Electr Electron Syst, an open
access journal, Vol. 6, Iss. 4, 2017.
[24] V.Chandra Naga Hasa, T.Padmavathi, ―Optimal Allocation of Distributed
Generation in Radial Distribution System‖, International Journal of
Electronics, Electrical and Computational System, Vol. 7, Iss. 3, March 2018.
[25] B. Srinivasa Rao and G.V.S. Krishna Rao, ―Optimal Placement of DG Unit for
Loss Reduction in Radial Distribution System: A Case Study‖, International
Journal of Electronic and Electrical Engineering, Vol. 4, pp. 313-320,
Number 2011.
[26] S. Satyanarayana, T. Ramana, S. Sivanagaraju & G. K. Rao, ―An Efficient
Load Flow Solution for Radial Distribution Network Including Voltage
Dependent Load Models‖, Electric Power Components and Systems, Vol. 35,
pp. 539–551, 2007.
[27] P. Dinakara Prasad Reddy, V. C Veera Reddy and T. Gowri Manohar, ―An
Efficient Distribution Load Flow Method for Radial Distribution Systems with
Load Models‖, International Journal of Grid and Distributed Computing, Vol.
11, No. 3, pp.63-78, 2018.
[28] M.H. Haque, ―Load flow solution of distribution systems with voltage
dependent load models‖ Electric Power Systems Research, Vol 36 pp. 151 –

DEPT. OF EEE Page 44

 ANALYSIS OF COST BY MITIGATING THE POWER QUALITY
PROBLEMSAT DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

156. 1996.
[29] Devender Singh, R. K. Misra, and Deependra Singh, ―Effect of Load Models
in Distributed Generation Planning‖, IEEE Transactions On Power Systems,
Vol. 22, NO. 4, November 2007.
[30] Rahul Sharma and Varsha Mehar, ―International journal of engineering
Sciences & management Power Flow Analysis With Voltage Dependent
Load‖, International Journal Of Engineering Sciences & Management, Dec
2016.
[31] Krischonme Bhumkittipich and Weerachai Phuangpornpitak, ―Optimal
Placement and Sizing of Distributed Generation for Power Loss Reduction
using Particle Swarm Optimization‖, Elsevier Energy Procedia, Vol 34, pp.
307 – 317, 2013.
[32] Wichit Krueasuk and Weerakorn Ongsakul, ―Optimal Placement of
Distributed Generation Using Particle Swarm Optimization‖, Research gate,
2014.
[33] K. Ras Guerriche and T. Bouktir, ―Optimal Allocation and Sizing of
Distributed Generation with Particle Swarm Optimization Algorithm for Loss
Reduction‖, Revue des Sciences et de la Technologie, Vol. 6, pp. 59-69, 2015.
[34] Naveen Jain, S.N. Singh and S.C. Srivastava, ―Particle Swarm Optimization
Based Method for Optimal Siting and Sizing of Multiple Distributed
Generators‖, 16th National Power Systems Conference, pp. 669-674,
December 2010.

DEPT. OF EEE Page 45

ANALYSIS OF COST BY MITIGATING THE POWER QUALITY PROBLEMSAT
DISTIBUTION SYSTEM USING OPTIMIZATION TECHNIQUE

AUTHOR’S PUBLICATIONS
[1] M. Ramalinga Reddy, U. Kamal Kumar and Suresh Babu Daram, “Enhancing
the performance of Distribution systems using PSO”, Reaching the Heights of
Technology Trends, 2020. (Communicating)

DEPT. OF EEE Page 46

 (AUTONOMOUS)
SreeSainath Nagar, A.Rangampet - 517 102

Department of Electrical and Electronics Engineering

PROJECT TITLE: Analysis of Cost by Mitigating the Power Quality

Problems at Distribution System Using Optimization Technique

ABSTRACT: In recent days the distribution network affected with load

uncertainties, which ever adhere the power quality problems. The quality
problems are isolates the healthy network into unbalanced and collapse the
devices in the different factors like, damages, power factor variations and
changes in the tariffs rates at end user side etc. Mostly due to sudden load
variations the voltage dips and long duration interruption will cause to changes in
cost at utility and end user. This work proposes the analysis of cost under the
power quality problems. Therefore in this scenario the quality of supply may
enhanced using distributed generations (DGs). The another approach is to sizing
and placement of DG, so far an optimized technique is proposed. The impacts of
DG placement at distributed networks are also analyzed. This proposed work
towards the detailed analysis of tariff at uncertainties and provide the best
solution to mitigate the power quality problems using optimized technique with
the help of MATLAB/SIMULINK software.

PROJECT CANDIDATE :
M. RAMALINGA REDDY 18121D0710

GUIDE: Dr. G.S MAHESH

Associate Professor, Dept. of EEE.

COURSE : PROJECT WORK SUBJECT CODE: 16MT30731

CO-POs-PSOs Attained:

Program Specific
Program Outcomes
Outcomes
Course
Outcomes PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 PO11 PSO1 PSO2 PSO3 PSO4
CO1.
demonstrate
capacity to
identify an √ √
advanced topic
for project work
in core and allied
areas

CO2. analyze
the problem and
derive an
√ √
optimal solution
pertinent to the
chosen topic
CO3. solve
engineering
problems and
provide a wide √ √ √
range of
potential
solutions
CO4.
comprehend
extracted
information
through the
literature survey
for design and √
development of
engineering
problems
pertinent to the
chosen topic.
CO5. use the
techniques, skills
and modern
√ √ √
engineering tools
necessary for
project work.
Co6. contribute
to
multidisciplinary
scientific work √ √
in the field of
Electrical power
Systems.
CO7. execute
the project
effectively and
efficiently √
considering
economical and
financial factors.
CO8. plan, √
prepare and
present effective
written and oral
technical report
on the topic.
Co9. engage in
lifelong learning
for development
of technical
competence in √
the field of
Electrical
power systems
and allied fields.
CO10.
understand
ethical
responsibility
towards
environment and √
society in the
field of
Electrical
Engineering.
CO11. adapt to
independent and
reflective
learning for √
sustainable
professional
growth.

Signature of the Guide

IET Generation, Transmission & Distribution

Research Article

Monte-Carlo based approach to consider the ISSN 1751-8687

Received on 30th June 2017
Revised 20th November 2017
cost of voltage dip and long duration Accepted on 20th December 2017
E-First on 19th February 2018
interruption in optimal planning of SDGs doi: 10.1049/iet-gtd.2017.1026
www.ietdl.org

Ahmadreza Eslami1 , Mohammad Esmail Hamedani Golshan1

1Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan, Iran
E-mail: a.eslami@ec.iut.ac.ir

Abstract: Voltage dip and long duration interruption (LDI) are among the most costly power quality phenomena. In this study, a
Monte-Carlo based approach is proposed to consider the cost of sensitive loads disruption caused by dip and LDI in the optimal
planning of synchronous distributed generations (SDGs). The idea is to link between trip probability due to dip and LDI and their
yearly costs by employing Monte-Carlo simulation and acquiring their total costs during the planning horizon. The addition of
disruption cost along with the traditional planning objectives like network upgrade cost and loss cost allows utilities to include the
customer's perception during planning. A formula for the probability of disruption due to dip is derived and a modified Monte-
Carlo approach is proposed. The methodology is illustrated on the distribution level of the IEEE 30-bus system and the
optimisation problem is solved by particle swarm optimisation algorithm. The results demonstrate that the sensitive loads
performance is improved from the dip standpoint in the presence of SDGs. However, the LDI cost is either not affected or
aggravated by the presence of SDGs depending on the protection model. The total disruption cost is decreased.

1 Introduction the multistage expansion planning problem of the distribution

system where investments in the distribution network and
Nowadays, power systems are facing new issues such as distributed generation were jointly considered [14]. Abu-mouti et
deregulated market, sensitive loads, environmental effects and al. presented an optimisation approach that employed an artificial
economic concerns which require special consideration. Therefore, bee colony algorithm to determine the optimal size, power factor
optimisation programs and optimal planning are going to play an and location of DG units in order to minimise the total system real
indisputable role in distribution networks. On the other hand, power loss [15]. In a class of DG planning, researchers maximised
distributed generations (DGs) are introducing interesting choices in DG penetration level subject to technical constraints [16, 17].
network expansion planning that can bring many technical and Pandi et al. presented a methodology to maximise DG penetration
economic benefits to power systems. DGs can improve power level considering allowable harmonic and protection coordination
quality and reliability and reduce system power loss. They can also constraints [17].
be considered as reinforcement options for system loadability From a sensitive load standpoint, voltage dip and long duration
problems and defer network investments. The DGs’ impacts will, interruption (LDI) could bear a large economic meaning and an
however, depend on several parameters, such as size, type, and enormous amount of money could be saved if they are mitigated.
location [1]. Installation of these resources in non-optimised In [18], an objective function for optimal siting and sizing of
locations and capacities may increase power loss [2], decrease photovoltaic DGs was proposed which considered real power loss,
power quality and reliability [3] and impose extra expenditure on voltage deviation, total harmonic distortion and voltage dip
the network. Therefore, prior-to-installation evaluation and optimal magnitude. The objective function was totally technical though
planning seem essential to make DG a cardinal player in economic assessment is of great importance in planning problems.
distribution networks. In [19], an index-based approach was proposed to consider voltage
In terms of classification, DGs planning literature is classified dip magnitude and the number of trips due to dip in the optimal
into some categories based on planning objectives including planning of SDGs. Although this reference presented useful indices
economic [1, 4, 5], technical [6–9] and combined objectives [10, and an economic interpretation of a sensitive load trip index, it did
11] along with environmental concerns [12]. In [1], a DG planning not consider a detailed stochastic model of dip. The focus of this
method was proposed which considered the cost of network study was mostly technical concerns. The authors of [20, 21]
investment, DG cost and the power loss cost as the objective proposed methods to consider voltage dip cost in DG planning
function. The effect of DGs on network investment deferral was while in both works, very few faults were simulated and the
examined while considering different types of ownership. probabilistic nature of an event like voltage dip was almost
Khanbabapour et al. proposed a multi-objective index as the ignored. Moreover, in all three works [19–21], the behaviour of
objective function of the synchronous distributed generation (SDG) sensitive loads was too simplified so that it might not track and
planning program which considered a wide variety of technical implement the real behaviour. Besides, they did not consider LDI
factors including power loss, voltage deviation, system loadability events.
and overcurrent and anti-islanding protection speed [7]. Voltage SDGs are capable of helping network in supporting sensitive
rise issue along with greenhouse gas emission and reliability loads. Synchronous generators have a positive effect on the
criterion in the optimal planning of DGs were addressed by the retained voltage (the lowest voltage during an event) by injecting
authors of [9,12], respectively. In [10], an approach was proposed reactive power into the network, increasing system short circuit
which maximises system loading margin along with economic level and contributing to fault current. In the present paper, a
profit subject to system constraints. In [13], several alternatives for modified Monte-Carlo simulation (MMCS) method with new
network expansion such as installation of new feeders, variables is proposed to involve the effect of voltage dip and LDI
transformers, and DGs were evaluated. A method was proposed to in the optimal SDG planning program. The traditional planning
consider the best expansion plan under uncertainties related to objectives including network investment deferral and real power
demand and renewable energy sources. Delgado et al. addressed loss cost are also taken into consideration. Accordingly, this study

IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865 1856
© The Institution of Engineering and Technology 2018
 sensitivity. Therefore, a typical type of sensitive equipment is
assumed by deriving an average rectangular curve between PLCs,
ASDs, and PCs

T min Typical = 1/3[T min PLC + T min ASD + T min(PC)], (1)

T max Typical = 1/3[T max PLC + T max ASD + T max(PC)], (2)

V min Typical = 1/3[V min PLC + V min ASD + V min(PC)], (3)

V max Typical = 1/3[V max PLC + V max ASD + V max(PC)] . (4)

The parameters for this typical sensitivity curve are Tmin = 25 

ms, Tmax = 260 ms, Vmin = 45% and Vmax = 75%. This typical
equipment is used wherever a sensitive load is involved in this
study. However, the proposed method is compatible with the
detailed development of process trip probability presented by
Cebrian et al. [23] and could be customised for any specified and
Fig. 1  Sensitivity curve of the typical sensitive equipment practical sensitive load scenario.
Let U1 be a uniform CDF in a manner that U1 V max = 0 and
U1 V min = 1 and U2 be a uniform CDF so that U2(T min) = 0 and
U2(T max) = 1, then for an event with a magnitude of V and duration
of T, the probability of the equipment trip is calculated by the joint
probability function of U1 and U2 as follows:

Ptrip T, V = U1 V × U2 T . (5)

Equation (5) shows that if T < T min or V > V max, Ptrip = 0 and if
T > T max and V < V min simultaneously, Ptrip = 1. If T > T max,
Ptrip = U1(V) and if V < V min, Ptrip = U2(T).
The time in which a network fault could vanish is a stochastic
variable which depends on several factors such as the nature of the
Fig. 2  Cumulative probability versus the fault extinction time [24] fault, the natural causes and the characteristics of the network. In
[24], a procedure is proposed to link the operation time of the
appends a cost term to the traditional planning objectives and protection system to the natural extinction time of a fault. Fig. 2
forms a comprehensive cost function to involve the costs of voltage shows the cumulative probability density function of the fault
dip and LDI in SDG planning programs. The behaviour of sensitive extinction time for which a uniform distribution function is
loads is simulated using the model of [22] which considers load assumed where an 80% probability corresponds to fault extinction
trips in a probabilistic manner. A formula for the probability of time in the range from 0 to 100 ms and a 20% probability
disruption due to dip is derived and the dip and LDI cost functions corresponds to the range from 100 ms up to 5 s [24]. For a
per event are calculated to be used in MMCS. The remainder of the protection system operation time of T i and a fault extinction time
paper is presented as follows. Section 2 presents the stochastic of T ext, if T ext < T i, then the fault would vanish before the
evaluation of sensitive loads and the MMCS method. Sections 3 protection system interference and the buses of the system only
and 4 present the comprehensive cost function formation and the experience dip events with the duration of T ext. On the other hand,
optimisation problem formulation, respectively. Section 5 describes for T ext ≥ T i, the protection system would operate and trip the
the system under evaluation. Section 6 presents and discusses the breaker which leads to supply interruption for a part of the system
results and finally, Section 7 presents the conclusion. depending on the location of the protective device and the topology
of the network. Other buses might only experience a dip event with
2 Stochastic evaluation of voltage dip and LDI the duration of T i. Once the protection time is known, the CDF of
Voltage dip and LDI are categorised among the most costly power T ext enables us to achieve a dip cumulative probability value (Pdip)
quality phenomena. However, as voltage dip occurs more for T i. For buses which will experience voltage interruption due to
frequently than LDI, it leads to more yearly economic loss. Short protection system operation, the interruption probability value
circuit faults are among the main causes of voltage dip and LDI. could be achieved as the complement of Pdip [24]
The same short circuit fault causing a dip in one node may cause an
interruption in another node due to the operation of the protection PLDI(T i) = 1 − Pdip(T i) . (6)
system. If the fault is permanent, this interruption will be an LDI.
For a specified T i, the buses experiencing LDI and their
2.1 Probability of disruption due to dip and LDI probability are determined. However, the equipment trip due to dip
requires a different treatment. Having Fig. 1 in mind, the range of
Every process is controlled by a series and parallel combination of
(T min, T i) could be divided into nr sub-ranges in the form of
control equipment such as adjustable speed drives (ASDs),
programmable logic controllers (PLCs), personal computers (PCs), (T j − 1, T j). According to [24], the probability of voltage dip for the
etc. The equipment sensitivity to voltage dip and LDI is usually sub-range of (T j − 1, T j) is equal to Pdip T j − Pdip(T j − 1). For each
expressed in terms of magnitude and duration of the event. sub-range, the probability of disruption due to dip is considered as
Milanovic and Gupta [22] have introduced a rectangular voltage the product of the probability of sensitive equipment trip (Ptrip) and
tolerance curve for ASDs, PLCs, and PCs as shown in Fig. 1. The the probability of dip occurrence in that range
probability of equipment trip inside the shaded area is modelled by
two cumulative distribution functions (CDFs) for voltage Tj−1 + Tj
magnitude and duration of the event. The aim of the present study PDV j = Pdip T j − Pdip T j − 1 × Ptrip , (7)
2
is to propose an approach to consider the disruption cost in SDG
planning and it is not focused on the detailed evaluation of process

IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865 1857
© The Institution of Engineering and Technology 2018
Table 1 Required values nr for errors <1.5% MMCS which ease the consideration of protection system
T i range, ms (0, 90) (90, 240) (210, 570) operation and LDI cost in sensitive loads analysis.
nr 2 10 20
2.3.1 MMCS uncertainty factors: The main uncertainty factors
max error (%) 1 × 10−13 1.12 1.489
of voltage dip and LDI evaluation in this study are as follows.

a. a Fault type: short circuit faults considered are (1) three-

phase, (2) single-phase-to-ground, (3) two-phase and (4)
two-phase-to-ground. In [25], the probability of fault types 1,
2, 3 and 4 are reported to be PLLL = 0.03, PLG = 0.78, PLL = 
0.04 and PLLG = 0.15, respectively. Let UN1 be a uniformly
distributed number, then the fault type indicator is calculated
as follows [25]:

FTi =

i = 1, ifUN1 ≤ PLLL, (11)

i = 2, ifPLLL < UN1 ≤ PLLL + PLG ,

i = 3, if PLLL + PLG < UN1 ≤ (PLLL + PLG + PLL),
i = 4, if PLLL + PLG + PLL ≤ UN1 .
b. b Fault location: fault location is a probabilistic parameter
and can be represented by a probability distribution. It is
supposed that faults can occur either in buses or lines. Line
faults close to a bus are assumed to be the same as faults in
that bus. Let L be the number of system lines and UN2 be a
Fig. 3  Flowchart of stochastic analysis for obtaining costs per event uniformly distributed number between [0,1). The line in
which fault occurs is obtained as follows:
where PDV j is the probability of disruption due to dip in sub-range
(T j − 1, T j). The total probability of disruption due to dip is FL = L × UN2 + 1, (12)
calculated as the summation of PDV j’s:
where the square bracket notation ([ ]) in (12) represents the
nr floor function. When the fault line is chosen that line with its
PDV T i, V = ∑ PDV j . (8) sending and receiving nodes are considered as a set which
j=1 can locate the fault anywhere in its domain. The per cent of
the line length (llp %) from the sending node to the fault
For different values of Ti and nr, PDV is calculated and the location is presented by (13) which acquires the decimal part
convergence error (relative to nr = 100 which provides complete of L × UN2
convergence) is obtained. Table 1 shows the required value of nr to
get errors <1.5% for different Ti ranges. llp% = L × UN2 + 1 − FL × 100% . (13)

2.2 Cost analysis To make it clear with an example, suppose that L = 10 and
UN2 = 0.58 then line l = 6 is chosen as the fault line by (12)
The linear cost model is used in this study. The costs of disruption and llP = 80%(the decimal part of L × UN2) which means
due to voltage dip and LDI are calculated as follows. They have that if line l is laid between buses i and j, the fault location is
been considered for one single event somewhere in the 80% of the line length from bus i. It is
worth mentioning that if llp was <10%, bus i and if it was
Cdip = PDV (T i, V) × Pload × Cudip, (9)
>90%, bus j would be chosen as the fault location.
c. c Fault impedance: this study models the fault impedance by
CLDI = PLDI(T i) × Pload × CuLDI × HLDI, (10) a normal distribution function with an average of μ = 5 Ω and
a standard deviation of σ = 1 Ω ( f imp) [25].
where Pload is the amount of disrupted load, Cudip is the cost of dip
d. d Duration of protection system operation: the time lasting
per kW, CuLDI is the cost of LDI per kW per hour and HLDI is the
for protection system to clear the fault and isolate the faulty
duration of power supply interruption in hours. Cdip and CLDI part is another key factor in our calculation. In this study, a
represent the costs of one dip and one LDI event at the sensitive definite-time model is considered for protection system
load busbar, respectively. As PDV and PLDI are both functions of T i which assigns two definite operation times to the primary
and V, by employing (9) and (10), the cost terms Cdip and CLDI are and backup protection systems and a probability of failure of
derived as functions of T i and V as shown in Fig. 3. primary protection is considered. It will lead to the reaction
of backup protection if the primary one fails to clear the
2.3 MMCS method fault. Table 2 shows the typical operation time for both
protections and the probability of probability of primary
Monte-Carlo simulation is an effective way of assessing stochastic protection failure (PPPF) for the 33 kV voltage level [26]. To
events and problems with a stochastic nature. In [25], a Monte involve T i in Monte-Carlo samples, let UN3 be a uniformly
Carlo simulation approach was introduced to evaluate voltage dip distributed number under [0,1], then the protection system
magnitude. In the present paper, an MMCS is proposed with a new operation time (T i) is expressed as
formula for fault location which makes it possible to model faults
both in buses and lines. Moreover, the terms of protection system
operation time and LDI duration are contributed in the proposed

1858 IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865
© The Institution of Engineering and Technology 2018
Table 2 Typical fault clearing time of primary and backup N N
∑i =b 2 ∑s =f 1 CLDI
protection [26] CLDI − year = , (19)
Nyear
Voltage level, Fault Typical fault clearing time, PPPF
kV location ms where Nb is the number of network buses. Cdip and CLDI for every
Primary Backup sample are calculated from the previous step. The cost of dip and
33 bus 150 550 0.0202 LDI for buses not hosting sensitive loads is assumed to be zero.
33 line 60 550 0.0119 The disruption cost is considered for one year of operation though
the planning program is performed for N years with the load
growth rate of rLG and the real discount rate of d. The dip and LDI
tp, if UN3 > PPPF, cost model in this study is a linear function of the load's active
Ti = (14) power. Therefore, it is reasonable to assume that the yearly cost of
tb, if UN3 ≤ PPPF,
disruption will grow proportional to the load growth as we advance
where tp and tb are the primary and backup protection times, in the planning horizon. This way, the disruption cost will be
calculated in each year of planning and by the real discount rate, it
respectively. Uniform distribution is chosen for UN3 because will be transformed to the present year. So, the total costs of
of the fact that the probability of UN3 being less than PPPF is disruption due to dip and LDI are calculated as follows:
equal to PPPF (P(UN3 < PPPF) = PPPF). It declares in other
words that the probability of primary protection failure is N
(1 + rLG)Y − 1
equal to PPPF which is our initial assumption. Cdip − T = Cdip − year × ∑ , (20)
e. e Power supply interruption duration: in order to analyse Y=1 (1 + d)Y
how an LDI affects the equipment, it is important to know
N
how long a customer is going to be without electricity after (1 + rLG)Y − 1
the fault. This duration consists of fault locating time, repair CLDI − T = CLDI − year × ∑ . (21)
Y=1 (1 + d)Y
time, etc. It is referred to as the cumulative probability curve
of power supply interruption duration (CSID) [23]. In this
study, the CSID is modelled by a normal distribution 3 Cost function formation
function with an average of 8 h and the standard deviation of In this study, two other cost terms are considered besides the cost
4 h. For this curve, the maximum supply off-time is 24 h and terms for voltage dip and LDI which were explained in the
in 24 h, 100% of LDIs would be repaired. Ninety per cent of previous section. The terms considered are the cost of system lines
LDIs will be repaired in 13 h and 70% will be repaired in 10  upgrade and the cost regarding the active power loss.
h. Let UN4 be a uniformly distributed number, then the
power supply interruption duration is formulated as follows: 3.1 Network upgrade cost
−1
HLDI = CSID UN4 . (15) The network upgrade is one of the main factors in distribution
system planning. The SDGs presence could defer the network
Equation (15) projects UN4 to HLDI by applying the described upgrade cost. This effect was modelled by the real discount rate in
CSID. [1] which transformed the deferred investment to the present year
value and gave an economic sense of the investment deferral. Let
Y
2.3.2 MMCS results analysis: An overall system sample which CUP be the total network upgrade cost due to load growth in year Y,
reflects all of the system uncertainties is expressed by the vector then the present value for such an investment is calculated using
Sk. f imp is a normally distributed number which can be generated the real discount rate and the yearly costs are added to obtain the
by a random number generator. By generating UN1, UN2, UN3 and total network upgrade cost during the planning horizon
UN4 and employing (11)–(15), the variables FTi, FL, llp, T i and N Y
CUP
HLDI would be determined for every sample. Hence, for each set of CDef = ∑ . (22)
generated random numbers, the MMCS sample is representing a Y=1 (1 + d)Y
specified fault
It is assumed that the transformers exhibit extra headroom for
SK = FT, FL, llp, f imp, T i, HLDI . (16) future demand so that only the cost of feeders upgrade is
considered. The feeders upgrade cost is accounted for by a linear
The MMCS sampling is performed. Each sample projects a model which is similar to [4] varies between 178 £/kW (for the
determined fault in the system. For the specified fault, during-fault shortest feeder) and 445 £/kW (for the longest feeder). The costs
voltages are obtained. T i is also specified from the previous step. are per kW of the newly installed capacity. The cost for any other
By employing Fig. 2 and applying (6)–(8), the probabilities of the feeder falls between these two values with regard to the feeder's
trip due to dip and LDI are calculated for all sensitive loads and the length.
costs of dip and LDI per event are obtained for every busbar
hosting a sensitive load using (9) and (10). The total costs of dip 3.2 Network loss cost
and LDI during MMCS are acquired as the summation of cost
The annual loss incentive values losses at CLossy (£/MWh) and
terms obtained from (9) and (10). Let Nf be the number of
rewards or penalises actual annul loss (ALossy, MWh) relative to
simulated faults which form the samples of MMCS, then the
number of years assigned to the simulation is calculated as the target level (TLossy, MWh). Equation (23) presents the annual
loss cost for year Y, based on the described strategy [1]
Nf
Nyear = , (17) LCY = CLossY ALossY − TLossY . (23)
LT × rf

where LT is the total length of the network's lines and rf is the fault If the annual loss is reduced with regard to the target level, LCy is
rate per km per year. The cost of LDI and dip for one year is negative which means that the network is profiting from reducing
formulated as follows: its loss. If the annual loss is increased with regard to the target
level, LCy is positive and the network is penalised for its excessive
N N
∑i =b 2 ∑s =f 1 Cdip loss. The present value of the total loss cost (LCT) during the
Cdip − year = , (18)
Nyear

IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865 1859
© The Institution of Engineering and Technology 2018
planning horizon is calculated by transforming the loss cost of each IEEE 1547 specifies this value to be less than one-third (1/3) of
year to the present value and summing all loss costs as total load in order to avoid unintended island. In this study, we
have considered that at the maximum, 35% of total active load
N being supplied by SDGs.
LCY
LCT = ∑ Y
. (24)
Y = 1 (1 + d)
4.2.4 SDG power factor limit: The power factor of SDGs’ output
is limited as follows:
3.3 Comprehensive cost function
The comprehensive cost function is the combination of the cost PFSDG ≥ PFMin . (30)
terms introduced above along with the terms in charge with voltage
dip and LDI costs. The planning program is performed for N years. 4.2.5 SDG location: The candidate bus for SDG location can be
The comprehensive function consists of the sensitive load chosen among all system buses except for slack bus. So SDG bus
disruption cost (dip and LDI), network upgrade cost and network tag number shall be a natural number between 2 and Nb
loss cost as
2 ≤ BSDG ≤ Nb & BSDG ∈ ℕ, (31)
Ctotal = α1 × Cdip − T + CLDI − T + α2 × CDef + α3 × LCT,
(25)
where αi ∈ 0, 1 for i = 1, 2 and 3 . where ℕ is supposed to be the set of natural numbers.

4 Optimisation problem formulation 4.3 Solution algorithm

The optimal planning of SDGs is performed to improve network The particle swarm optimisation (PSO) algorithm is applied to
support for sensitive loads and lowering disruption cost by solve the optimisation problem. It starts with a random population
reducing the number of costly dips and LDIs if possible. In matrix X in the search space. In each iteration, the load flow
addition, the formulated problem is also aimed to enhance the program is run and the MMCS is performed for every single
traditional planning objectives including network loss cost and particle of the population. The corresponding cost terms will be
lines’ upgrade cost. The objective of the optimisation problem is calculated to attain the objective function. The problem constraints
the comprehensive cost function presented in (25). Either zero or are verified and if any violation found, those violations are added
one coefficient is assigned to each term of the objective function to the objective function using penalty factors. The local optimal
which enables the user to consider that term separately or (pbest) and global optimal solutions (gbest) will be determined. By
simultaneously with other terms. As all of the described cost terms velocity and position equations of the PSO algorithm, a new
are dependent on SDGs’ active power, reactive power and location, population will be generated and the procedure will be reiterated
the comprehensive cost term is a function of the control variables until the termination condition of the algorithm is satisfied. The
as follows: details of the PSO algorithm are presented in [17] and equations
are as follows:
Minimise Ctotal = ℱ PSDG, QSDG, BSDG , (26)
vik + 1 = ωvik + c1r1 pbestki − xik + c2r2 gbestki − xik , (32)
where PSDG, QSDG and BSDG are the vectors of active power,
reactive power, and buses of SDGs, respectively. xik + 1 = xik + vik + 1, (33)

4.1 Equality constraints where vik is the velocity of the ith particle at kth iteration; xik is the
The active and reactive power balance for every system bus must position of an ith particle at kth iteration; ω is the inertia weight
be satisfied. The equations for active and reactive power balance factor; c1, c2 are cognitive and social coefficients equal to 0.9; r1, r2
can be found in [27] which are not mentioned here for brevity. are random numbers in the range of [0,1]; pbestki is the local best
position of an ith particle at kth iteration; and gbestki is the global
4.2 Inequality constraints
best position of the swarm at kth iteration.
4.2.1 Bus voltage limit: The voltage magnitude must be kept To have a better balance between the global exploration and
within the permissible limit at each bus local exploration, the inertia weight ω is decreased linearly with
the number of iterations [17]. The inertia weight is updated in each
V mini ≤ V i ≤ V maxi, i ∈ Nb, (27) iteration as follows:

where V mini and V maxi refer to the minimum and maximum voltage ωmax − ωmin
ω = ωmax − × k, (34)
limit at bus i. itermax

where ωmax and ωmin are the minimum and maximum values of
4.2.2 Thermal constraint: This is a stand-alone constraint; simply
put, the rated current of lines must not be exceeded inertia weight set to be 0.4 and 0.9, respectively, and itermax is the
maximum iteration number. Fig. 4 shows the overall flowchart of
Ii < Iirated, i ∈ NL, (28) the proposed method.

where Ii and Iirated stand for the current flowing into line i and the 5 System under evaluation
rated current of line i, respectively, and NL is the set of system The proposed methodology was tested on the 33 kV level of the
lines. If any of the lines’ currents are exceeded, they will be IEEE 30-bus meshed distribution system as shown in Fig. 5. The
upgraded. system data can be found in [28]. As there is no data available on
the lines’ length of the system, the system lines are modified and
4.2.3 SDG penetration limit: According to the related standards, the lengths are estimated based on some manufacturers’ data. The
the permissible penetration level of SDGs in the system must be modified lines’ data for 33 kV level of IEEE 30-bus test system is
bound to a proportion of the system load presented in Table 3. The interface transformer of the SDG is a
33/0.4 kV (DYn), uk = 5% transformer. A zigzag-grounding wye
33/0.4 kV transformer with zigzag connection on the primary side
∑ Max
PSDGn ≤ PSDG . (29) (33 kV) and Yn connection on the secondary side (0.4 kV) is
n ∈ NSDG

1860 IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865
© The Institution of Engineering and Technology 2018
installed at bus 8 to provide the path to ground for ground faults. manufacturers such as ABB are used. The load growth rate is
The positive and zero sequence reactances of the zigzag rLG = 3% and the real discount rate is d = 5%.
transformer are assumed to be uk1 = 2% and uk0 = 1%,
respectively. The target loss level is the present loss of the network
6 Results and discussion
equal to 18.52 kW and the bus voltage limits are 0.95 and 1.05 pu.
As mentioned before, the maximum permissible SDG capacity is Three sets of sensitive loads are evaluated to study the effect of
chosen 35% of the total active load and PFmin = 0.85. For the different configurations of sensitive consumers. The first set forms
reactances of different SDG sizes, the data from some a small loop in the system including buses 1, 2 and 3. The second
one forms a big loop including buses 1, 2, 3, 4, 5, 6, 7, 8 and 15
and the last one embraces all system buses. The purpose of
defining two loops inclined toward one side of the network is to
evaluate how dip and LDI costs are influenced by the quantity and
distribution of sensitive loads and whether it leads to variation in
optimised size and location of SDGs or not. The data about
sensitive loads’ type, size and location are input to the proposed
method. Here, to analyse the usefulness of the method, we have
made the following assumption. However, to implement the
method in a real system, the data of real sensitive loads and their
corresponding cost functions shall be considered. It is assumed that
buses with their tag numbers in the form of 3k (a multiplier of
three), 3k + 1 and 3k + 2 are hosting industrial, residential and
commercial loads, respectively. For the dip cost evaluation, 2.31 
£/kW is considered for all consumers and for the interruption cost
evaluation, 0.978, 2.364 and 0.886 £/kWh are considered for
industrial, commercial and residential consumers, respectively
[24]. The planning horizon is assumed to be 20 years and the fault
rate is rf = 1/km/year. An initial study is also performed which
ignores the effect of voltage dip and LDI on SDG planning and
only takes the traditional planning objectives into consideration. In
addition, a scenario with all buses hosting sensitive loads is
performed which evaluates the optimal installation of three SDGs
instead of one SDG. Different cases taken into consideration are
listed in the following:

• Case 1: optimised allocation of one SDG with the objective

function of network upgrade cost (CDef ) and no sensitive load
present.
• Case 2: optimised allocation of one SDG with loss cost (LCT) as
Fig. 4  Overall flowchart of the proposed method
the objective function and no sensitive load present.
• Case 3: optimised allocation of one SDG with both network
upgrade cost and loss cost as the objective function and no
sensitive load present.
• Case 4: optimised allocation of one SDG with dip and LDI costs
as the objective function and sensitive loads assigned to the
small loop.
• Case 5: optimised allocation of one SDG with comprehensive
cost function as the objective and sensitive loads assigned to the
small loop.
• Case 6: optimised allocation of one SDG with voltage dip and
LDI costs as the objective function and sensitive loads assigned
to the big loop.
• Case 7: optimised allocation of one SDG with the
comprehensive cost function as the objective and sensitive loads
assigned to the big loop.
Fig. 5  33-kV level of IEEE-30 bus system

Table 3 Modified lines’ data for 33 kV level of IEEE 30-bus test system
Line Length, km R, Ω X, Ω Line Length, km R, Ω X, Ω
1–2 7.457 2.89 1.11 8–16 4.375 1.69 0.65
2–3 6.811 1.31 0.9 9–16 0.691 0.27 0.1
2–4 5.778 2.24 0.86 3–17 5.919 2.29 0.88
1–3 7.822 3.03 1.17 10–16 5.587 2.16 0.83
4–7 5.234 2.03 0.78 10–17 7.892 3.05 1.18
3–5 6.392 2.47 0.95 10–18 9.962 3.86 1.38
5–6 3.785 1.46 0.56 11–18 12.009 4.65 1.79
6–15 1.996 0.77 0.3 12–18 6.187 2.39 0.92
8–15 6.014 2.33 0.9 12–13 12.339 4.78 1.84
7–8 2.377 0.92 0.35 12–14 17.922 6.94 2.67
8–9 2.979 0.8 0.41 13–14 13.468 5.21 2.01

IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865 1861
© The Institution of Engineering and Technology 2018
• Case 8: optimised allocation of one SDG with voltage dip and number of buses with sensitive loads increases, the cost of dip rises
LDI costs as the objective function and sensitive loads assigned more sharply than the cost of LDI.
to all buses.
• Case 9: optimised allocation of one SDG with comprehensive 6.1 No-sensitive-load scenario
cost function as the objective and sensitive loads assigned to all
buses. In this scenario, the traditional objectives of network upgrade cost
• Case 10: optimised allocation of three SDGs with voltage dip and power loss cost are considered. Table 6 shows the results of
and LDI costs as the objective function and sensitive loads this scenario. The predominant cost of the network is in regard to
assigned to all buses. the network upgrade and the cost of power loss is almost
negligible. The network upgrade cost is alleviated by about £7m in
• Case 11: optimised allocation of three SDGs with
the presence of SDG. For both cases 1 and 3, the capacity of SDG
comprehensive cost function as the objective and sensitive loads
is nearly set at 22 MW and bus 10 is chosen as the optimal location
assigned to all buses.
of SDG. In case 2, the loss level will remain at the target level
during the planning horizon so that the loss cost is almost zero. All
Table 4 shows the coefficients of the objective function for
costs are in million pounds.
different cases in every scenario. Table 5 depicts different expected
costs of the system without SDG during the next 20 years. As seen
from the table, the major cost of the network is due to the lines 6.2 Sensitive-load scenarios
upgrade regarding the load growth. Moreover, it seems that as the The results of sensitive-load scenarios are presented in Table 7.
They include small-loop, big-loop and all-bus scenarios. All-bus
scenario consists of two sub-scenarios including one SDG
installation and three SDGs installation.
Table 4 Objective function coefficients for different case of
SDG planning
6.2.1 One SDG allocation: In case 4, bus 1 is chosen as the
Scenario Case α1 α2 α3
optimal location which reduces the cost of dip for £0.36 m. In case
no SDG 0 — — — 5, although the network upgrade cost term determines bus 10 as the
no sensitive load (traditional) 1 0 1 0 optimal location of SDG, but the capacity of the resource is
2 0 0 1 increased due to the disruption cost term of the objective function
3 0 1 1 and the optimal capacity is set to 26.5 MW. Moreover, the
small-loop 4 1 0 0 sensitivity of CDef to control variables is more than Cdip − T. That is
5 1 1 1 why CDef has more parts in determining the optimal solution. LDI
big-loop 6 1 0 0 cost does not alter significantly. This was expected. The probability
of LDI depends only on T i and with a definite-time protection
7 1 1 1
model, SDG has no effect on protection time. In case 6, the SDG
all-bus (one SDG) 8 1 0 0
capacity nearly hits the maximum permissible capacity and bus 6 is
9 1 1 1 chosen as the optimal location. For case 7, again the optimal bus is
all-bus (three SDGs) 10 1 0 0 10 which is forced by the term of network upgrade cost. The term
11 1 1 1 of dip cost has made the program to increase the capacity of SDG
by 6 MW relative to case 3 which considers only the traditional

Table 5 Costs of loss, feeder upgrade, dip and LDI without SDG in million pounds
Scenario LIT CDef Cdip − T CLDI − T CTotal
small-loop 0.0145 13.316 2.164 1.557 17.052
big-loop 0.0145 13.316 5.247 3.43 22.008
all-bus 0.0145 13.316 9.678 5.756 28.765

Table 6 Results of no-sensitive-load scenario

Case PSDG, MW QSDG, MVA BSDG CDef (106 £) LIT (106 £) CTotal (106 £)
1 22.07 12.01 10 5.040 0.0107 5.0507
2 35.76 22.16 16 8.820 0.000037 8.820
3 22.86 12.28 10 5.055 0.01059 5.0656

Table 7 Results of sensitive-load scenarios

Case PSDG, MW QSDG, MVA BSDG CDef (106 £) LIT (106 £) Cdip − T (106 £) CLDI − T (106 £) CTotal (106 £)
4 20.32 10.80 1 9.375 0.0230 1.804 1.563 12.765
5 26.50 6.085 10 5.186 0.0119 2.093 1.552 8.843
6 35.98 22.08 6 8.164 0.0127 3.956 3.455 15.588
7 28.51 2.67 10 5.281 0.0125 4.848 3.44 13.582
8 26.68 11.51 15 10.70 0.0053 8.502 5.745 24.952
9 28.39 11.14 10 5.427 0.0109 8.685 5.761 19.884
— 24.4 1.75 9 — — — — —
10 8.50 −0.02 13 8.340 −0.00273 8.477 5.750 22.564
— 1.90 0.58 6 — — — — —
— 15.10 0.28 10 — — — — —
11 11.86 −0.71 6 5.111 −0.00273 8.513 5.748 19.369
— 7.65 0.53 5 — — — — —

1862 IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865
© The Institution of Engineering and Technology 2018
 Table 8 Dip and LDI evaluation for VI relay characteristic
Case Cdip − T (106 £) CLDI − T (106 £) Number of
Dips LDIs
no SDG 18.619 8.375 1496 131
case 9 16.64 8.6 1323 138

remained SDG capacity is allocated to buses 6 and 5 which are

adjacent to the optimal location of case 8. It is worth mentioning
that by applying three SDGs, the contribution of SDGs in reactive
power mitigation is reduced and more capacity is transformed to
useful power.
Fig. 6  Costs of dip and LDI for different cases
6.3 Overall discussion
The dip and LDI costs for studied cases are depicted in Fig. 6. In
different cases, the presence of SDGs has improved the system
performance from voltage dip standpoint and reduced the cost
imposed on the system by sensitive loads disruption. LDI cost is
not affected by SDGs. This is because of the fact that SDGs do not
alter the protection system operation time. Three SDGs cases
reduce the network expenditure a little bit more than one SDG
cases.
Another interesting assessment could be the effect of SDGs on
the total number of disruptions caused by dips and LDIs. Fig. 7
depicts the number of dips and LDIs for different cases during the
planning horizon. The number of dips is reduced significantly in
the presence of SDGs while the number of LDIs remains
Fig. 7  Number of dips and LDIs for different cases with and without SDG unchanged.
With a definite time protection model, LDI is not affected by
planning objectives. LDI cost is not affected in this scenario as the presence of SDGs. This is because of the fact that PLDI is only
well. In all-bus scenario, all system buses are considered except for dependent on T i and SDG installation does not alter T i. The inverse
buses 12, 16 and 18 which have no load. For case 8, bus 15 and for characteristic of overcurrent relays and protection scheme and
case 9, bus 10 is chosen as the optimal location. In case 8, the coordination might be considered in future works. However, here a
objective is to improve the system sensitive loads performance. raw evaluation is performed by employing very inverse (VI)
About 73% of the system loads are assigned to the big loop and bus characteristic of IEC 60225 as in (35) for case 9 and no SDG case
9. Therefore, the choice of bus 15 by the optimal program seems with sensitive equipment assumed on all buses. The cost of LDI
obviously reasonable. In case 9, the optimal capacity of SDG is set and dip for VI relay scenario is shown in Table 8
to 28.39 which demonstrate an 8 MW raise relative to case 3. As is (Is = 320 A&TMS = 0.05 s)
perceived, though the dip cost term does not affect the SDG
location, but it has increased the optimised capacity of SDG for all 13.5
scenarios. All network cost terms are reduced in the presence of t = TMS × ,
Ir − 1
one optimal SDG.
where Ir = I/Is, Is is the relay setting current, and TMS is the time
6.2.2 Three SDGs allocation: Cases 10 and 11 represent the
multiplier setting.
results for optimal allocation of three SDGs with the coefficients of
According to the table, the number and cost of LDIs have
cost function equal to the figures of Table 4. As shown in Table 7,
increased in the presence of SDG. This is because of the fact that
the cost of dip is improved by £0.025 m which is not remarkable in
SDG raises fault current which leads to decrease in protection time.
comparison with the same case for one SDG allocation (case 8).
By a lower T i relative to no SDG case, the probability of LDI
However, the interesting thing is that the cost of network upgrade
is reduced by £2.36 m, although network upgrade cost (CDef ) is not increases. However, the total disruption cost in the presence of
SDG has declined again.
the objective for this case. Moreover, the loss level for both cases
The SDG effect on other cost terms is also assessed to give a
10 and 11 is at its minimum level and the network is profiting from
visual proof of the improvement brought by SDGs. The different
its lowered loss level in both cases. For all-bus scenario, in case 11
cost terms for cases 5, 7, 9 and 11 are compared with the related no
the network experience the lowest cost level during the planning
SDG cases. As shown in Fig. 8, the presence of SDGs leads to the
horizon. However, the improvement brought by three SDGs case of
reduction in all costs except for LDI and improves the network
11 is only £0.52 m more than the related one SDG case of 9. In
performance. The main improvement occurs in the network
cases 10 and 11, the reactive power supplied by SDGs has been
upgrade cost and the location of SDGs is totally determined by the
reduced significantly by dividing the SDG capacity into three
upgrade cost term of the objective function. The value of loss cost
shares which are installed in three different locations. The total
is negligible in proportion to other cost terms. However, the
active capacity of SDG installed in the network has been increased
penalised cost of loss could reach zero in the presence of SDG in
in comparison with one SDG case and the total SDG capacity
case 2. The interesting fact is that network is profiting from its
almost hits the maximum permissible level. In case 10, the main
reduced loss in case 11. The improvement in sensitive loads
share of SDG is allocated to bus 9 which is nearly at the same
performance takes place only from voltage dip standpoint and LDI
electrical distance from buses 15 and 10 which are optimal
cost is not influenced by the presence of SDGs. Generally, the total
locations for one SDG allocation. The second big portion of SDG
cost of the network is reduced significantly in the presence of
is allocated to bus 13. Probably the excessive capacity allocated to
optimal SDGs. The lowest cost for the all-bus scenario is achieved
bus 13 is to supply and support the large loads in buses 13 and 14
by the allocation of three SDGs in case 11. However, case 9 also
and to satisfy the optimisation constraints. In three SDGs case 10,
shows a good performance.
all network costs including voltage dip, network upgrade and
Fig. 9 shows the reduction in different cost terms is million
power loss are improved relative to one SDG case 8. In case 11, the
main capacity is allocated to bus 10 as the same as case 9. The pounds (106 £) per total MVA of SDGs installed for different
optimisation cases. As perceived from the figure, the most
IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865 1863
© The Institution of Engineering and Technology 2018
Fig. 8  Different cost terms for small-loop, big-loop, and all-bus scenarios

SDGs has almost remained constant. In fact, the big-loop scenario

is the most sensitive scenario to dip cost and considering a dip in
planning matters most for this scenario. The other interesting fact is
that with cost reduction per MVA of SDGs as a criterion, three
SDGs cases show less total improvement than one SDG cases,
although the total network cost for three SDGs cases is less. This is
because of the rise in SDG capacity for three SDGs cases.
A technical factor which may represent dip performance is the
average dip magnitude during the simulations. However, according
to [19], a trip number due to dip is a better technical representative
of network performance from power quality standpoint. Fig. 10
depicts the average dip magnitude of every sensitive busbar during
the MMCS for all-bus scenarios including cases 8, 9, 10, 11. No-
SDG case is also shown to give a sense of the improvement
Fig. 9  Reduction in cost terms per total MVA of SDG installed
brought by the presence of SDGs. The voltage of buses 12, 16 and
18 which have no load is set to zero. As depicted by the figure,
some buses such as bus 6 and bus 10, show an improvement of
>7% in their average dip magnitude in the presence of SDGs. All
voltages are improved by SDGs. Buses which are closer to the
slack bus do not observe any remarkable change in their average
voltage dip magnitude as they are still supplied by the upstream
network. The buses where SDGs are installed show the most
improvement in dip magnitude. Although the improvements
brought by SDG cases are obvious with regard to no-SDG case,
average dip magnitude does not present a proper criterion to
compare different SDG cases with each other.
Fig. 10  Average voltage dip magnitude of sensitive buses for SDG cases 8,
9, 10, 11 and no SDG case 6.4 MMCS convergence
The convergence of the MMCS method depends on the number of
simulated faults in each iteration. For case 4, the annual costs of
dip and LDI versus the number of faults are presented in Fig. 11.
As seen from the figure, for the number of faults >7000, the
convergence is almost obtained. The number of faults in this study
is 8000.

7 Conclusion
This study proposed a Monte-Carlo based methodology to consider
sensitive loads disruption caused by dips and LDIs in SDGs
planning programs. Additionally, the traditional planning
objectives including network upgrade cost and power loss cost are
also taken into consideration. The results illustrate that the dip cost
Fig. 11  Convergence of MMCS in annual dip and LDI cost calculation term could affect the optimal solution and is improved in the
presence of SDGs. It conveys such an economic burden which
improvement per total MVA of SDGs is assigned to network could not be neglected in the planning of networks hosting
investment upgrade cost. The dip cost is ranked at the second sensitive loads. The cost term for LDI is not affected by the
place. The improvement in loss per MVA of SDGs is presence of SDGs for a definite time protection model. However,
inconsiderable and no improvement is observed for LDI cost. The the cost and number of LDIs could be aggravated for an inverse
interesting fact is that the sensitivity of dip cost to SDGs’ location time characteristic of the protection system. The total cost of
has reduced in an all-bus scenario with regard to cases 8 and 9. sensitive loads disruption is mitigated by the presence of SDGs.
Even in three SDGs cases, the dip cost reduction per MVA of Three SDGs allocation shows a mild improvement relative to one

1864 IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865
© The Institution of Engineering and Technology 2018
SDG allocation and more capacity is transformed to sellable useful [13] Muñoz-Delgado, G., Contreras, J., Arroyo, J.M.: ‘Multistage generation and
network expansion planning in distribution systems considering uncertainty
power. However, by the criterion of cost reduction per MVA of and reliability’, IEEE Trans. Power Syst., 2016, 31, (5), pp. 3715–3728
SDGs installed, one SDG scenario is more economical than three [14] Muñoz-Delgado, G., Contreras, J., Arroyo, J.M.: ‘Joint expansion planning of
SDGs scenario. distributed generation and distribution networks’, IEEE Trans. Power Syst.,
Finally, this study tried to slightly make up the lack of power 2016, 30, (5), pp. 2579–2590
[15] Abu-Mouti, F.S., El-Hawary, M.E.: ‘Optimal distributed generation allocation
quality in economic DG planning programs. The authors plan to and sizing in distribution systems via artificial bee colony algorithm’, IEEE
contribute the protection system coordination and process Trans. Power Deliv., 2011, 26, (4), pp. 2090–2101
sensitivity in the SDG planning program and employ methods to [16] Keane, A., O'Malley, M.: ‘Optimal allocation of embedded generation on
reduce the calculation burden in future works. distribution networks’, IEEE Trans. Power. Syst., 2005, 20, (3), pp. 1640–
1646
[17] Pandi, V.R., Zeineldin, H.H., Xiao, W.: ‘Determining optimal location and
8 References size of distributed generation resources considering harmonic and protection
coordination limits’, IEEE Trans. Power. Syst., 2013, 28, (2), pp. 1245–1254
[1] Piccolo, A., Siano, P.: ‘Evaluating the impact of network investment deferral [18] Daud, S., Kadir, A.F.A., Gan, C.K., et al.: ‘A comparison of heuristic
on distributed generation expansion’, IEEE Trans. Power Syst., 2009, 24, (3), optimization techniques for optimal placement and sizing of photovoltaic
pp. 1559–1567 based distributed generation in a distribution system’, Sol. Energy, 2016, 140,
[2] Harrison, G.P., Piccolo, A., Siano, P., et al.: ‘Exploring the tradeoffs between pp. 219–226 Volume
incentives for distributed generation developers and DNOs’, IEEE Trans. [19] Eslami, A., Golshan, M.E.H.: ‘Index-based voltage dip consideration in
Power Syst., 2007, 22, (2), pp. 821–828 optimal planning of SDGs by applying a modified Monte Carlo simulation
[3] Viral, R., Khatod, D.K.: ‘Optimal planning of distributed generation systems method’, Int. Trans. Electr. Energy Syst., 2017, 28, (1), pp. 1–17
in distribution system: a review’, Renew. Sustain. Energy Rev., 2012, 16, (7), [20] Biswas, S., Goswami, S.K., Chatterjee, A.: ‘Optimal distributed generation
pp. 5146–5165 placement in shunt capacitor compensated distribution systems considering
[4] Wang, D.T.C., Ochoa, L.F., Harrison, G.P.: ‘DG impact on investment voltage sag and harmonics distortions’, IET Gener. Transm. Distrib., 2014, 8,
deferral: network planning and security of supply’, IEEE Trans. Power Syst., (5), pp. 783–797
2010, 25, (2), pp. 1134–1141 [21] Biswas, S., Goswami, S.K., Chatterjee, A.: ‘Optimum distributed generation
[5] Celli, G., Ghiani, E., Mocci, S., et al.: ‘A multiobjective evolutionary placement with voltage sag effect minimization’, Energy Convers. Manage.,
algorithm for the sizing and siting of distributed generation’, IEEE Trans. 2012, 53, (1), pp. 163–174
Power Syst., 2005, 20, (2), pp. 750–757 [22] Milanovic, J.V., Gupta, C.P.: ‘Probabilistic assessment of financial losses due
[6] Singh, D., Singh, D., Verma, K.S.: ‘Multiobjective optimization for DG to interruptions and voltage sags-part I: the methodology’, IEEE Trans. Power
planning with load models’, IEEE Trans. Power Syst., 2009, 24, (1), pp. 427– Deliv., 2006, 21, (2), pp. 918–924
436 [23] Cebrian, J.C., Milanović, J.V., Kagan, N.: ‘Probabilistic assessment of
[7] Khanbabapour, S., Golshan, M.E.H: ‘Synchronous DG planning for financial losses in distribution network Due to fault-induced process
simultaneous improvement of technical, overcurrent and timely anti-islanding interruptions considering process immunity time’, IEEE Trans. Power Deliv.,
protection indices of the network to preserve protection coordination’, IEEE 2015, 30, (3), pp. 1478–1486
Trans. Power Deliv., 2017, 32, (1), no. 99, pp. 474–483 [24] Cebrian, J.C., Kagan, N.: ‘Hybrid method to assess sensitive process
[8] Abri, R.S.A., El-Saadany, E.F., Atwa, Y.M.: ‘Optimal placement and sizing interruption costs due to faults in electric power distribution networks’, IEEE
method to improve the voltage stability margin in a distribution system using Trans. Power Deliv., 2010, 25, (3), pp. 1686–1696
distributed generation’, IEEE Trans. Power Syst., 2013, 28, (1), pp. 326–334 [25] Faried, S.O., Billinton, R., Aboreshaid, S.: ‘Stochastic evaluation of voltage
[9] Mitra, J., Vallem, M.R., Singh, C.: ‘Optimal deployment of distributed sag and unbalance in transmission systems’, IEEE Trans. Power Deliv., 2005,
generation using a reliability criterion’, IEEE Trans Ind. Appl., 2016, 52, (3), 20, (4), pp. 2631–2637
pp. 1989–1997 [26] Aung, M.T., Milanovic, J.V.: ‘Stochastic prediction of voltage sags by
[10] Akorede, M.F., Hizam, H., Aris, I., et al.: ‘Effective method for optimal considering the probability of the failure of the protection system’, IEEE
allocation of distributed generation units in meshed electric power systems’, Trans Power Deliv., 2006, 21, (1), pp. 322–329
IET Gener. Transm. Distrib., 2011, 5, (2), pp. 276–287 [27] Stevenson, W.D., Grainger, J.J.: ‘Power system analysis’ (McGraw-Hill, New
[11] Abbasi, F., Hosseini, S.M.: ‘Optimal DG allocation and sizing in presence of York, 1994)
storage systems considering network configuration effects in distribution [28] ‘Power systems test case archive’, available at https://
systems’, IET Gener. Transm. Distrib., 2016, 10, (3), pp. 617–624 www2.ee.washington.edu/research/pstca/pf30/pg_tca30bus.htm/, accessed 14
[12] Tan, W.S., Hassan, M.Y., Rahman, H.A., et al.: ‘Multi-distributed generation May 2017
planning using hybrid particle swarm optimisation-gravitational search
algorithm including voltage rise issue’, IET Gener. Transm. Distrib., 2013, 7,
(9), pp. 929–942

IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 8, pp. 1856-1865 1865
© The Institution of Engineering and Technology 2018
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS

Enhancing the performance of

Distribution systems using PSO

M. Ramalinga Reddy 1a*, U. Kamal Kumar 2b,

Suresh Babu Daram 3c
a
PG Student,EEE, Sree Vidyanikethan Engineering College, Tirupati
517102, India
b
Assistant Professor, EEE, Sree Vidyanikethan Engineering College,
Tirupati 517102, India
c
Associate Professor, EEE, Sree Vidyanikethan Engineering College,
Tirupati 517102, India

 M.Ramalinga Reddy. Tel.:+91 9849918850;

email:0117ramalinga@gmail.com

Abstract:
The electricity demand is increasing day by day as well that the
distribution network configuration varies simultaneously. The rapid
uncertainties at generation to distribution networks will impacts on
the consumer devices. In this the identification of uncertainty at
network location and mitigating with respective compensating
devices are passionately done by many researcher, This paper
proposes that the integration of optimizing techniques into the power
system will solve the problems at utilities and end user side. Particle
Swarm Optimization have been proposed to analyze the distribution
network in all aspects like voltage deviations, power losses and
power quality issues and their mitigations by optimal placing and
sizing of distribution generation(DG).

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
Key Words: Distribution System, Distributed Generation,
Optimization Technique.
1. INTRODUCTION

Nowadays, Distribution systems are facing new issues like as

sensitive loads, economic concerns and environmental effects
which need special consideration.
Increase in load demand in customer side leads to
development in distribution network, this is additional leads to
increase in power loss, voltage deviations, decrease in bus
voltage, increase in load imbalance. There are many methods
of supporting voltage and loss reduction techniques like
capacitor placement, network reconfiguration, high voltage
distribution system, conductor grading and DG unit
placement. Distribution generation units are improves the
system performance in all aspects The placement of DG at
near to the is acquiring to specific issue that can be identified
by the following types. They are
Type 1: DG is capable of supplying only real power
Type 2: DG is capable of supplying only reactive power
Type 3: DG is capable of supplying real power and absorbing
reactive power
Type 4: DG is capable of supplying of both real and reactive
power
The placement and size of DGs calculation is another
challenging job. There are several optimization algorithms are
exacting to placement and sizing of DGs.

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
2. LITERATURE SURVEY

So, many authors can do research on placement and sizing of

DGs to reduce power loss and improve voltage profile. There
are some methods proposed to optimization problem solve in
terms of optimal location and sizing of DG, which both can be
conventional or stochastic search algorithms.
The authors A. K. Saonerkari, and B. Y. Bagde proposed the
optimal placement of distribution generation using Genetic
Algorithm [1]. In this paper capacitor banks placement along
with DGs proposed. The authors Adel A. Abou El-Ela, Ragab
A. El-Sehiemy , and Ahmed Samir Abbas proposed the
Optimal Placement and Size of Distributed Generation and
Capacitor Banks in Distribution Systems Using Water Cycle
Algorithm [2]. Authors P V.C. Dinakara Prasasd Reddy, Dr. T.
Veera Reddy and Dr. Gowri Manohar proposed Optimal
renewable resources placement in distribution networks by
using Whale optimization algorithms [3]. The authors Mukul
Dixit, Prasanta Kundu and Hitesh R. Jariwala proposed
optimal placement and sizing of Distribution generation by
using Artificial Bee Colony Algorithm [4]. C. Hari Prasad, K.
Subbaramaiah and P. Sujatha proposed placement of DG by
using elephant herding optimization algorithm [5]. The authors
analysis cost at the time of DG placement. The authors John
Edwin Candelo-Becerra and Helman Enrique Hernández
Riano proposed placement of DG in radial distribution system
using bat-inspired algorithm [6]. The author Vasiliki Vita

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
proposed placement and size of DG unit in radial distribution
system by using Decision-Making Algorithm [7]. Mr.
Rayabarapu Rakesh, Dr.P. VenkataPapana and Mrs. Sravani
Keerthi proposed location and size of DG by using Hybrid
algorithmusing Power Loss Index (PLI) & Genetic Algorithm
(GA) [8]. Sayyad Nojavan, Mehdi Jalali and Kazem Zare
proposed a placement of DG units and capacitors by using
mixed integer nonlinear programming (MINLP) approach [9].
Authors Bikash Das and V. Mukherjee Debapriya Das
proposed real power loss reduction by placement of DG using
symbiotic organism search algorithm [10]. Abdel-Rahman
Sobieh1, M. Mandour, Ebtisam M. Saied, and M.M. Salama
proposed a location and size of DG in distribution system
using Grey Wolf Optimizer (GWO) [11].
Many researchers worked on this placement of DG and
proposed so many algorithms those algorithms take more
time for solving and it becomes more complex.

3. PARTICLE SWARM OPTIMIZATION (PSO)

Particle swarm optimization (PSO) algorithm is inspired by
nature. PSO algorithm is one of the optimization technique,
problem solving based on movement and intelligence of
swarms. PSO consists of a swarm (group) of particles
(individuals) moving in the search space looking for the best
solution. For each particle, proposed objective function is

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
calculated by its position and each particle‟s position &
velocity can be updated by using following equations
( ) ( )

where, c1, c2: The weighting factor, r1 , r2 : The random

numbers between 0 and 1, w : The weighting function, vik :
The current velocity of particle i at iteration k, vik+1 : The
modified velocity of particle i, sik : The current position of
particle i at iteration k, sik+1 : The modified position of particle i,
pbesti : The personal best of particle i, gbest : The global best of
the group.
3.1 PSO Algorithm
Calculation of optimal size & placement of DG using PSO
algorithm has been used following steps:
Step 1: Give Input line & bus data and voltage
Step 2: Calculate node voltages and total power loss in radial
distribution network using forward backward sweep method
Step 3: Randomly generates an initial population
Step 4: Initialize number of particles to be optimized
Step 5: Set bus count x=2
Step 6: Set iteration count y = 0
Step 7: Generate random position and velocity for each
particle
Step 8: Calculate total power loss at each particle
Step 9: Initialize current position of each particle as „Pbest‟
Step 10: Assign „Gbest‟ as best amount „Pbest‟

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
Step 11: Update position and velocity of each particle
Step 12: If generation count reaches maximum limit, go to
Step 13 or else increase the counter by one and go to Step 7
Step 13: If bus count reaches maximum limit, go to Step 14 or
else increase the counter by one and go to Step 6
Step 14: Display the results
The advantages of Particle Swarm Optimization over other
conventional techniques:
 PSO is easy to implement
 It does not require the good initial solution to start its
iteration process.
 PSO is faster to reach optimal solution

4. TEST SYSTEM AND RESULTS

To evaluate the performance of PSO algorithm in the

application of DG planning problem, the IEEE 15-bus radial
distribution system is simulated using MATLAB 2014a
software. The Particle Swarm Optimization algorithm is used
to find the placement and sizing of DG for the minimization of
total power loss and improve voltage profile of the system.

4.1 IEEE 15-bus distribution system

The single line diagram of IEEE 15-bus system is shown in

Figure 1. The system voltage is 11 kV and total system active

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
and reactive loads are 1401.6 kW and 1051.2 kVAR. This test
system consists of 15 buses and 14 branches.

. Fig. 1 IEEE 15 – Bus Single line diagram

4.2 Enhancing the Voltage Profile
The bus voltage values of IEEE 15 bus radial distribution
system without and with DG unit using PSO shown in table 1.
The bus 1 has maximum voltage of 1pu. The weakest voltage
level at bus 13 with 0.9446pu, then after placement of DG that
voltage level improves to 0.9591pu. The life time of different
customer loads are increased by voltage profile improved.

Without DG With DG

1.05

0.95

0.9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Fig.2 Bus Voltages without and with DG of Test system

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
4.3 Total Power loss reduction
The total real and reactive power loss of the IEEE 15 bus
system is 61.76 kW and 57.27 kVAR without connecting DG.
The total active and reactive power loss of 15 bus system
after connection of DG is 44.59 kW and 48.03 kVAR.
Placement of DG is cost effective method and future
expansion is also possible to reduce power loss at load
centres.

5. ACKNOWLEDGEMENTS

The author would like to thank Mr. U. KAMAL KUMAR,

M.Tech., (Ph.D.) Assistant Professor, Department of EEE for
his valuable guidance. And Sree Vidyanikethan Engineering
College, Tirupati.

6. CONCLUSION

In this paper, a particle swarm optimization for optimal

placement of DG is minimizing the total real power loss. This
method is fast and accurate in determining the sizes and
locations. The voltage profile of the IEEE 15 bus systems is
enhanced by reducing the power losses by installing DG.

Bibliography:
[1] A. K. Saonerkari and B. Y. Bagde, “Optimized DG

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
Placement in Radial Distribution System with
Reconfiguration and Capacitor Placement Using Genetic
Algorithm”, IEEE International Conference on Advanced
Communication Control and Computing Technologies,
2014.
[2] Adel A. Abou El-Ela, Ragab A. El-Sehiemy , and Ahmed
Samir Abbas, “Optimal Placement and Sizing of Distributed
Generation and Capacitor Banks in Distribution Systems
Using Water Cycle Algorithm”, IEEE Systems Journal,
Volume: 12, Issue:. 4, pp: 3629-3636, December 2018.
[3] P V.C. Dinakara Prasasd Reddy, Dr. T. Veera Reddy, and
Dr. Gowri Manohar, “Optimal renewable resources
placement in distribution networks by combined power loss
index and Whale optimization algorithms”, Journal of
electrical systems and information technology, March 2017.
[4] Mukul Dixit, Prasanta Kundu and Hitesh R. Jariwala,
“Optimal Placement and Sizing of DG in Distribution
System using Artificial Bee Colony Algorithm”, IEEE, 2016.
[5] C. Hari Prasad, K. Subbaramaiah and P. Sujatha, “Cost–
benefit analysis for optimal DG placement in distribution
systems by using elephant herding optimization algorithm”,
Springer open, 2019.
[6] John Edwin Candelo-Becerra and Helman Enrique
Hernández Riano, “Distributed generation placement in
radial distribution networks using a bat-inspired algorithm”,
DYNA pp: 60-67. August, 2015.
[7] Vasiliki Vita, “Development of a Decision-Making Algorithm
for the Optimum Size and Placement of Distributed
Generation Units in Distribution Networks”, Energies, 2017.

Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
[8] Mr. Rayabarapu Rakesh, Dr.P. VenkataPapana and Mrs.
Sravani Keerthi, “A Hybrid Algorithm for Optimal allocation
of DG in Radial Distribution System”, IEEE, 2017.
[9] Sayyad Nojavan, Mehdi Jalali and Kazem Zare, “Loss
Reduction in Distribution System Using Simultaneous
Allocation of DG Units and Capacitor”, Research gate,
May2014.
[10] Bikash Das and V. Mukherjee Debapriya Das, “DG
placement in radial distribution network by symbiotic
organism search algorithm for real power loss
minimization”, Applied soft computing, 2016.
[11] Abdel-Rahman Sobieh1, M. Mandour, Ebtisam M. Saied,
and M.M. Salama, “Optimal Number Size and Location of
Distributed Generation Units in Radial Distribution Systems
Using Grey Wolf Optimizer”, International Electrical
Engineering Journal (IEEJ), Volume: 7, Issue: 9, pp: 2367-
2376, 2017, ISSN 2078-2365.
[12] Krischonme Bhumkittipich and Weerachai
Phuangpornpitak, “Optimal Placement and Sizing of
Distributed Generation for Power Loss Reduction using
Particle Swarm Optimization”, Elsevier Energy Procedia,
Volume: 34, pp: 307 – 317, 2013.
[13] Wichit Krueasuk and Weerakorn Ongsakul, “Optimal
Placement of Distributed Generation Using Particle Swarm
Optimization”, Research gate, 2014.
[14] K. Ras Guerriche and T. Bouktir, “Optimal Allocation and
Sizing of Distributed Generation with Particle Swarm
Optimization Algorithm for Loss Reduction”, Revue des
Sciences et de la Technologie, Volume 6, pp: 59-69, 2015.

1
0
Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 REACHING THE HEIGHTS OF TECHNOLOGY
TRENDS
[15] Naveen Jain, S.N. Singh and S.C. Srivastava, “Particle
Swarm Optimization Based Method for Optimal Siting and
th
Sizing of Multiple Distributed Generators”, 16 National
Power Systems Conference, pp: 669-674, December 2010.

1
1
Accepted for publication through e-techno symposium conducted by HPFG on 25 & 26 July 2020
 BIODATA

NAME : MUCHUGUTTA RAMALINGA REDDY

FATHER NAME : M. NARASIMHA REDDY

DATE OF BIRTH : 17-01-1996

NATIONALITY : INDIAN

CONTACT DETAILS

Contact No. : 9849918850

Email : 0117ramalinga@gmail.com

Contact Address : 12-4, Panjurivari palli (Village), Renimakula palli (Post),

Thamballa palli (M), Chittoor (D), AP.