ABSTRACT
This research work presents development of Firefly algorithm (FA) and application of voltage
stability index (VSI) for optimal planning of Distribution Generation (DG) in an unbalanced threephase
distribution network. The VSI was used to find the DG location while the FA was used for
the DG sizing. The developed method was implemented on standard IEEE 37-bus Radial
distribution network test system and a local 19-bus Mahuta feeder. The results obtained from the
IEEE 37-bus were validated by comparing with a similar work. For the standard IEEE 37-bus
unbalanced radial distribution network (URDN), the total power loss obtained are 31.3543 kW and
15.2829 kVAr for active and reactive power respectively without DG in the network. When the
developed method is applied, a DG optimal location was found at bus 34 and a DG size of 356 kW
and 170 kVAr for active and reactive respectively are obtained. The active and reactive power loss
were found to be 19.8329 kW and 10.0014 kVAr respectively. The developed method recorded a
loss reduction of 36.75% and 34.56% for both active and reactive power respectively over the base
case. Also, the maximum loadability of the network was found to be 18% and 8% of the initial
loading with and without DG respectively. For the 19-bus Mahuta feeder, the location for the DG
is bus 17 and the DG size of 201.58 kW and 115 kVAr are obtained for active and reactive power
respectively. A power loss reduction of 4.48% and 5.62% for active and reactive power were
recorded over the base case respectively. The maximum loadability of the network for both the
developed method and base case were found to be 119% of the initial loading. When compared
with research on similar work, the developed method achieved a loss reduction of 8.14% and
30.42% for active and reactive power respectively over the method applied in the work.
TABLE OF CONTENTS
TITLE PAGE
DECLARATION i
CERTIFICATION ii
DEDICATION iii
ACKNOWLEDGEMENT iv
ABSTRACT vi
TABLE OF CONTENTS vii
LIST OF TABLES ix
LIST OF FIGURES x
LIST OF APPENDICES xi
LIST OF ABBREVIATIONS xii
CHAPTER ONE: INTRODUCTION
1.1 Background of Study 1
1.2 Significance of Research 4
1.3 Statement of Problem 5
1.4 Aim and Objectives 5
1.5 Methodology 6
1.6 Dissertation Organization 7
CHAPTER TWO: LITERATURE REVIEW
2.1 Introduction 8
2.2 Review of Fundamental Concepts 8
2.2.1 Power System Background 8
2.2.2 Distribution System 9
2.2.3 Three-Phase Distribution Network 9
2.2.4 Bus Classification 13
2.2.5 Distribution Generation 15
2.2.6 Power Flow (Load Flow) 20
viii
2.2.7 Load Models 25
` 2.2.8 Objective Function 27
2.2.9 Methods for Optimal DG Placement and Sizing 29
2.2.10 Sensitivty Based Method 31
2.2.11 Voltage Stability Index 31
2.2.12 Firefly Algorithm 35
2.2.13 Radial Distribution Test System 39
2.3 Review of Similar Works 43
CHAPTER THREE: MATERIALS AND METHODS
3.1 Introduction 50
3.2 Materials 50
3.2.1 Personal computer 50
3.2.2 MATLAB 2013a software 50
3.2.3 Distribution network parameters 50
3.3 Methodology 51
3.3.1 Development of three-phase power flow based on backward and forward sweep
technique 51
3.3.4 Voltage Stability Index (VSI) application 52
3.3.3 Model of the Distributed Generation Planning 53
3.3.4 Development of the firefly algorithm for optimal DG sizing 54
3.3.5 Developed method for DG planning 55
3.3.6 The Network Maximum Loadability 58
3.3.6 Test Systems 58
3.3.7 Performance Evaluation
CHAPTER FOUR: RESULTS AND DISCUSSIONS
4.1 Introduction 59
4.2 The IEEE 37-Bus Unbalanced Radial Distribution Test System 59
4.2.1 Case I: Network without DG (Base Case) 59
4.2.2 Network with DG 60
ix
4.2.3 Network Voltage Profile with and without DG 62
4.2.4 IEEE 37-Bus Loadability 64
4.3 19-Bus Mahuta Radial Distribution Network Feeder 67
4.3.1 Case I: Network without DG (Base Case) 67
4.3.2 Case II: Network with DG 68
4.3.3 19-Bus Mahuta Feeder Network Voltage Profile 69
4.3.4 19-Bus Mahuta Feeder Loadability 71
4.4 Validation of the Optimized Method 73
CHAPTER FIVE: CONCLUSION AND RECOMMENDATIONS
5.1 Summary 76
5.2 Conclusion 76
5.3 Significant Contribution 78
5.4 Limitations 78
5.5 Recommendations 78
REFERENCES 79
APPENDIX 83
CHAPTER ONE
INTRODUCTION
1.1 Background of Study
The electric power system majorly includes a generating plant, a transmission system and the
distribution network (Subramanyam et al., 2015). Modern power systems are evolving from the
centralized bulk systems, with generation plants connected to the transmission network, to more
decentralized systems, with smaller generating units connected directly to distribution networks
close to demand site. The distribution network is mainly a passive network where the flow of both
real and reactive power is unidirectional (Satish & Navuri, 2012). However, with significant
penetration of distributed generation, the power flow may become reversed, hence the distribution
network is no longer a passive system but an active system. In this active system, the power flows
and voltages are determined by the topology of the network generation sources as well as the loads
(Mahmud et al., 2011). Distributed Generation (DG) also termed as embedded generation,
dispersed generation or decentralized generation is defined as a small electric power source that
can be connected to a distribution network by a distribution company (DISCO) at any node or by
customer at the customer side of the meter (Payasi et al., 2012). DG, unlike conventional
generation, aims to generate part of required electrical energy on small scale, closer to the area of
consumption, and also to augment the electrical power from the grid within the network. It
represents a change in the conceptual framework of electrical energy generation. DG can be an
alternative for residential, commercial, and industrial applications (Murthy & Kumar, 2013).
Electrical Distribution Systems (EDS) are expected to experience considerable growth in the near
future, with respect to the penetration of DG. This will be mainly due to several factors, ranging
2
from environmental concerns to new technologies such as fuel cells and other alternative energy
sources. In spite of the additional complexity in DS planning and operation in the presence of DG,
it is of paramount importance that the performance of these systems should be continuously
improved to ensure increasing levels of power quality to the customers (Penido et al., 2008). It is
well-known that most DS are considerably unbalanced, and in high density load areas such as city
centers, the network topology can be highly meshed. Under these circumstances, the three-phase
four-conductor configuration with multiple neutral grounding has been largely adopted, due to low
installation costs and better sensitivity for fault protection, when compared with the three-phase
three-conductor configuration. The presence of neutral conductors and grounding affect not only
the system operation but also equipment and human safety (Penido et al., 2008). A distribution
system is basically unbalanced because of many factors, such as untransposed feeders, conductor
bundles, single-phase loads, unequal three phase loads and single- and double-phase ‘radial spurs’
on primary feeders (Segura et al., 2011). Furthermore, even when a network is balanced,
asymmetrical faults can introduce imbalance. To avoid significant errors arising from inherent
system imbalance, a rigorous analysis of the distribution system using detailed component models
is required (Segura et al., 2011). Conventional load flow methods like Gauss-Seidel, Newton-
Raphson and fast decoupled techniques are inefficient in solving such networks (Prakash &
Khatod, 2016). These methods are not very suitable for distribution networks, because of the
following characteristics of such systems (Elsaiah et al., 2012):
1. Radial structure with sometimes weakly-meshed topology.
2. High resistance to reactance ( R
X
) ratio which sometimes causes the Newton-Raphson (NR)
and the Fast-Decoupled (FD) methods to diverge.
3
3. Untransposed or rarely transposed lines where it is often inappropriate to neglect the mutual
coupling between phases.
4. Unbalanced loads along with single-phase and double phase laterals.
5. Unbalanced distributed loads.
6. Dispersed generation.
These characteristics, combined with the large number of nodes and branches of distribution
networks make the direct use of the aforementioned techniques unsuitable and inefficient for
power flow studies of unbalanced distribution systems (Elsaiah et al., 2012).
The loading of a distribution feeder is inherently unbalanced due to a large number of unequal
single-phase loads and the nonsymmetrical conductor spacing of three-phase underground and
overhead line segments. Due to these factors, conventional power flow programs used for
transmission system studies do not show good convergence properties for distribution systems .
Single phase representation of three phase system is used for power flow studies on transmission
system which is assumed as a balanced network in most cases. But due to the unbalanced loads,
radial structure of the network and untransposed conductors makes the distribution system as an
unbalanced system. Hence three phase power flow analysis need to be used for distribution
systems. The three phase power flow analysis can be carried out in two different reference frames
namely phase frame and sequence frame. The phase frame deals directly with unbalanced
quantities while the sequence frame deals with three separate phasor systems which, when
superposed, give the unbalanced conditions in the circuit (Balamurugan & Srinivasan, 2011).
Different methodologies and approaches have been developed to find the optimal size of DG and
identify the optimal location to install the DG (Kansal et al., 2013). (Das, 2015) developed optimal
sizing and placement of DG in a radial distribution system using loss sensitivity factor and firefly.
4
In the work, the loss sensitivity factor is used for optimal DG location and the firerfly algorithm
are used for DG sizing. (Bhimarasetti & Kumar, 2014) presented DG planning in unbalanced mesh
distribution system with different unbalances, where variational algorithm is used to find the
optimal DG capacity and voltage index is used to obtained the optimal DG size. (Chou & Butler-
Purry, 2014) showed that loading unbalance degree and DG power output has significant effect on
voltage stability. (Reddy & Manohar, 2013) developed optimal placement of DG on unbalanced
distribution network. (Al-Sabounchi et al., 2011), (Anwar & Pota, 2012), (Abdelaziz et al., 2015),
(Othman & Hegazy, 2015), (Gómez-González et al., 2015), (Dahal & Salehfar, 2016) and a lot of
researchers has shown that DG placement on unbalanced system has significant effect on the
voltage and the power flowing in the system.
In order to ensure a balance between generation and consumption with a minimum power loss in
an unbalanced distribution systems, this research tends to introduce voltage stability index and
firefly algorithm method for optimal location and sizing of DG in the network. Also, Backward-
Forward Sweep (BFS) based method will be used to achieve a convergence guaranteed of power
flow analysis. The BFS method is the most suitable method in solving power flow problem due to
its simplicity and better convergence performance compared to Gauss-Siedeland Newton-Raphson
based methods under the assumption of radial network structures (Demirok et al., 2012).
1.2 Significance of Research
The significant of this research is the development of Firefly algorithm and voltage stability index
for optimal DG planning in an unbalanced three-phase distribution network considering the
loading conditions of the phases, thereby minimizing and maximizing the losess and loading of
the network respectively. Other researchers had not consider the loading on each phase of the
network.
5
1.3 Statement of Problem
The unbalanced loading between lines causes unsymmetrical current to flow and irregular voltage
drop in a network causing branch losses which differ in each phase. Also, the low voltage and high
current in the distribution network causes active and reactive power loss. The losses in the branches
can be reduced by either external power injection or distribution automation system. As such,
many researches have been carried out on optimal placement of DG, from analytical, heuristic and
meta-heuristic to hybrid technique. Most of the researchers assumed a balanced network but in
practic distribution network is unbalanced in nature due to different loading and asymmetrical
component between the phases. This research work tend to address unbalanced and loading
conditions in a radial distribution system in the presence of DG placement using firefly algorithm
and voltage stability index. Also, in order to ensure a simple and convergence guaranteed power
flow, a Backward-forward sweep based method is used in this work.
1.4 Aim and Objectives
The aim of this research is to develop firefly algorithm based method for optimal Distributed
Generation planning in unbalanced three phase distribution network using voltage stability index.
The objectives of the research are:
1. To perform a three-phase power flow algorithm based on the unbalanced three-phase
distribution network.
2. To apply voltage stability index for optimal DG location
3. To develop firefly algorithm for optimal DG sizing.
4. To validate the developed method by comparing with the work of Othman et al. (2016)
using power loss and voltage profile as the performance metric.
6
1.5 Methodology
The following methodology are used to carried out this research work:
1. Collection of necessary data (such as conductor length, conductor cross sectional area,
conductor current carrying capacity, transformer rating and loads) from Ungwan Boro
distribution network Barnawa area 1 office, Kaduna and modelling of the ungwa Boro
distribution network.
2. Performance of three-phase power flow based on forward-backward sweep by:
i. For backward sweep: sum currents or power flows (and possibly updates voltages).
ii. For forward sweep: calculate voltage drops (and possibly update currents/power flows).
iii. Repeat steps i and ii above until convergence criteria is met
3. Perform base case power flow analysis on the radial distribution feeder (IEE 37-node
feeder and Mahuta 19-node feeder)
4. Modelling and establishment of an appropriate DG model that will suit the loading
condition of the given network
5. Application of voltage stability index for optimal DG location by:
i. Running a three-phase power flow and calculate the voltage stability index (VSI).
ii. Rank the node based on VSI obtain in (5.i) in descending order.
iii. Select the node with the highest VSI value as the DG location.
6. Development of a Firefly algorithm (FA) that will be suitable for optimal DG sizing by:
7
i. Generating the initial fireflies (DG active power, system loadability)
ii. Consider the fitness function as the objective function
iii. Define the firefly algorithm parameters based on (Yang & He, 2013)
iv. Calculating the active power loss and system loadability that correspond to all the initial
fireflies based on the three-phase power flow and the VSI obtained in (5) above
v. Select the best DG size that achieve a minimum active power loss and system loadability
(the brightest firefly)
vi. Updates the firefly (DG size) and keep the best DG size and system loadability as a new
fireflies (system variable)
vii. Continue steps (iv-vi) until the convergence criteria is met
7. Implementation of the developed models on unbalanced three-phase IEEE 37-node feeder
and Ungwa Boro radial distribution network
8. Validation of result obtained by comparing with the result of Othman et al. (2016).
1.6 Dissertation Organization
Chapter One presents the general introduction of the research work while chapter two gives a detail
explanation and mathematical modelling that govern this research work and review on work
related to this dissertation.Chapter three gives the full details on the step taken to achieved this
research work and chapter four presents the results and discussion of the work. Finally, conclusion,
recommendations of further work and limitation of the work are presented in chapter Five while
list of cited work, data and MATLAB codes are given in appendices provided at the end of this
dissertation.
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
