ABSTRACT
This research work presents the Application of Cuckoo Search Algorithm for the simultaneous placement of distributed generation units and shunt capacitor banks in radial distribution networks. The approaches used in most literatures for determining the optimal allocation of DG units and Capacitor banks did not consider the simultaneous placement of multiple combinations of DG units and capacitor banks in order to obtain the best combination of optimal sizes and installed an appropriate location for optimal network performance. The Cuckoo Search Algorithm was applied to determine the optimal location and sizes of Micro-DG units and Shunt Capacitor Banks with an objective of minimizing the total active and reactive power loss and maximizing voltage profile improvement and voltage stability of distribution networks. The DGs units and Capacitor banks where separately and simultaneously applied on standard IEEE 33 and 69 test bus systems and on 50 bus Canteen Feeder in Zaria distribution network. The result obtained from the simultaneous allocation was validated by comparing with that obtained from the separate allocation of Distributed Generators and Capacitor Banks. For the 33 bus test system, the Cuckoo search algorithm found the optimal sizes and locations of the best simultaneous combination of the DG units and capacitor bank allocated to be 515.69 kW at bus 25, 214.01 kW at bus 32 and 572.27 kVAr at bus 30 with 63.29% and 59.38% reduction in active and reactive power loss, 6.32% Improvement in average voltage profile and 7.89% in overall VSI, as compared to 31.65% and 31.25% active and reactive power loss reduction, 5.11% improvement in average voltage profile and 6.13% in overall VSI for separate DG placements, and 53.16% and 53.13% power loss reductions, 3,95% improvement in average voltage profile and 3.85% improvement in overall VSI for separate capacitor bank placement when compared to the base case result. For the standard IEEE 69 test bus system the optimal sizes and locations of the best combinations of DG units and CBs that were simultaneously allocated were found to be 239.83 kW at bus 53, 885.58 kW at bus 50 and 1408.79 kVAr at bus 50 with 74.29% and 79.17% reduction in active and reactive power loss, 2.34% Improvement in average voltage profile and 3.79% in overall VSI as compared to 51.43% and 54.17% active and reactive power loss reduction, 2.02% improvement in average voltage profile and 2.69% in overall VSI for the separate DG placements and 28.57% and 31.25% active and reactive power loss reductions, 1.65% improvement in average voltage profile for separate capacitor banks placement when compared to base case results. Finally for the 50- bus Canteen Feeder in Zaria distribution network, the optimal sizes and locations of the best combinations of DG units and capacitor banks simultaneously allocated was 247.19 kW at bus 23, 137.82 kVAr at bus 25 and 131.78 kW at bus 25 with 17.77% and 17.76% reduction in active and reactive power loss, 0.47% improvement in average voltage profile and 0.32% improvement in overall VSI as compared to 15.70% and 15.79% active and reactive power loss reductions,0.45% improvement in average voltage profile and 0.30% improvement in overall VSI for the separate DG placements, 9.92% and 9.87% active and reactive power loss reductions, 0.38% improvement in average voltage profile and 0.28% improvement in overall VSI for separate capacitor banks placements when compared to the base case results. From results comparison, it is evident that the simultaneous placement of DGs and Capacitor banks using the cuckoo search algorithm gave a better performance to the separate placement of the DG units and capacitor banks.
TABLE OF CONTENTS
TITLE PAGE II DECLARATION III CERTIFICATION IV DEDICATION V ACKNOWLEDGEMENT VI ABSTRACT VIII LIST OF FIGURES XVI LIST OF TABLES XVIII
LIST OF ACRONYM XIX
CHAPTER ONE: INTRODUCTION
1.1 Background of study 1
1.2 Motivation 5
1.3 Statement of Problem 6
1.4 Aim and Objectives 7
1.5 Methodology 8
1.6 Dissertation Organization 9
CHAPTER TWO: LITERATURE REVIEW
2.1 Introduction 10
2.2 Review of Fundamental Concepts 10
2.2.1 Radial distribution network (RDN) 10
2.2.2 Power flow analysis 11
2.2.3 Inclusion of DG unit in the Power Flow solution 15
2.2.4 Distributed generation 15
2.2.4.1 Types of Distributed generation 18
2.2.4.2 Classification of Distributed Generations 182.2.5 Distributed Generation model types 19
viii
2.2.5.1 DG Modelled as a PV type 19
2.2.5.2 DG Modelled as a PQ Node 20
2.2.6 Capacitor Banks 22
2.2.6.1 Shunt Capacitor Banks 23
2.2.7 Voltage Stability 24
2.2.7.1 Voltage Stability Index 25
2.2.8 Standard IEEE Test Benchmarks 28
2.2.8.1 Standard IEEE 33 Bus Distribution Network 29
2.2.8.2 Standard IEEE 69 Bus Distribution Network 29
2.2.8.3 Zaria Distribution Network 30
2.2.9 Methods for Optimal DG and Capacitor bank Placement and Sizing 31
2.2.10 Cuckoo Search Algorithm 32
2.2.11 Performance Metrics 36
2.3 Review of Similar Works 38
CHAPTER THREE: MATERIALS AND METHODS
3.1 Introduction 45
3.2 Materials 45
3.2.1 Personal Computer 45
3.2.2 Matlab 2013a Software 45
3.2.3 Distribution Network Parameters 46
3.3 Methods 46
3.3.1 Acquisition of Relevant Data 46
3.3.2 Base Case Power Flow Analysis 46
3.3.3 Application of the Cuckoo Search Algorithm 47
ix
3.3.4 Comparison of Results obtained 48
3.4 Distributed Generator Model 48
3.5 Model of DG units and Shunt capacitors in Radial Distribution Networks 48
3.6 Objective Function 50
3.7 Cuckoo Search Algorithm 51
3.8 Cuckoo Search Algorithm for Simultaneous DG and Capacitor bank allocation 52
3.9 Voltage Stability Index 54
3.10 Standard Test Systems 54
3.10.1 IEEE 33 Bus System 54
3.10.2 IEEE 69 Bus System 55
3.10.3 Zaria Distribution Network 55
3.11 Performance Evaluation 55
CHAPTER FOUR: RESULT AND DISCUSSIONS
4.1 Introduction 56
4.2 IEEE 33 bus system 56
4.2.1 Base case system 56
4.2.2 Effect of Capacitor allocation 58
4.2.3 Voltage and VSI Profiles after Capacitor allocation 59
4.2.4 Effect of DG allocation 61
4.2.5 Voltage and VSI Profiles after DG allocation 62
4.2.6 Allocation of Multiple DGs and CBs combinations Simultaneously 64
4.2.7 Voltage and VSI Profiles after simultaneous DG and CB allocation 66
4.2.8 Power loss reduction profiles for 33- bus System 68
x
4.2.9 Summary of results for 33- bus System 70
4.3 IEEE 69 bus system 71
4.3.1 Base case system 71
4.3.2 Effect of Capacitor allocation 74
4.3.3 Voltage and VSI Profiles after CB allocation 76
4.3.4 Effect of DG allocation 77
4.3.5 Voltage and VSI Profiles after DG allocation 80
4.3.6 Allocation of Multiple DGs and CBs combinations Simultaneously 82
4.3.7 Voltage and VSI Profiles after simultaneous DG and CB allocation 85
4.3.8 Power loss reduction profiles for 69-bus System 87
4.3.9 Summary of results for 69- bus System 89
4.4 Zaria 50-bus Canteen Feeder 89
4.4.1 Base case system 90
4.4.2 Effect of Capacitor allocation 91
4.4.3 Voltage and VSI Profiles after CB allocation 93
4.4.4 Effect of DG allocation 95
4.4.5 Voltage and VSI profiles after DG allocation 97
4.4.6Allocation of Multiple DGs and CBs combinations simultaneously 99
4.4.7 Voltage and VSI Profiles after simultaneous DG and CB allocation 102
4.4.8 Summary of results for 50- bus Zaria Canteen Feeder System 104
CHAPTER FIVE: CONCLUSION AND RECOMMENDATION
5.1 Conclusion 108
5.2 Significant Contributions 110
xi
5.3 Limitations 111
5.4 Recommendations for Further Work 111
xii
REFERENCES 110
CHAPTER ONE
INTRODUCTION 1.1 Background of Study
Electrical power systems are evolving from the conventional power systems known, where generation plants are connected to a distribution or transmission network to new trends involving decentralized systems consisting of small generating units connected directly to a distribution system near demand consumption. This type of generating unit is termed Distributed Generation (DG) (Pepermanset al., 2005). Among various DG types, the renewable energy (Micro- DG) units are becoming more common since majority of other DG resources (power electronic and synchronous generator based DG units) are capable of injecting harmonics to the network and are not environmentally friendly. A capacitor bank is another device that can be directly connected to a distribution network to provide a proper balance between active and reactive power injection and consumption. Among the types of capacitor banks, the shunt compensating capacitor bank is mostly suitable for distribution networks because of its ability to inject as well as absorb reactive power when needed thereby protecting against low power factor associated with network loads that leads to power losses, voltage drops and voltage instability.
Distribution system is a network that provides a final link between the high voltage transmission system and the consumers. In distribution system, among the various network topologies, the radial network topology is common because of its simplicity. The challenges of radial distribution system include; high power losses, voltage drops, transients and harmonic distortion on the distribution networks (Ogunyemi and Adejumobi, 2012). Solving
2
these power quality problems using conventional solutions such as distribution network expansions and substation upgrades are more capitally incentive and requires years of planning and implementation. The adverse effects of harmonic injection and low power factor in power systems include the accumulation of zero sequence current in the neutral line causing overheating of the neutral cable, overheating of transformers and power system equipments, additional losses and voltage instability. This therefore provides a motivation to select a Distributed Generation and shunt capacitor bank (Heydari et al, 2013). Distribution networks are mostly radial in nature with very rare exceptions, with the customers having one source of supply. A typical distribution network consists of distribution substation, feeders, switches, fuses, transformers, voltage regulators, meters and circuit breakers as shown in Figure 1.1
Figure 1.1: Schematic Diagram of a Simple Distribution Network (Kersting, 2012)
The deregulation of power networks to augment the imbalances between generation and consumption results in loss of the networks passive nature. This scenario subsequently results in technical issues that may affect the power quality of the entire network (Machowskiet al., 2011). Poor voltage quality in distribution networks which can be due to variations in
3
consumer‟s demand forces electrical equipment to draw more than their rated current, resulting in excessive heating capable of inflicting severe damages to Power system equipments(Watson and Miller, 2015). Due to low voltage level and high current in distribution systems, the largest power loss among the three (3) power system sections of generation, transmission and distribution, belongs to the distribution section. Such that the line losses at the distribution level in Nigeria constitute about 15% to 30% of total power generation (NERC). Nowadays, because of the industrialization of societies and overloading of distribution networks, the voltage stability of distribution networks has become an important subject for consideration. Over the years, many efforts have been made to reduce the losses in distribution networks, which include the use of series voltage regulators, capacitor placements, distributed generation (DG) resources. Shunt capacitors have been employed in several literatures to locally compensate for the reactive power in the network and in consequence reduce the power loss in the lines and improve their voltage profile as in the works of Gailegoet al, (2001), Ellithyet al,(2008) and Khodret al., (2008), where shunt capacitors optimal placement was applied as effective economic tool to an improved and stable distribution network.
Distributed generation (DG) units are employed at the distribution level to supply power and reduce losses. The optimal sizing and placement of these resources to minimize the power loss
4
improve voltage profile; enhance systems reliability indices have been proposed in several researches (Pepermanset al.,2005, Archarya,et al., 2006, Akorede, 2010, Hung et al, 2014).
With increase in economic growth and development, load demands in distribution networks are susceptible to sharp increment. Hence, the distribution networks in most developing nations like Nigeria are operating very close to the voltage instability boundaries. The decline of voltage stability margin is one of the important factors which restrict increment in loads served by distribution companies (Jain, et al. 2014). The rapidly increasing electricity demand and difficulties in providing the required capacity using traditional solutions, such as transmission network expansions and substation upgrades, the need to protect power systems from voltage associated disturbances and provide an argumentation to grid generated power lead to the emergence of Distributed Generation (DG) and other network compensating devices (Pepermans, 2005) Several DG types have been used in finding solutions to power loss reduction and voltage profile improvements and voltage stability in networks over the years but some have been found to create further disturbances in the network during power injection and thus might not be suitable for voltage support in power networks. In this work renewable energy DG units would be made use of because of its low harmonics injectioninto power systems and as an environmentally friendly alternative to other DG units.
Capacitors alongside DGs have been used in previous work to provide active and reactive power compensations on a network and have been confirmed to be a good approach for
5
optimum network operation and planning (Heydariet al, 2013, Esmaeilianet al, 2014, Harisha and Lakshmi, 2015). Analytical and numerical techniques have been employed in the search for optimal position and size of distributed generation and capacitors in power networks. Their computational complexity and inability to fully incorporate the nonlinearity of the power networks prompted the search for better and more robust techniques (Ming et al., 2014). The evolution of meta-heuristics algorithms has led to a new, faster and robust approach of solving complex power system optimization problems (Binitha and Sathya, 2012). Out of the many meta-heuristics algorithm presently developed, the Cuckoo search algorithm was applied to the location and sizing of Type 1DG units and shunt capacitor banks in radial distribution networks with respect to a multi-objective function encompassing power loss minimization, improvement of voltage profile and voltage stabilityina radial distribution network. 1.2 Motivation
A sharp increase in demand for energy has caused suppliers of energy to search for a quicker and less expensive alternate means of improving the sustainability and stability of power distribution networks. Also the need to provide additional reactive power compensation due to network loads that are majorly inductive in nature, in order to provide proper balance between active and reactive power injection and consumption, motivated the choice of renewable Distributed Generationand shunt capacitor banks as an emergency approach to solve this problem and provide augmentation to the grid generated power in a less expensive means as
6
compared to the conventional transmission expansion or network reconfiguration and environmentally friendly compared to some other DG options. Secondly, large proportions of electric machinery used in industries have inherently low power factor and so also are most loads on power networks. When the overall power factor of generating stations is low, systems become inefficient, corresponding cost of electricity becomes high, increase in power losses and poor voltage regulation of loads. In a bid to address issues on poor power factor of loads the shunt capacitor bank, which have been employed mostly for the purpose of power factor correction purposes, thus improvingthe low power factor problems in power systems and reducing power losses will also be employed.In this work the DG units and shunt capacitor banks will be simultaneously placed on radial distribution networks to obtain the best combinations suitable for reduction in power loss and improving the voltage profile of the networks using the cuckoo search algorithm. 1.3Statement of Problem
With DG mainly considered as an active source of energy by most power distribution companies in order to get maximum benefit from electricity purchased, drawing maximum active power can result into deficiency in fulfilling the kVAr of the reactive load requirement which when worst may lead to voltage collapse situations. Most researchers have addressed the issue of voltage instability in distribution networks with the use of DGs alone, which studies have shown might not be able to meet the kVAr requirements especially since networks are having loads that are mostly inductive in nature and may require the use of additional compensators to meet the kVAr demands from loads. The use of shunt capacitor banks for reactive power compensation, power factor correction and voltage stability improvement have shown to be efficient in meeting the kVAr demands in situations where
7
inductive loads dominate the loads in a network. Another problem widely faced when allocating DG units and capacitor banks in distributionssystems is how to efficiently integrate them optimally in order to achieve improved system performance. The search for an optimal location and the size of DG units and capacitor banks to be placed can be quite challenging. In most cases, the methods employed for locating and sizing DG units and capacitor banks in distribution network are either analytical method or heuristic which arecomputationally exhaustive and time consuming optimization techniques and are characterized by slow convergence of power flow especially when used for complex system analysis. The meta-heuristic method tends to provide a lesser time consumption and faster convergence of power flow analysis. 1.4Aim and Objectives The aim of this work is to apply a Cuckoo search algorithm for the placement ofDistributed Generation (DG) units and shunt capacitor banks in radial distribution networks with the intent of minimizing power loss (active and reactive), improving the voltage profile and voltage stability. To achieve the set aim, theobjectives are:
i. To Run power flow for base case power loss and voltage at each bus
ii. To integrate the Cuckoo search algorithm to the power flowfor separate and simultaneous (combined)DG units and shunt capacitor bank location and sizing on radial distribution networks in MATLAB 2013a environment.
iii. To compare the results obtained from the separate sizing and location of DG units and capacitor banks with that from the simultaneous DG units and shunt capacitor bank sizing and locationon the basis of power loss minimization, voltage profile and voltage
8
stability improvement using cuckoo search algorithm on IEEE 33 and 69 radial test buses and on 50-bus Canteen Feeder in Zaria distribution network.
1.5 Methodology The following steps which comprises the methodology adopted for this research are as follows
i. Base case load flow program for base case power loss and voltage values
ii. Application of the CSA for simultaneous DG units and shunt capacitor bank location and sizing on standard IEEE- 33 and 69 radial distribution networks and 50 bus Canteen Feeder in Zaria distribution network with corresponding loss reduction and voltage profile improvement.
iii. Validation through results comparison
a. Comparison of results obtained from the simultaneous placement of the DG units and shunt capacitor banks to that obtained from the separate (singular) placements of DG units and CBs
1.6Dissertation Organisation
A general introduction has been presented in chapter one, while chapter two presented a review of relevant fundamental concepts such as definition of a radial distribution network distributed generation, capacitor bank, voltage stability and voltage stability indices, power flow analysis for radial distribution network, methods for DG allocation etc. and a review published similar works directly related to this research. Chapter three presents the materials and methodology used such as hardware and software employedand the application of DG units and shunt capacitor banks on radial distribution test networks were presented. In chapter
9
four, results obtained after the simultaneous allocation of the DGs and shunt capacitor banks were presented while conclusion and recommendations for further works were presented in chapter five. Quoted references and appendices were also presented at the latter end of this research report.
10
Do you need help? Talk to us right now: (+234) 08060082010, 08107932631 (Call/WhatsApp). Email: [email protected].
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
