BAHIR DAR UNIVERSITY

INSTITUTE OF TECHNOLOGY

SCHOOL OF RESEARCH AND GRADUATE STUDIES

FACULTY OF ELECTRICAL AND COMPUTER ENGINEERING

Master of Science in power system engineering

Msc thesis on:

Fuzzy logic Based network Reconfiguration for Reliability analysis and
improvement of Jigjiga Distribution system
BY

Habitye Amide
Advisor

Dr.Ing Belachewu Bantiyrga

June 2016 E.C

Jigjiga, Ethiopia
 BAHIR DAR UNIVERSITY
BAHIR DAR INSTITUTE OF TECHNOLOGY
SCHOOL OF RESEARCH AND GRADUATE STUDIES
FACULTY OF ELECTRICAL AND COMPUTER ENGINEERING

Approval of thesis for defense result

I hereby confirm that the changes required by the examiners have been carried out and incorporated
in the final thesis.

Name of Student ________________ Signature ______________ Date______________

As members of the board of examiners, we examined this thesis entitled
“____________________________________________________________________________”
by_______________________________. We hereby certify that the thesis is accepted for fulfilling
the requirements for the award of the degree of Masters of Science in
“_____________________________________________________________________”.

Board of Examiners

Name of Advisor Signature Date

__________________ ____________ ___________

Name of External examiner Signature Date

__________________ ____________ ___________

Name of Internal Examiner Signature Date

__________________ _____________ ___________

Name of Chairperson Signature Date

______________ ____________ _____________

Name of Chair Holder Signature Date

_______________ _____________ _____________

Name of Faculty Dean Signature Date

_______________ _____________ _____________

I
Acknowledgements
First and foremost, my praises and thanks go to God, the almighty, for his showers of blessings
throughout completion of the thesis successfully.
I would like to express my deeper and sincere gratitude to Bahirdar institute of Technology for
providing Msc chance and giving me the opportunity to do thesis, and all instructors who provided
excellent lectures throughout Msc sessions.
My grateful thanks to my advisor Dr.ing Belachewu bantiyrga (PHD) for providing invaluable
guidance professional expertise and suggestion throughout this thesis. He has taught me the
methodology to carry out the thesis and present the thesis as clearly as possible.
I very much thankful to my wife for motivation, understanding and continuous support to complete
this thesis.
I would also want to take this opportunity to thank the staff, supervisors and managers of EEU and
EEP for their hospitality and assistance in providing me with numerous valuable data and useful
information for my work. I would especially like to thank Mr Demlewu Teshome ,jigjiga substation
manager , Abebawu Adamu jigjiga district EEU customer service head, for providing the necessary
information and data as well as making the necessary arrangment to visit jigjiga substation site while
collecting the necessary information and recording data.

II
Abbreviation

AAC All Aluminum conductor

AENS Average energy not supplied
ASAI Average Service Availability Index
ASUI Average Service Unavailability Index
CAIDI Customer Average Interruption Duration Index
CLS Connected load supply
DHF Distance of nearby healthy feeder
EEU Ethiopian electric utility
EF Earth fault
EENS Expected energy not supplied
ETAPS Electrical Transient Analysis Program Software
FL Fuzzy Logic
MTTF Mean-time-to-failure
MWH Megawatt hour
SAIDI System average interruption duration index
SAIFI System average Interruption frequency index
SC Short circuit
SL Section line length

III
Contents
Approval of thesis for defense result................................................................................................. I
Acknowledgements........................................................................................................................... II
Abbreviation..................................................................................................................................... III
List of Figures .................................................................................................................................. VII
List of Table ...................................................................................................................................... IX
Abstract ............................................................................................................................................ XI
1. INTRODUCTION ............................................................................................................................. 1
1.1 Back ground ............................................................................................................................. 1
1.1 .1 Overview of Jigjiga town Distribution system ................................................................. 3
1.2 Problem of statement ............................................................................................................. 4
1.3. Objectives of the thesis .......................................................................................................... 5
1.4 Significance of the Study ......................................................................................................... 5
2. LITERATURE REVIEWS ........................................................................................................... 7
2.1 Distribution System Reliability ............................................................................................... 7
2.2 Distribution system Reconfiguration .................................................................................... 10
2.3 Markove chain Model ......................................................................................................... 12
2.3.1 Series System ................................................................................................................ 14
2.3.2 Parallel System .............................................................................................................. 15
2.4 Power Quality, Reliability and Availability ............................................................................ 15
2.4.1 Power Qualities............................................................................................................... 16
2.4.2 Reliability ........................................................................................................................ 16
2.4.3 Availability ...................................................................................................................... 17
2.5 Reliability Improvement Methods ........................................................................................ 17
2.5.1 Application of Distribution network reconfigurations ................................................... 17
2.6. Methods of Distribution Network Reconfigurations ........................................................... 19
2.6.1 Fuzzy Logic Method ........................................................................................................ 19
2.6.2 Heuristic Optimazation methods.................................................................................... 20
2.6.3 Artificial Neural Network ................................................................................................ 20

IV
 2.6.4 Tabu Search Method ...................................................................................................... 20
2.6.5 Simulated Annealing Method ......................................................................................... 21
2.6.6 Ant colony optimization Method ................................................................................... 21
CHAPTER THREE ........................................................................................................................ 22
3. METHODOLOGY, DATA COLLECTION & ANALYSIS ..................................................... 22
3.1 Distribution System Reliability Indices .................................................................................. 22
3.1.1 Load Point Indices ........................................................................................................... 23
3.1.2 System Reliability Indices.............................................................................................. 23
3.2 Power Outage Cost Evaluation .............................................................................................. 25
3.3 Impacts of Improvement Techniques and Protection System on Reliability ...................... 26
3.4 Reliability Evaluation ............................................................................................................. 26
3.4.1 Historical Reliability Assessment .................................................................................... 26
3.4.2 Predictive Reliability Analysis ......................................................................................... 27
3.5 Distribution System Failures Cause Analysis ......................................................................... 28
3.6 Factors Influencing Power System Reliability ....................................................................... 30
3.6.1 Component Statistics ...................................................................................................... 30
3.6.1.1 Repair failure rate .................................................................................................. 31
3.6.1.2 Repair time ............................................................................................................. 33
3.6.2 System Configurations .................................................................................................... 33
3.6.3 Time-varying Load .......................................................................................................... 33
3.6.4 Environmental Conditions .............................................................................................. 33
3.7 Failure Modes and Effects Analysis (FMEA) .......................................................................... 33
3.8 Reliability Improvement Strategy ......................................................................................... 34
3.9 Modeling of jigjiga Distribution System with ETAP software ............................................... 35
3.9.1 Study of jigjiga Power Distribution System .................................................................... 40
3.9.2. Interruption Data from 2011 E.C to 2014 E.C ................................................................ 41
3.9.3 Comparison of the calculated values of Reliability Indices with benchmarks ............... 52
3.9.4 Outage Cost Evaluation of Jigjiga substation 33kv Feeder2 Line ................................. 54
3.9.5 Method of fuzzy logic based Distribution Network Reconfiguration ............................. 56
4 .CHAPTER FOUR ........................................................................................................................... 62
4.1 Result and Discussion ........................................................................................................... 62

V
 4.2 System Modeling ................................................................................................................... 62
4.3. Simulation and Result Analysis of jigjiga substation Togowuchale Feeder ......................... 66
4.3.1 Optimal placement of sectionalize switch ..................................................................... 66
4.3.2 Optimal placement of tie switch .................................................................................... 74
4.4 Economical Cost Analysis ...................................................................................................... 91
CHAPTER FIVE.................................................................................................................................. 97
CONCLUSIONS AND RECOMMEDATIONS ....................................................................................... 97
5.1 CONCLUSIONS ....................................................................................................................... 97
5.2 Recommendation .................................................................................................................. 98
5.3 Future works ......................................................................................................................... 99
References .................................................................................................................................... 100
Appendex A ................................................................................................................................... 103
Appendex B ................................................................................................................................... 103

VI
List of Figures
Figure 1.1 single line diagram of jigjiga substation .................................................................... 3
Figure 2.1 system reliability subdivision .................................................................................... 8
Figure 2.2 Radial Distribution Systems .................................................................................... 10
Figure 2.3. Ring Feeder with Two Energy Sources ................................................................... 11
Figure 2.4 Interconnected Systems ........................................................................................... 12
Figure 2.5 Transition Diagram of Component .......................................................................... 13
Figure 2.6 Average State Cycle................................................................................................ 14
Figure 2.7 Hierarchies of Power Quality, Reliability and Availability ..................................... 16
Figure 2 .8 Structure of Fuzzy Logic System ............................................................................ 20
Figure 3.1 Definitions of Failure .............................................................................................. 29
Figure 3.2 Total Times for Repair/Replacement of Passive Failure ........................................ 30
Figure 3.3 Bath-Tub Component‟s Life ................................................................................... 32
Figure 3.4 ETAP Algorithm for Calculation of Reliability Indices ......................................... 37
Figure 3.5 Modeling of Togowuchale line 33kv feeder2 ......................................................... 38
Figure 3.6 Load flow of modeling 33kv feeder2 line ............................................................. 39
Figure 3.7 Percentage average frequency interruption of each type of fault .......................... 46
Figure 3.8 Percentage average duration interruption of each type of fault ........................... 47
Figure 3.9 four years average value of SAIFI in jigjiga substation ...................................... 50
Figure 3.10 four years average value of SAIDI in jigjiga substation ....................................... 51
Figure 3.11 Fuzzy logic control for benefits of sectionalizer switch allocation ..................... 57
Figure 3.12 Fuzzy logic control for benefits of tieswitch allocations ...................................... 59
Figure 3.13 overall work flow of methodology ..................................................................... 61
Figure 4.1 system modeling of togowuchale line ................................................................ 64
Figure 4.2 window of etap19 existing system opened page ................................................. 65
Figure 4.3 Fuzzy set used to out put variables ..................................................................... 67
Figure 4.4 for sectionalize switch allocation problem ......................................................... 68
Figure 4.5 Rule created for optimal placement of sectionalize switch ................................ 69
Figure 4.6 surface high wind speed and number of branches .............................................. 70
Figure 4.7 surface high wind speed and total connected loads ............................................. 71
Figure 4.8 surface high wind speed and section line length ................................................... 71
VII
Figure 4.9 Rule viewer of fuzzy optimization for optimal placement of section switch ........ 72
Figure 4.10 fuzzy input-out put variable and the Fl designer for tieswitch allocation ............ 75
Figure 4.11 Rule created for optimal placement of tieswitch allocation ................................ 75
Figure 4.12 surfaceviewer of DHF and LMF for tie switch allocation .................................... 76
Figure 4.13 surfaceviewer of DHF and TCLF for tie-switch placement decisions ............... 76
Figure 4.14. Rule viewer for tieswitch placement decisions .................................................... 79
Figure 4.15. placing one sectionalizer switch on the existing system ..................................... 82
Figure 4.16. placing two sectionalize switch on the existing system ....................................... 84
Figure 4.17. placing three sectionalize switch on the existing system ..................................... 86
Figure 4.18. NR of 33kv F2 with 33kv F4 .............................................................................. 88

VIII
List of Table
Table 3.1 Number of Transformers and Rating of Each Transformer ......................................... 40

Table 3.2 Planned and Unplanned power interruption 2011 E.C ............................................... 41

Table 3.3 Planned and Unplanned power interruption 2012 E.C ............................................... 42

Table 3.4 Planned and Unplanned power interruption 2013 E.C ............................................... 42

Table 3.5 Planned and Unplanned power interruption 2014 E.C ............................................... 43

Table 3.6 Total planned and unplanned power interruption from 2011-2014 ........................... 43

Table 3.7 Frequency Interruptions from 2011 E.C to 2014 E.C .............................................. 44

Table 3.8 Duration Interruptions from 2011 E.C to 2014 E.C ................................................ 45

Table 3.9 Percentage of the cause of average duration and frequency interruptions ......... 46

Table 3.10 Calculated reliability indices of 2011 jigjiga Substation ......................................... 48

Table 3.11 Calculated reliability indices of 2012 jigjiga Substation ......................................... 48

Table.3.12 Calculated reliability indices of 2013 jigjiga Substation ........................................ 49

Table 3.13 Calculated reliability indices of 2014 jigjiga Substation ....................................... 49

Table 3.14 Calculated reliability indices the average of four years at jigjiga Substation ....... 50

Table 3.15 Calculated reliability indices of jigjiga city feeder for year 2011-2014 E.C ......... 51

Table 3.16 Comparisons of SAIFI and SAIDI values with different countries .......................... 52

Table 3.17 load with number of customer at jigjiga substation out going feeders in 2011 ..... 53

Table 3.18 load with number of customer at jigjiga substation out going feeders in 2012..... 53

Table 3.19 load with number of customer at jigjiga substation out going feeders in 2013 .. 53

Table 3.20 load with number of customer at jigjiga substation out going feeders in 2014 .. 54

Table 3.21 Ethiopia electricity tariff (Birr/KWh) since Dec 1, 2014 E.C ............................ 55

Table 3.22 Summary of estimated interruption cost from 2011 to 2014 E.C in 33kv F2 ...... 56

Table 3.23 Set of rule switching allocation ......................................................................... 58

IX
Table 3.24 The Sets of Rules Used for Allocation of tie switch ............................................... 59

Table 4.1 Simulation result of reliability indices for existing system .................................... 73

Table 4.2 Length of near by health feeder,buses and total connected load to bus ............. 77

Table 4.3 normalized value of table table 4.2 ....................................................................... 80

Table 4.4 one sectionalize switch errection cost estimation ................................................. 91

Table 4.5 two sectionalize switch errection cost estimation ............................................... 92

Table 4.6 three sectionalize switch errection cost estimation .............................................. 93

Table 4.7 10km mv line constraction cost estimation .......................................................... 94

Table 4.8 summary of the results of reliability improvements of all scenarios ..................... 95

Table 4.9 summary of the results of energy oriented reliability improvements ................. 96

X
Abstract
The unreliability of the electrical grid has been a major challenge in Jigjiga City. The existing
substation has encountered frequent power interruption problems. The interruptions are caused
mainly by the short circuit (SC) and earth fault (EF).There are also planned outages for operation
and maintenance and the interrupted electric power distribution reduce power consumption
,affect daily activity and drags the modern life style. It impacts societal development and
individual income. Basically, a stable and reliable electric power supply system is an inevitable
pre-request for the technological and economic growth of any nation. And the utilities must
strive and ensure that the customers are getting the reliable power supply. In this thesis the
reliability of jigjiga distribution system was done on both 15 and 33kv system assesses the
performance of existing system. More over predictive reliability analysis for the future system
considering optimal network reconfiguration to be annualized. For predictive reliability
improvement, Matlab modeled fuzzy logic based optimization techinques has been applied,
which delivers optimal number and placement of switches on the selected feeder of the
distribution system. Among 8 feeders of the distribution system under cause study, 33kv feeder2
is selected as a test feeder as it has higest frequency and duration of interruption and also its cost
of interruption is very high compared to other feeders .As the result of network reconfiguration
in strategic places reduce the outage time in case of abnormal conditions and improve the
reliability of the network and the optimal placement of switches and optimal tie switch of near by
healthy feeder The SAIDI and SAIFI are reduced from 368.5failure/cust,yr to 166.37
failure/cust.yr and 621hr/cust.yr to 269.94hr/cust.yr .from the test results, the proposed technique
present better interruption cost in comparation with that of existing system. 1,609,907.89ETB to
840,145.5 ETB Birr reduction along with improvement in reliability indices. This method is
implemented and tested on Togowuchale feeder of jigjiga town distribution network, Results of
both initial and optimal configuration of expected energy not supplied (EENS) was improved
from 1343MWH./yr to 701 MWh./yr this alternative which gives low system average
interruption index (SAIDI), system average interruption frequency index(SAIFI) and customer
average interruption duration index(CAIDI) was being assessed and considered. The reliability
analysis indicated that of jigjiga distribution feeder was further improved installation
/reconfigured/ of the radial network in strategic place by using fuzzy logic artificial intelligence
method.
Keyword: Network Reconfiguration, fuzzy logic optimization, distribution system improvement

XI
 CHAPTER ONE

1. INTRODUCTION
1.1 Back ground
The basic function of an electric power system is to supply customers with electricity;
modern society demands that electrical energy should be as economical as possible
with a reasonable degree of continuity and quality. The continuity of energy supply
can be increased by improved system structure, increased investment during the
planning phase, operating phase or both over investment can lead to excessive
operating costs. A power system has a number of generating stations of different
types interconnected by a system of transmission lines, sub transmission line and
distribution networks to supply different types of loads to various consumers. The
distribution systems in practices have a single circuit main feeder and are radially
configured [1]. The radial distribution system is widely used because of its simple
design, generally low cost and supportive protection scheme. This configuration
suggests from a reliability point of view that all components between a load point and
the supply point must work and therefore poor reliability can be expected as the
failure of any single component causes the load points to be disconnected .Reliability
of an electric power distribution system is defined as the ability to deliver
uninterrupted service to the end customers. Reliability is an integral component of
modern power system design, planning and control [2]. The reliability in distribution
system can be improved by network reconfiguration, which is accomplished by
closing normally open switches and opening normally closed switches, these switch
play important role in reducing interruption duration in the event of a system failure.
Two types of switch are normally installed along the main feeders and laterals
sectionalizing switches (normally closed switch) and tie switch (normally open
switch) the former is a device that isolates a faulted part from the system so that the
healthy part can still be electrically supplied the later is adevice that recovers loads
that has been disconnected by transferring some of the loads to other supporting
distribution feeders without violating operation and engineering constraints.

1
The techniques used in distribution system reliability evaluation can be divided into two basic
catagories-analytical and simulation methods. The difference between these methods is in the
way the methodology uses the input data in which the reliability indices are evaluated. Analytical
techniques represent the system by simplified mathematical models derived from mathematical
equations and evaluate the reliability indices using direct mathematical solutions. Simulation
techniques, estimate the reliability indices by simulating the actual process and stochastic
behavior of the system. Therefore, this methods treates the problem as a series of real
experments conducted in simulated time .Reliabltiy improvement strategy has to be developed
for each utilities depending up on their requirements also with strategies, also with strategies ,
outages mitigation techniques in distribution system and affect the distribution system analysis
and these techniques includes addition of protective devices (recloser and fuses) and
switching(manual and automated switches), system reconfiguration , feeder upgrading and
integration of distributed generation, on the other hand,non-electric mitigation techniques donot
have any impact on other engineering analysis tools and can be evaluated solely with reliability
studies and these techniques includes vegetation management, installation of lightning and
animal guards[3]. The impact and the efficiency of the mitigation techniques could be assessed
through quantitative reliability evaluation of the distribution system such as measuring of the
past performance and predicting the future performance. This thesis would be focuses on the
evaluation of the distribution system of jigjiga town distribution system. This is assess whether
there is already a need for the switching scheme to be reconfigured or not. The computed
reliability indices such as system average interruption duration index (SAIDI), system average
interruption frequency index (SAIFI), customer average interruption duration index (CAIDI) and
Average Service Availability Index (ASAI) with the Average System Curtailment Index (ASCI)
will be used as a basis on why the protective scheme of the system is going to be reconfigured.
Maintaining a specific level of reliability is very important so that the customer satisfaction is
high as possible and the efficiency of the system will improve and reliability evaluation can be
used to provide a measure of the overall ability of a power system to perform its intended
function. The concept of reliability can be subdivided into the two main aspects of system
adequacy and system security [4]. System security relates to the ability of the system to respond
to disturbance arising within the system. System adequacy relates to the existence of sufficient
facilities within the system to satisfy the customer demands within the system operating
constraints

2
The other main relevant issue for the analysis of quality of power supply is the fact that the level
of system reliability is interrelated with the economic aspects. Reliability costs are used for
reviews and request for rate increase. The economic analysis of system reliability can be a very
useful planning tool in determining the capital expenditures required to improve the real value of
additional (and incremental) investment in to the system [3]. It is necessary to combine the
reliability criteria with certain cost considerations.
1.1 .1 Overview of Jigjiga town Distribution system
Jijiga city is the capital of Ethiopian Somali national regional state. Jijiga is located about 630km
to the east of Addis Ababa, about 105 km east from the state city of Harar. The geographic
coordinates are 90 20‟ North and 420 56‟ east.
The distribution feeders in the Jigjiga town are madeup of 33 kV and 15 kV primary feeders. The
33 kV feeders have four outgoing lines, these are: fafen (feeder1), Togowuchale (Feeder2)
Kebribeyah (Feeder3) and chinaksen (Feeder4) and 15 kV feeders have four outgoing lines such
as Feeder1, Feeder2, Feeder3 and Feeder4. These feeders are connected to a total of more than
400 distribution transformers; most of them are pole mounted, for further step down to 380 V
three-phase and 220 V single-phase for secondary distribution purpose.

Fig 1.1 Single line diagram of jigjiga substation

3
 1.2 Problem of statement
Electricity networks are a critical part of our energy infrastructure and power utility company
has the responsibility to ensure that are developed consistently and in a manner that meets future
demands of existing and new customers. The process of network development should be directed
towards a long term vision aligned with the expectations of the existing and new customers.
The main existing problem electric power utilities in developing countries today is that the power
demand is increasing rapidly where as supply growth is constrained by scarce resources,
environmental problems and other societal concerns. This has resulted in a need for more
expensive justifications of the new system facilities, improvements in production and use of
electricity. Currently in jigjiga city there are frequent interruption which affects both the utility
and customers due to this unplanned, unwanted and long power interruptions, all residential,
commercial and industrial customers are victoms of the problem. Especially for industries and
factories and power reliability become a matter of fatality and if these challenges for the last
many years and it will continue to be a challenge in the future if an amicable solution is not
found to the problem. This unreliable power supply does not only slow down or damages
production or results in shutting down of the plant but also leads to equipment damage,
additional maintenance and the industry reputation for the quality of the product. Achieving
reliability of power supply in jigjiga city has a major challenge. The event of aging of the
electrical network, severity of the damages, repetition of repairs,and a lack of strategic
maintenance trends, under standard long line constraction and environmental condition of the
region all contributing to the decline to the distribution network.
An intelligent placement of sectionalizers and tie switches in the distribution feeder has
significant impact on reliability improvement and this would be further assessed along with
outage mitigation techinques for the distribution system. Therefore, in this thesis the distribution
system of jigjiga town has been considered, particularly focusing jigjiga distribution system.

4
 1.3. Objectives of the thesis
. 1.3.1 General objective
The main objective of this thesis is to assess and evaluate the methods of reliability
improvement of jigjiga town distribution system and utilization of artificial intelligence
optimization (AI) or fuzzy logic techinques to improve system reliability by making the
optimal configuration of the existing distribution network. The main goal of this master
thesis work is to develop, evaluate and assesses a frame work for reliability
improvement of distribution system in jigjiga town.
1.3.2 Specific objective
The specific objectives of this thesis are

 To investigate the reliability of the power distribution system in Jigjiga City

 To model a feeder in the distribution system and analyze it using an appropriate software
for reliability improvement

 To develop an optimal configuration for the network and evaluate its performance and
reliability improvement of jigjiga distribution system by simulating with etap software

 To develop and evaluate effective power outage mitigation strategies, including

proposing solutions to minimize the occurrence and impact of power outages in the
future, and draw relevant conclusions based on the findings.

 Comparation of the results obtained from the existing and optimal configuration in order
to determine the extent of reduction of reliability indices.

1.4 Significance of the Study

The main importance of this thesis is to reduce reliability indices for improving system
performance in a given distribution system when a network reconfiguration algorithm is
implemented, reliability indices would be reduced.
Generally expected importance of this thesis paper study is:
 To indicate the influence of power interruption on economy of both customers
and utility.
 It is regarded as a tool for the planning senior electrical engineers to assure a fair
level of service continuity and to select from a variety of expansion plans that
are cost –effective when compared to system investment and loss costs.

5
  To apply fuzzy logic optimization (FL) systems for improvement of system reliability.
 To show the feasibility of this distribution feeder network reconfiguration on
reliability improvement with comparing other approaches.
Thus, this thesis is to present a methodology for network reconfiguration with objective of
identifying, in normal operating conditions, the optimal configuration for the distribution
system that minimizes interruption costs while also enhancing reliability.
1.5 Scope of the Study
The scope of this thesis paper Focused on:-
 Only distribution system reliability would be analyzed, neither generation system nor
transmission system reliability evaluations will be conducted.
 Using the ETAP software simulation on the distribution network in order to minimize the
customer reliability indices based on fuzzy expert system output
 Only the effect of the feeder line with regard to reliability will be considered. The effect
of distribution system components would not be considered
 Using fuzzy logic optimaization with matlab to analyze the optimal placement of switch
and tie switch for reliability improvement
 Only the effect of the feeder line with regard to reliability would be considered. The
effect of distribution system component would not be considered.

6
 CHAPTER TWO

2. LITERATURE REVIEWS
2.1 Distribution System Reliability
Electric power is a vital element in any modern economy. The availability of a reliable power
Supply at a reasonable cost is crucial for the economic growth and development of a country.
Electric power utilities throughout the world therefore endeavor to meet customer demands as
economically as possible at a reasonable service of reliability. To meet customer demands, the
Power utility has to evolve and the distribution system have to be reconfigured, operated and
maintained accordingly. An analysis throughout the world shows that around 90% of all
Customer reliability problems are due to the problem in distribution system, hence, improving
Distribution reliability is the key to improving customer reliability [3]. There are two main
approaches applied for reliability evaluation of distribution system, namely. Monte Carlo
Simulation method based on drawings from statistical distributions and Analytical Methods
based on solutions of mathematical models. The Monte Carlo techniques are Simulation Method;
however, it is normally time consuming due to large number of drawings necessary in Order to
obtain accurate results. The usual method of evaluating the reliability indices is an Analytical
approach based on failure modes assessment and the use of equations for series and Parallel
networks. The common indices used for evaluation are the expected failure rate (λ), the Average
outage time(r), and the expected annual outage time (U) which is adequate to the Simple radial
system.

Tempa Dorji [3] in paper entitled “Reliability assessment of distribution systems including a case
Study on Wang due Distribution system in Bhutan”, in this paper he used a probabilistic
Approach to estimate different improvement techniques to increase the reliability of distribution
System. Power system reliability is defined as ability of electrical power system to supply the
System load with reasonable continuity and quality of supply. Major subdivisions of power
system reliability are system adequacy and system security as shown in Figure 2.1. The term
adequacy relates to the existence of sufficient facilities within the system to satisfy the
Consumers load demand and system operational constraints. Thus, adequacy majorly deals with
Static conditions and not the dynamic and transients of power system. Security is associated with
System dynamics and disturbances in the system. Security is therefore related to the response of
the system to perturbations it is subjected to. The concept of power-reliability is extremely broad
and covers all aspects of the ability of the system to satisfy the customer requirements.

7
There is a reasonable subdivision of the concern designated as system reliability which is shown
in Figure.2.1

Fig 2.1 system Reliability subdivision

Reliability assessment and improvement of electric power distribution system has been studied
by different researchers using different approaches (methodologies). Some of these are reviewed
as follows:

Chaichan Pothisarn etal [5] in paper makes analysis of reliability with the inclusion of DG. They
used DIGSILENT to analyze the impact of reliability with the installing DG into the distribution
system. They used only the size of load, and the distance of load which is factors able to impact
to SAIFI, SAIDI, and interruption cost but they didn‟t consider the branch of the main section as
the input variable.

Bordin Bordeerath [6] in his paper entitled “reliability evaluation of composite Systems with
renewable energy sources”, presented simple effective methodologies for accelerating Monte
Carlo simulation (MC) in power system reliability assessment and also studied the impact of
correlation between generated renewable power and loads towards estimated reliability indices.

As a case study of Harer town distribution system, Mandefro Teshome (2019) [7] studied
"power distribution system reliability assessment and improvement." His work shows that by
employing reliability improvement strategies that are financially justified, the system's reliability
has been significantly increased. The main drawback in this work is that the distribution system
was reconfigured to increase reliability by placing tieswitches and recloses in random places.
Xuebei Yu [8] he discusses in this thesis on the reliability enhancement in power distribution
systems. An algorithm was built to realize optimal relay setting for several test systems.

8
Both overcurrent relays and distance relays and their coordination were considered in the
optimization problem. In addition, fuse saving scheme was applied to avoid long-time power
outage. The cost benefited due to this relay setting was not evaluated.

Benyamin Moradzadeh, [9] in his paper entitled “Optimal Distribution Reconfiguration and
Demand Management” has shown how it is possible to find optimal state of the switches for
which the observed network has smaller total losses and smaller total customer interruption
costs. This approach can also be applied to weakly-meshed system be removing the radially
constraint for particular cycles in the system.

Desta Tegegne Asradew, 2021 [10] he proposed Fuzzy Logic Controller optimization for benefit
of switching allocation using fuzzy logic optimization technique, the paper shows it optimal
section switch allocation using fuzzy logic technique by considering three Linguistic Inputs
variables i.e. 1. Number of branches connected to each bus (NBC) 2. Connected, Load to the
line segment (CLS) & 3. Section line Length (SLB) of main line segment, this paper is more
recently and good compared to above listed thesis paper, because the paper used artificial
intelligence for the selected optimal placement of sectionalize switch but the main gap of this
thesis is not include environmental effect of the region and optimal placement of tieswitch in the
network reconfiguration in cause of that the improvement of the solution effectiveness is below
10%.

In general, it is possible to say that the aforementioned reliability analysis, evaluation, and
improvement include the distribution system's reliability but do not do so well because they
demonstrate the addition of devices but not in an optimal number and locations. In this thesis
work, the problem of tieswitch and sectionalize switch placement in a real distribution network
was considered for minimizing SAIFI, SAIDI, EENS and ECOST in order to achieve a minimum
level of reliability.

9
 2.2 Distribution system Reconfiguration[11]
All distributions of electrical energy are done by constant voltage system. It is important to
distribute power to consumers at constant voltage level.

I. Radial main distribution systems: There are several configurations of distribution

systems. Most distribution circuits are radial. Radial circuits have many advantages :


Easier voltage control


Easier prediction and control of power flows

Lower cost

A radial system is connected to only one source of supply and is exposed to many interruption
possibilities. The most important of which are installing the system is relatively cheap, the
maintenance is easy, the distribution system needs least amount of conductance etc

Figure 2.2 Radial Distribution Systems

This is the simplest distribution circuit and has the lowest initial cost. However, it suffers
from the following drawbacks.

The end of the distributors nearest to the feeder point will be heavily loaded.

The customers at the distant end of the distribution would be subjected to serious voltage
fluctuations when the load on the distributor is changed.
The customers are dependent on a single feeder and single distributor. Therefore, any fault on the
feeder distributor cuts off supply to the customers who are on the side of the fault away from the
substation.

10
 
.
II. Ring main distribution system

In this system the primaries of the transformers form a loop. The loop circuit starts from the
substation bus bars, make a loop through the area to be served, and returned to the substation.
The figure below show the single line diagram of ring main system for a.c distribution where a
substation supplies to the closed feeder LMNOPQRS. The distributors are tapped from M, O, &
Q of the feeders through distribution transformer.[11]

Figure 2.3 Ring Feeders with Two Energy Sources

The ring main system has the following advantages:

• The system is very reliable at each distributor is fed from two feeder

• There are very less voltage fluctuation at customer terminals.

In case of fault in any section of feeder the continuity of supply is maintained. For instance if the
fault occurs at point F of section SLM of the feeder. The section SLM of the feeder can be
isolated for repair and at the time continuity supply to the customer is maintained through the
feeder SRQPON.

11
 III. Interconnected distribution system
When the feeder ring is energized by two or more than generating stations or substations, it is
called inter connected system [11] The inter-connected system has the following advantages
 It increases the service reliability
 Any area fed from one substation during peak load hours can be fed from the other
substation this reduce reserve power capacity and increase efficiency of the system

Figure 2.4 Interconnected Systems

2.3 Markove chain Model
A markov model is quite popular in the quantitative reliability analysis, and that is suitable to
give fair idea about reliability analysis principle. On the basis of markov models formulas can be
developed and used to calculate the reliability of the radial distribution network, it basically
operates with two central concepts
 Failure frequency(λ)
 Repair time(r)
It is assumed for example that a component-wise reliability can only be in one of the following
conditions; Condition 1: Component is in the function (in); Condition 2 Component is in repair
out). (This is illustrated in two state model diagram in figure 2.5 represented by “0‟‟ (component
in failed state) and “1‟‟ (component is in a normal state).

12
The two main approaches used are analytical and simulation. The vast majority of techniques
have been analytically based and simulation techniques have taken minor role in specialized
applications [12] The main reason for this is because simulation generally requires large amount
of computing time, and analytical models and techniques have been sufficient to provide
planners and designers with results needed to make objective decisions. Analytical techniques
represent the system by a mathematical model and evaluate the reliability indices from this
model using direct numerical solutions. A Markov model is quite popular in the quantitative
reliability analysis, and that is suitable to give fair idea about reliability analysis principle. On the
basis of Markov models, a simple formula can be developed that can be used to calculate the
reliability of the radial distribution network. The method is called like duration-frequency
technique, and the starting point is the failure of the individual component. In a so-called
stationary Markov process, it basically operates with two central concepts Failure frequency (λ)
Repair time (r). It is assumed for example that a component-wise reliability can only be in one of
the following conditions; Condition 1: Component is in the function (in); Condition 2:
Component is in repair (out). This is illustrated in two state model diagram in figure 2.5
represented by 0(component in failed state) and 1(component is in a normal state).

In condition µ=1/r Out condition

Figure 2.5 Transition Diagram of Component
States λ = fault frequency
m = mean time to failure
µ=repair frequency
r = mean time to repair

13
The figure 2.6 illustrates expected functional and outage time for a component (so called state
cycle).The system can be represented by Markov process and equations developed for the
probabilities of residing in each state in terms of state transition rates are as follows

Figure 2.6 Average State Cycle

Definitions: Mean-time-to-failure (MTTF): MTTF is described as the time to failure counted

from the moment the component begins to operate to the moment it fails. Figure.2.6 shows
typical time to failure and time to repair cycle of a component. Mean-time-to-Repair (MTTR):
MTTR is the time counted from the moment the component fails to the moment it is returned
back to an operable condition. Failure rate: The failure rate is the reciprocal of the mean time to
failure.
2.3.1 Series System

All distribution systems in jigjiga are basically designed, constructed and operated in radial
system. A radial system basically consists of set of series components like; breakers, lines,
switches, transformers and at the end of „Customers‟. In the series structure all components must
be intact for the system to function, while in the parallel structure both must fail for the system to
stop functioning

14
 2.3.2 Parallel System

In this case the failure modes of the load point involve overlapping outages, i.e. two or more
components must be on outage at the same time in order to interrupt a load point. It is assumed
that the failures are independent and that restoration involves repair or replacement, the
equations used to evaluate the indices of the overlapping outage.

2.4 Power Quality, Reliability and Availability

Power quality is an ambiguous term that means many things to many people. From a consumer
perspective, a power quality problem might be defined as any electric supply condition that
causes appliances to malfunction or prevents their use. From a utility perspective, a power
Quality problem might be viewed as non-compliance with various standards such as RMS
voltage or harmonics. Perfect power quality is a perfect sinusoid with constant frequency and
amplitude. The power Quality is affected when a voltage waveform is distorted by transients or
harmonics changes its amplitudes or deviates in frequency [13] Customer interruptions are power
quality concern since it reduces voltage to zero.
Reliability is primarily concerned with customer interruptions and is therefore a subset of power
quality. Availability is defined as the percentage of time a voltage source is uninterrupted. The
hierarchy of power quality, reliability and availability is shown in figure 2.7. The figure indicates
availability is a subset of reliability and reliability is a subset of power quality. Power quality
deals with any deviation from a perfect sinusoidal voltage source. Reliability deals with
interruptions.

15
.

Figure 2.7 Hierarchies of Power Quality, Reliability and Availability

2.4.1 Power Qualities
There are different defintions for power quality [14]
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
This depends up on the end user. according to end user of view, it is defined as any power
problem manifested in voltage, current or frequency deviation that result in failure or mis
operation of customer equipment
In IEEE dictionary, power quality is defined as “the concept of powering and grounding
sensitive equipment in a matter that is suitable to the operation of the equipment”
2.4.2 Reliability

Distribution reliability primarily relates to equipment outages and customer interruptions. In

normal operating conditions, all equipment (except stand by) is energized and all customers are
energized. Schedule and unscheduled events disrupt normal operating conditions and can lead to
outages and interruptions. The unscheduled events are caused either due to human error or due to
equipment failures. The schedule events are meant for periodic maintenance of the equipment
and shall be notified in advance to the customers from definition of reliability, power outages are
one of the measures of reliability performance. Unavailability of power can be reduced in two
ways: 1) by reducing the frequency of power outages or 2) By reducing the outage time.

16
 2.4.3 Availability

Availability is the probability of something being energized and Unavailability is the probability
of not being energized. It is most basic aspect of reliability and is typically measure in percent or
per unit. Unavailability can be computed directly from interruption duration information. If a
customer experiences 10 hours of interrupted power in a year, unavailability is equal to 10 /8760
= 0.110% (8760 hours in a year). Then availability is equal to 100% - 0.11% = 99.89%.

2.5 Reliability Improvement Methods

The main purpose of reliability data quantification and information extraction is to take
reliability improvement measures. Reliability of distribution system can be improved by
increasing distribution system protection, decreasing equipment failure, system automation,
installment of reclosing and switching devices and system configuration. System automation
improves short interruption duration. Restoration time of momentary outage event will be small.
Hence the duration that an outage will last will be diminished. Similarly system configuration
produces effective improvement in reliability. During occurrence of fault that lasts long,
distribution system can be configured into set of network topologies. Distribution network will
have alternative supplying network after reconfiguration.

2.5.1 Application of Distribution network reconfigurations

Distribution network reconfiguration is an important tool to reduce the reliability indices,

system‟s power loss, and to do the load balancing in distribution system. This operation is to
transfer load from one feeder to another, which will significantly improve the overall system
operating conditions. Configuration must be done from time to time, since the line distribution
shows different characteristics as each distribution feeder consists of residential, commercial and
industrial type, load. Some sections of the light distribution system loaded at specific times of the
day and many loaded at other times.
In a distribution system, each feeder has a different mixture of commercial, residential, and
industrial type loads. These load types have different daily patterns which make the peak load of
feeders occur at different times. In normal operating conditions, part of loads can be transferred
from heavily loaded to relatively less heavily loaded feeders by network reconfiguration.
Distribution feeders contain number of switches that are normally closed (sectionalized switches)
and switches that are normally open (tie switches).

17
Distribution network reconfiguration is the process of altering the topological structure of
distribution network by closing the open/ close status of sectionalizing and tie switches
When the operation conditions change, network reconfiguration is performed by the
opening/closing of the network switches under the constraints of transformer capacity, feeder
thermal capacity, and voltage drop and radially of the network. The distribution network
reconfiguration is a complex combinatorial optimization problem. This is because there are
multiples of constraints which must not be violated while finding an optimal or near optimal
solution to that problem.
When system reliability needs to be evaluated more closely .multiple failures shoud be taken in
to account and actual reconfiguration for restoration analysis should be carried out as part of this
assessment. In this work a research effort has been devoted to developing a reconfiguration for
restoration program.
The distribution reconfiguration studies could be divided into following.

 Capacitor placement
 Network reconfiguration
 Voltage and reactive control
 Optimal placement of switch
 Distributed generation placement
 Service restoration
In this research, optimal placement of sectionalizer switch and tie switches is the major
concern. Service restoration aims at minimizing service down time to out of power loads during
post fault conditions. the location of protective components on the distribution system and their
response to failures can have an important impact on reliability indices. With Protective devices
together with switches will be referred to as sectionalizing devices. Sectionalizing devices are
used to group components .thus a sectionalizing devices is any component that is designed to
break the electrical circuit topology

18
 2.6. Methods of Distribution Network Reconfigurations

2.6.1 Fuzzy Logic Method

In [15] described a calculation strategy that allows determining the optimal number and
placement of sectionalizing switches in MV radial distribution networks, in correspondence to
technical, regulatory and economical aspects. A formulation that takes into account the
investment, maintenance and power interruption costs has been developed, seeking for a
reduction in total costs while taking care of the regulatory and technical aspects. A multicriteria
optimization procedure allows incorporating in the calculating process various quality indicators
which can be either global or individual indexes. This way of formulation makes the proposal
flexible as well as applicable to allow including aspects that were not considered in this papers.
T. Juhana Hashim etal [16] discussed the implementation of a fuzzy logic based coordinated
voltage control for a distribution system connected with distributed generations (DGs). The
connection of DGs has created a challenge for the distribution network operators to keep the
voltage in the system within its acceptable limits. Intelligent centralized or coordinated voltage
control schemes have proven to be more reliable due to its ability to provide more control and
coordination with the communication with other network devices. In [17] proposed a new type of
Evolutionary search methodology is proposed for determining the minimum loss configuration
of a radial distribution system. To improve the performance of evolutionary programming a
fuzzy controlled evolutionary programming (FCEP) based on heuristic information has been
proposed. The designed mutation fuzzy controller adaptively adjusts the mutation rate during the
evolutionary process. This method reduces combinatorial explosive switching problem into a
realizable one and reduces the switching combination to a few number. A mutation fuzzy logic
controller is developed to speed up the evolutionary process by adaptively adjusting the mutation
rate.The following advantages of the fuzzy systems are to be selective [18]

 Capacity to represent inherent uncertainties of the human knowledge with linguistic

 Simple interaction of the expert of the domain with the engineer designer of the system
 Easy interpretation of the results, because of natural rules representation
 Easy extension of the base of knowledge through the addition of new rules and
Robustness in relation of the possible disturbances in the system.

19
 Figure 2 .8 Structure of Fuzzy Logic System

2.6.2 Heuristic Optimazation methods

Heuristic and meta-heuristic technique is the randomized searching procedures. It described that
swarm optimization techniques and evolutionary techniques.
2.6.3 Artificial Neural Network
This shows the strategy of feeder reconfiguration to reduce the power loss by artificial neural
network (ANN). This approach developed is basically different where are the aspects that the
load transfer and the corresponding load flow solution during the search process are not required.
and this artificial intelligence used to test various load patterns which minimize the loads under
given conditions.
2.6.4 Tabu Search Method
M.Gkumar, P.sangamewara Raju, P.Venkatesh, P.Ramanjaneyulu Reddy [19] proposed a tabu
search algorithm for the reconfiguration of distribution system for the minimizing the real power
loss. A new characterization of the neighbor hood structure avoids the exploration of an
excessive number of configurations, thus reducing the computation effort with out reducing the
quality generated configuration and one of the main characterstics of this aligorithm is use short
term or long term memory.

20
 2.6.5 Simulated Annealing Method
Young –Jae jeon .jae-chulkim [12] proposed an efficient algorithm for loss minimization by
using an automatic switching operation in large scale distribution system. Simulated annealing is
particularly well suited for large combinatorial optimization problem so it can avoid local
minima by accepting improvement in cost and it requires a meaningful cooling schedule, special
strategy which makes use of the property of distribution system in finding optimal solutions.
2.6.6 Ant colony optimization Method
R.srinivasa Rao,(etal) [20] a new population based artificial bee colony algorithm (ABC) was
proposed to solve the network reconfiguration problem in a radial distribution system. The main
advantage of ABC aligorithm is that it does require external parameters such as cross over rate
and mutation rate etc, as in cause of genetic algorithms. Differential evolution and other
evolution and other evolutionary algorithms and these are hard to determine in prior. The other
advantage is that the global search ability in the algorithm is implemented.

21
 CHAPTER THREE
3. METHODOLOGY, DATA COLLECTION & ANALYSIS

To evaluate and assess the reliability of the distribution system the historical statistics data of
power interruption frequency and duration is required .for this thesis data are collected from EEP
jigjiga substation monthly report data. From jigjiga substation of the four years (2011, 2012,
2013 and 2014) data are taken for selected eight feeders. In the report every permanent and
temporary power interruption, duration of each power outages with their cause in each month are
included. Reliability analysis of electrical distribution system is considered as a tool for the
planning engineer to ensure a reasonable quality of service and to choose between different
systems expansion plans that cost wise were comparable considering system investment and cost
of losses. There are two main approaches applied for reliability evaluation of distribution system
namely simulation approaches and analytical approach [5] The usual method of evaluating the
reliability indicies is analytical approach based on failure modes assessment and the use of
equations for series and parallel networks. The common indices used for evaluation are
 The expected failure rate(λ)
 The average outage time(r),
 The expected annual outage time (U)
Which are adequate to the simple radial systems, in the distribution system whether the networks
are radial or meshed, mostly they are operated in radials since it is simple to assess

3.1 Distribution System Reliability Indices

Reliability indices are statistical aggregation of reliability data for a well-defined set of loads,
components or customers. Most reliability indices are average values of particular reliability
characteristic for an entire system. Operating region, substation, service terriotory, or feeder.
This section presents some basic reliability indices that were used in this thesis as general terms
for quantitative measure of reliability. Some of these indices could be evaluated using computer
program. They are therefore defined as follows

22
 3.1.1 Load Point Indices

The basic distribution system reliability indices of a load point p are average failure rate λp,
Average outage duration rp and annual outage time U. these three basic indices are calculated
using the principle of series system and given by

n
λ𝑝 = ∑ λi …………………………………………………………... (3.1)
i=1
n
U = ∑ λ𝑖r ……………………………………………………………. (3.2)
i=1

r𝑝 = U /λ𝑝 = λ𝑖𝑟𝑖/ ∑ λi ……………………………………………………………… (3.3)

Where n = number of outage events affecting load point p

λi = failure rate of component i (failure/yr or, in short,
f/yr) ri= repair time of component i (hr.)
3.1.2 System Reliability Indices
The electric utility industry is moving toward a deregulated, competitive environment where
utilities must have accurate information about system performance to ensure that maintenance
dollars are spent wisely and that customer expectation are met. To measure system performance,
the electric utility industry has developed several measures of reliability. These reliability indices
include measures of outage duration, frequency outages, and system availability
[4]. The most common distribution indices include the system average interruption duration
index(SAIDI), customer average interruption duration index(CAIDI), system average
interruption frequency index(SAIFI), Momentary average interruption frequency index(MAIFI),
customer average interruption frequency index(CAIFI) customer interrupted per interruption
index(CIII), and the average service availability index(ASAI)
1) The System Average Interruption Frequency Index (SAIFI) (interruption/yr.)
The index represents the average number of sustained interruptions experienced by a
customer in a unit time (generally per year). The definition of service area is flexible in the sense
that the number of customers and the interruptions experienced by them changes with the
definition of the enclosed area. For instance, a feeder SAIFI indicates the average number of
interruptions a customer serviced by the particular feeder would experience in a year. Similarly
SAIFI reported for a substation or a distribution system encloses the total customers in the
service area. The system average interruption frequency index is given in Equation (3.4)

23
 SAIFI= = 3.4
Where, λi is the failure rate and Ni is the number of customers at load points i.

2) System Average Interruption Duration Index (SAIDI, h/yr.)

The index indicates the average time a customer has an interruption during a time cycle (1 year).
It is usually specified in customer minutes or customer hours of interruption /year. SAIDI
(system average interruption duration index) is the average interruption duration per customer
served as given in Equation (3.5). It is determined by dividing the sum of all customer
interruption durations during a year by the number of customers served.

SAIDI = = 3.5
Where Ui the annual outage time and Ni is the number of customers at load point i.
SAIDI can be improved by reducing the number of interruptions or the duration of the
interruptions.
3) Customer Average Interruption Duration Index (CAIDI) (h/interruption).
Customer Average Interruption Duration Index (CAIDI) is the average interruption duration for
those customers interrupted during a year. It is determined by dividing the sum of all customer
interruption durations by the number of customers experiencing one or more interruptions over a
one year period. The index is the ratio of SAIDI to SAIFI as given in Equations (3.6). It
represents the average time taken to restore service to the customers when a sustained
interruption occurs.

CAIDI = = = 3.6

Where λi is the failure rate, Ui is the annual outage time and Ni is the number of customers at
load point i.
4) Average Service Available Index (ASAI)

The average service availability index (ASAI) gives the fraction of time the customer has power
during the reporting time. Higher ASAI values reflect higher levels of reliability. Equation (3.7)
is used to calculate the value of ASAI for a given service area.

ASAI= 3.7

24
 5) Average Service Unavailable Index(ASUI)

ASUI= 1-ASAI= 3.8

6) Expected Energy not supplied index (EENS)

EENS= ∑Li ∗Ui 3.9

Where Li and Ui respectively are the average connected load and the average annual outage time
at load point i.
(7) Average Energy not supplied index (AENS) (KWh/yr.)

AENS = = 3.10

The first five indices are customer-oriented indices and the last two are load and energy-oriented
indices. These indices can be used not only to assess the past performance of a distribution
system but also to predict the future system performance

3.2 Power Outage Cost Evaluation

Power outage cost is the cost of customers and utility due to power interruption as discussed in
chapter one. A variety of methods can be utilized to evaluate customer impacts due to
interruption. These methods can be grouped, based on methodological approach used, into three
broad categories, namely; various analytical evaluations, case studies of black outs, and customer
surveys. Even if a single approach has not been universally applied, customer surveys are the
main approach to determine specific information for outage cost evaluation purposes. In practice,
the cost of interruption from the customer‟s perspective is related to the nature and degree to
which the activities interrupted are dependent on electrical supply. In turn, this dependency is a
function of both customer and interruption characteristics. Customer characteristics include type
of customer, nature of customers‟ activities, size of operation and other demographic data,
demand and energy requirements, energy dependence as function of time of day, etc. Interruption
characteristics include duration, frequency, and time of occurrence of interruptions;

25
 3.3 Impacts of Improvement Techniques and Protection System on
Reliability

A properly coordinated protection system is vital to ensure that an electricity distribution

network can operate within predetermined requirements for safety for individual items of
equipment, staff and public, and the network overall. Suitable and reliable equipment should be
Install on all circuits in electrical power system. A better over-current protection scheme can
reduce number of customers affected by temporary and permanent faults. So, historical data can
be used to quantify improvements and predict the best locations for sectionalizing devices for
reliability improvements. Adding numbers of re-closer at optimal locations can reduce SAIFI
and SAIDI but it should be economically viable. The location and installation of number of Auto
re-closer, Switches, Load Break Switches and sectionalizes either manual or automatic helps to
reduce fault rate, repair time and sectioning time which directly reduces the impacts on the
system when fault occurs.

3.4 Reliability Evaluation

The ultimate goal of reliability analysis should be to answer questions like: Is the system reliable
enough? Which schemes will be effective? And where high capital should spent to improve the
system? Reliability in power system can be divided in to two basic aspects; historical and
predictive. The predictive reliability is then followed to predict the changes in reliability
measures after a change in system configuration or any improvement strategy is planned to be
implemented
3.4.1 Historical Reliability Assessment

Two approaches to reliability evaluation of distribution systems are normally used, namely,
historical assessment and predictive assessment [9]. Historical assessment involves the collection
and analysis of distribution system outage and customer interruption data. It is essential for
electric utilities to measure actual distribution system reliability performance levels and define
performance indicators to assess the basic function of providing cost-effective and reliable power
supply to all customer types. Historical assessment generally is described as measuring the past
performance of a system by consistently logging the frequency, duration, and causes of system
Component failures and customer interruption.

26
 3.4.2 Predictive Reliability Analysis

The predictive reliability analysis (PRA) estimates the future performance of the distribution
system based on system topology and failure data of the components. Its assessment provides a
basis to select the best options from several computing alternatives.
The main advantage of the PRA is its ability to forecast the reliability impacts of the system
expansion and to quantify the impact of the reliability improvement projects.
The improvement methods can be; load transfer between feeders, new feeder tie points, feeder
automation, replacement of aging equipment, changing the configuration of the system and
changing the connection scheme of feeders. A number of alternatives are considered and the one
which gives best result in term of reliability and minimum energy interrupted shall be chosen for
the future system
3.4.2.1 Reliability improvement Technique
These techniques have direct impact on the distribution system and affect the distribution system
Analysis.This technique include: addition of protective devices (re-closers and fuses) and
switching devices, system reconfiguration and feeder reconnecting.
(a) Reclosing devices, sectionalizes switches

Re-closers: They are protective devices which have two basic functions on the system reliability
and over current protection. Mostly they are mainly used for reliability reasons. Reclosing was
normal for all utilities since most lines were overhead and most temporary faults could be
cleared by the re-closer before the fuse operated. Most of the overhead faults are temporary in
nature any feeders with primarily overhead exposure should be protected by a reclosing relay on
its main circuit breaker. Placing a line recloser on a feeder will improve the reliability of all the
upstream customers by protecting them from downstream faults. However, the placement of the
re-closer should be such that a maximum benefit is obtained to improve reliability.

27
Automatic sectionalizes: Automatic sectionalizes automatically de-energize a faulted section so
that non-faulted line sections can be safely energized. It helps to overcome the coordination
problems of other protection devices like fuse and reclosers near substations. When the fault
current exceeds the pre-set value, the sectionalizers will open instantaneously in order to isolate
the fault to restore power back to non-faulted line segment customers. This also helps to save the
fuse from blowing which is also known as a “fuse saving” scheme.
Sectionalizing and Tie Switch: Sectionalizing switches are basically used to make or break
loads under normal conditions. They are used for immediate restoration of power to healthy part
of distribution system. Similarly, the tie switch is normally open switch applied for network
Reconfiguration. In case of power failure, distribution network is routed to alternative power
supplying path by the use of Tie switches. Both sectionalizing and tie switches are applied in
primary distribution network to reduce power interruption. The sectionalizing and Tie switches
can improve the SAIDI and CAIDI of a feeder.

Load Break switch: The installation of additional Load Break switch at strategic locations
would be more beneficial for making step restoration of the system at the time of fault. The
installation of load break switch have the potential to improve reliability by allowing faults to be
isolated and the customer service to be restored before the fault is repaired.
System configuration: A distribution system can be reconfigured by changing the location of
the normally open switches, effectively changing the allocation of the customers and the flow of
power for the effective feeders. It is not only improves reliability, but, also minimizes the losses
and the operation costs.

3.5 Distribution System Failures Cause Analysis

Outages in distribution systems caused by different factors significantly impact the reliability of
the system. It is important to investigate this outages.specifically; analysis of reliability
performance of the distribution system over the past year related to outages caused by different
factors is very useful for the utilities.The supply interruptions in this thesis are referred to as
outages or failures. These denote the state of a component when it is not available to perform its
intended due to some event directly associated with component.

28
The general definition of failure can be structured as figure 3.1 failure can be divided primarily
into damaging faults and non-damaging faults. Outages caused by damaging faults are usually
called permanent forced outages, while outages caused by non –damaging faults are categorized
again after the action of restoration into [21]

(i) Transient forced outages when the system is restored by automatic switching and the
outage time is negligible, and
(ii) Temporary forced outages when the system is restored by a manual switching or fuse
replacement.

Figure 3.1 Definitions of Failure

Long interruptions are often caused by damaging faults (permanent faults) and short
interruptions are often caused by transient faults. Furthermore, damaging faults can be separated
into two models of failure; Passive failure and active failure defined as follows:
An active failure of an item is the one which causes the operation of the protection devices
around it and results in the opening of one or more fuses and also active failure is a failure mode
that occurs when a component fails actively and causes interruption through its impact on other
components.
A passive failure is a failure that is not an active failure. The failed item (component) by an
active failure is consequently isolated and protection breakers are re-closed. This leads to service
restoration to some or all of the load points.

29
However, for the passive failure, service is restored by repairing or replacing the failed
component (or by re-closing a disconnect or and using another feeder for supply). The outage
time of a failure is made up of various items depending on the cause. Figure 3.2 shows two
different time sequences following active and passive failures. As can be seen in the figure, the
active failures can be restored by either repair or replacement, or by switching.

Figure 3.2 Total Times for Repair/Replacement of Passive Failure

3.6 Factors Influencing Power System Reliability

The factors influencing power system reliability can be broken down into four catagories. They
are component reliability, system configuration, environmental conditions, loading
3.6.1 Component Statistics

A power system consists of various components, such as lines, cables, transformers, breakers,
switches, reactors and capacitors bank.
Any single component outages may cause a partial or even the entire system outages.
The availablity of component functionblity is characterized by failure rates and repair or
replacement times.
Reliability of a power supply system can be assessed statistically, based on the information on
past system performance and use of historical data and records, or it can be assessed
stochastically, based on the information on the past system performance and use of historical
data and records, or it can be assessed stochastically, based on a predication analysis and use of
probabilistic variables and parameters.

30
 3.6.1.1 Repair failure rate
Hazard function h(t) is the conditional probability of failure in time interval „t’ to (t+dt), given
that there was no failure at time „t’ divided by the length of the time interval dt.

h(t)= 3.11

Where f (t) is probability density function and R (t) is reliability function. The cumulative hazard
function h (t) is the conditional probability of failure in the interval 0 to„t‟.

If the total number of failures during the time interval 0 to t.

H (t) = ∫ ℎ( t )d 3.12
0

Hazard function is also referred as hazard rate or instantaneous failure rate in reliability theory
it is very important for power system design engineers, repair and maintenance people. Hazard
rate is a function of time and it is a bathtub-shaped function shown figure 3.3 component failures
can be divided into aging failures and chance failures, aging failures is a conditional failure that
depends on the component history figure 3.3 shows a bathtub curve of a component failure rate
change during its life time. An aging failure can happen suddenly after components enter its
wear-out period. Figure 3.5 indicates that a component failure rate is not a constant failure.
Rate distribution is different from component type to component type .some expensive
component like transformers, come with a set of reliability data provided by the manufacturer,
including the component life cycle statistical distribution. Nowadays, the infant mortality period
of some expensive components is usually consumed by manufacturers so that when these
components are put into service they are already in reliable state.
The occurrence of failures of a component will vary during its lifetime and often being

31
 “Bathtub curve

7)

Figure 3.3 Bath-Tub Component‟s Life

1. The life of components follow three major periods:

Infant mortality period or decreasing failure rate period.

h (t) =-λt 3.13

2. Useful life period or constant failure rate period

h (t)= λ 3.14
3.Wear-out period or increasing failure rate period

h(t)=λt 3.15

Many components in power systems exhibit constant failure rate during their lifetimes,
this occurs at the end of the early failure region. Burn-in is performed by subjecting components
to stress slightly higher than the expected operating stress for a short period in order to weed out
the failure due to manufacturing defects.
The CDF of the life of a component is represented by:
F (t) =P [T≤T] 3.16
And
R (t) =1-F (t) 3.17
The reliability is

F (t) =P [T>t] 3.18

Thus, the hazard rate function is

3.19

32
 3.6.1.2 Repair time
There are no good models for repair time(down time),since the repair time for failed equipment
depends upon many things, such location, crewdispatch policy, different failed parts in a type of
component and so on . One of the common practices is to use the exponential model, which
assumes reparations are statistically independent events and repair time can be represent by the
global average. Historical data shows that the repair time is also affected by weather conditions.
Stormy conditions usually prolong the process of customer down time.
3.6.2 System Configurations
System configuration include various issues, including topology, transportation capacity,
protection/coordination schemes, and DG placement
3.6.3 Time-varying Load
The load demands in distribution systems vary from time to time, and each class of customers
follows a different pattern.
3.6.4 Environmental Conditions
Power system components are exposed to various weather conditions and hazards. Animals,
Motor vechicle accidents, rain, ice and tree contact can all lead to faults and failures,
environment dependent failures may be of short duration.

3.7 Failure Modes and Effects Analysis (FMEA)

In order to determine the system reliability, it is important to have a strategic methodology for
defining the impacts of different events. One qualitative analysis principle, called failure modes
Effects and Analysis (FMEA), is a bottom-up technique that is effective to identify component
failures with in a system [12] A successful FMEA helps to identify each failure mode.
Probability of occurrence of each failure mode, and necessary actions required mitigating such
failure modes, the steps followed in FMEA are listed as follows
 Identify all failure modes
 Figure out their probability of occurrence λ.
 Select a contingency and its impact on all loads
 Weigh the impact of contigent by multiplying with λ.
 Follow the previous steps to the rest of all contingencies


.

33
 3.8 Reliability Improvement Strategy
Due to increase in dependence on electricity and the growth of sensitive loads in all customer
sectors (residential, commercial and industrial) the requirement of continuous power supply
increases. The first step to maximize reliability is identifying the root cause of the outages .the
main strategy to improve reliability and power quality to customers are to eliminate faults and
then to minimize the effects on customers even if it occurs. After developing the reliability
improvement strategy, it is important to apply the interruption improvement techinques in order
to obtain better results. The improvement techniques can be basically in to two categories,
electric and non electric Figure 3.4 shows a simple radial distribution system consisting of
transformers, transmission lines (or feeders), breakers, fuses, and disconnects. Disconnects and
transmission lines such as s1 and l2 are designated as a main section. The main sections deliver
energy to the different power supply points. An individual load point is normally connected to a
power supply point through a transformer, fuse and lateral transmission line. A combination such
as f1, t2, and l5 is called a lateral section. A simple distribution system is usually represented by
a general feeder which consists of n main sections, n lateral sections, and a series component as
shown in Figure 6. In this feeder, Si , Li , Mi , and Lpi represent series component i , lateral
section i , main section i; and load point i , respectively. Li could be a transmission line, a line
with a fuse or a line with a fuse and a transformer. Mi can be a line, a line with one disconnect
switch, or a line with disconnect switches on both ends.

34
 3.9 Modeling of jigjiga Distribution System with ETAP software
ETAP software powerful electrical engineering analysis and management tools established as a
world leader in power system design, analysis and monitoring [22]
Modeling: to model the system with ETAP software Togowuchale feeder2 (33kv) feeder is
selected for modeling .this feeder is selected from other feeders because the high interrupted
cause to failure record is higher than the other feeder as shown table 4.2. To calculate the failure
rate of each component in the feeder for active failure rate and passive failure rate the failure rate
data and its cause is required .in order to model the feeder the cumulative load, sample feeder
line, for each type of transformer one sample and other component and device are modeled . The
reliability analysis and assessment using ETAP software for radial and looped system is very
efficient analytical algorithm. The algorithm is used based on reliability evaluation and
component modeling presented in probabilities all listed below equations, model reliability
characteristics of each component defined parameters and settings are calculated. To model
distribution system reliability system reliability analysis the following assumptions are
considered
• Only AC systems are considered
• All switching device operate successfully when required
• Switching devices can be opened whenever possible to isolate a fault
• All failures are statistically independent. the reliability indices which are used to calculate
for each component in component modeling are expressed as follow

35
 1. Expected Failure Rate (λ):

𝑟 𝑝 𝑖 𝑖 𝑖
3.20
𝑖 𝑝 𝑖 𝑝 𝑟 𝑖

λ is the frequency of load interruption per year.As shown in equation . it is the sum of the active
failure rate (λA) and passive failure rate (λp).

Active failure rate causes the operation of the protection devices around the failed component (like
short circuit fault) or is the failure of the component itself that restore services after replacement

2. Mean time to repair (MTTR) (r): r is the time (hours) required to repair a component outage and/or
restore the system in to normal operating state.
MTTR = r 3.21

3. Expected Repair Rate (µ): µ, is the frequency of a repair

µ= 3.22

4. Mean time to failure (MTTF)

MTTF is expected time (years) that the component will be in failure
MTTF = 3.23
Mean Time between Failure (MTBF)
MTBF is the expected time (years) between a component failure
MTBF = MTTF+ 3.24

5 Forced Outage Rate (FOR) The basic parameter used is the probability of finding the component on
forced outage at some distant time in the future. Its unavailability is evaluated from the previous indices
equation:

For=U= = 3.25
∗ ∗

6. Expected Energy not supplied index at load point (EENS)

Where Lai, and Ui, are the average connected load and the average annual outage time at load point i
respectively.
7. Expected interruption cost index at load point (ECOST):
ECOSTi = ∑ f(rij) ∗ λej 3.26
£
ECOST is the cost of not supplied energy at that load point. Interrupted Energy assessment Rate
index at load point {IEAR ($/KWh)}.
36
 𝑖
IEARi = 3.27

The IEAR is important index in order to identify the weak points of the network. The failure rates
and repair duration of different components such as transformers feeder circuit breakers, s bars,
feeder lines, isolators and fuses are summarized in table 4.1 below.
ETAP Algorithm: Figures 3.6 and 3.7 show the algorithm for distribution network reliability
assessment using ETAP and Togowuchale feeder of jigjiga Distribution Network in ETAP
environment respectively.

Figure 3.4 ETAP Algorithms for Calculation of Reliability Indices

37
Fig 3.5 Modeling of togowuchale line 33kv feeder

38
Fig 3.6 load flow of Modeling of togowuchale 33kv feeder line

39
 3.9.1 Study of jigjiga Power Distribution System
The existing power distribution in jigjiga town is radaial distribution system type. Power is
delivered to the customer from the utility in a one path way. There are no lateral and interconnection
or mesh type network topology.even through radial power distribution system is less costly in terms
of design and protection, it vulnerable to disturbance hence less reliable because of it‟s radially this
substation has a frequent interruption to the customer. Due to this reason the mesh or interconnected
distribution distribution is highly recommended to improve customer based reliability and power
availability.
Jigjiga substation has seven outgoing feeders‟ .it works in 15kv and 33kv voltages level in the
secondary side of the substation transformer. The figure below shows that different rating of
transformers used for each feeder.
Table 3.1 Number of Transformers and Rating of Each Transformer

Feeder ‟s Capacity (KVA)

Name 25 50 100 200 315 630 800 1250
KVA KVA KVA KVA KVA KVA KVA 400kva KVA Total
F1(15kv) 11 24 9 12 16 3 1 0 76
F2(15kv) 4 9 11 12 13 2 0 1 1 52
F3(15kv) 7 15 6 34 4 2 1 1 70
F4(15kv) 15 24 15 15 15 3 1 0 91
F1(33kv) 8 10 17 9 5 1 1 2 53
F2(33kv) 13 19 19 9 6 2 0 0 0 68
F3(33kv) 20 22 21 24 6 2 2 4 0 101
F4(33kv) 31 30 15 0 93
14 3 0 0 0

40
 3.9.2. Interruption Data from 2011 E.C to 2014 E.C

Among the different types of power system faults, frequently occurring faults at jigjiga substation
include permanent and transient earth fault, permanent fault and transient short circuit, and
interruptions due to operation/mentenance (planned).Table 3.9 shows the duration and frequency
of these different types of faults such as permanent earth fault (PEF), Permanent Short circuit
(PSC), Transient Earth fault (TEF) and Transient Short circuit (TSC) [12].
Table 3.2: Planned and Unplanned power interruption 2011 E.C

Feeder planned TSC PSC TEF PEF

Name Dur(hr
Freq(n) Dur(hr) Freq(n) ) Freq(n) Dur(hr) Freq(n) Dur(hr) Freq(n) Dur(hr)

15KV F1 70 213.76 15 1.04 38 49.57 30 4.81 28 67.49

15K F2 84 221.5 18 1.99 23 26.65 30 6.63 34 38.08
15KV F3 62 208.11 11 1.64 18 16.76 15 3.4 15 22.17
15KV F4 104 243.39 21 1.36 34 58.72 16 1.31 23 45.04

33KV F1 45 153.12 8 2.29 28 133.15 23 53.03 73 352.3

33KV F2 51 82.84 27 6.56 120 215 130 75 245 532
33KV F3 59 169.26 51 15.76 40 121.22 101 62.92 121 216.2
33KV F4 16.7 274.7
30 138.02 15 4.42 30 94.24 30 2 136 4
Total 223.8 1548.0
505 1430 166 35.06 331 715.31 375 675
2 2

41
 Tabl 3.3 Planned and Unplanned power interruption 2012 E.C

Feeder planned TSC PSC TEF PEF

Name Freq(n Freq(n Dur(hr

Freq(n) Dur(hr) Freq(n) Dur(hr) Freq(n) Dur(hr) ) Dur(hr) ) )

15KV F1 4 4.85 5 1.87 58 88.15 2 1.87 4 3

15K F2 8 22.75 3 1.05 66 89.7 2 0.58 6 6.8
15KV F3 7 4.6 6 1.4 30 57.25 2 1.5 3 3.15
15KV F4 6 5.5 1 0.03 65 61.95 2 2.38 19 31.6

33KV F1 3 3.07 0 0 17 75.22 1 0.83 25 42.3

33KV F2 15 60 25 10 103 215 25 5 88 215
33KV F3 6 7.67 0 0 41 80.4 3 2.03 120 167.12
33KV F4 5 6.07 2 0.77 61 143.8 2 1.83 96 77.12
Total 54 114.51 42 15.12 441 811.47 39 16.02 361 546.0
9

Table 3.4 Planned and Unplanned power interruption 2013 E.C

Feeder planned TSC PSC TEF PEF

Name Freq(n) Dur(hr). Freq(f) Dur(hr) Freq(f) Dur(hr) Freq(n) Dur(hr) Freq(n) Dur(hr)

15KV F1 17 37.956 5 0.21 96 185.25 39 2.03 104 155

15K F2 17 22.41 8 0.47 133 143 23 0.86 62 56
15KV F3 7 6.6 74 3.5 47 39.65 2 0.6 42 45.1
15KV F4 7 16.6 13 0.46 46 58.5 36 1.5 59 56.3
33KV F1 5 6.1 11 0.23 33 61.3 12 0.26 98 203
33KV F2 17 15.46 9 0.48 109 220 13 0.63 95 182
33KV F3 12 26.06 13 0.81 59 107 11 0.56 107 121.4
33KV F4 11 15.5 5 0.86 88 150 18 22.16 106 100
Total 93 146.686 138 7.02 611 964.7 154 28.6 673 918.

42
 Table 3.5 Planned and Unplanned power interruption 2014 E.C

Feeder planned TSC PSC TEF PEF

Name Freq( Dur(hr

Freq(n) Dur(hr) Freq(n) Dur(hr) n) Dur(hr) Freq(n) Dur(hr) Freq(n) )

15KV F1 48 91.63 33 0.86 127 238.61 33 1.18 107 232.26

15K F2 38 49.16 21 0.683 125 134.46 11 0.31 62 56.18
15KV F3 30 77.86 33 0.83 53 73.33 42 1.13 38 54.01
15KV F4 34 38.35 32 0.98 69 67.81 43 1.28 74 90.96

33KV F1 18 25.55 15 0.616 46 74.1 63 2.25 171 375.01

33KV F2 20 29.53 52 1.75 89 194.1 82 2.56 159 413.05
33KV F3 21 27.21 50 2.05 132 259.98 91 3.2 130 216.71
33KV F4 14 15.98 25 0.68 93 170.9 127 9.66 201 502.8
6
Total 223 355.27 261 8.449 734 1213.2 492 21.57 942 1941.
9 04

Table 3.6 Total planned and unplanned power interruption from 2011-2014

planned TSC PSC TEF PEF

Name
Freq(n Dur(hr) Freq(n Dur(hr) Freq(n) Dur(hr) Freq(n) Dur(hr) Freq Dur(hr)
) ) (n)
15KV
F1 139 348.196 58 3.98 319 561.58 104 9.89 243 457.75
15K
F2 147 315.82 50 4.193 347 393.81 66 8.38 164 157.06
15KV
F3 106 297.17 124 7.37 148 186.99 61 6.63 98 124.43
15KV
F4 151 303.84 67 2.83 214 246.98 97 6.47 175 223.9
33KV
F1 71 187.84 34 3.136 124 343.77 99 56.37 367 972.61
33KV 1342.0
F2 103 187.83 113 18.79 421 844.1 250 83.19 587 5
33KV
F3 98 230.2 114 18.62 272 568.6 206 68.71 478 721.43
33KV
F4 60 175.57 47 6.73 272 558.94 177 50.37 539 954.72
3704.7 265 4953.1
Total
875 2046.466 607 65.649 2117 7 1060 290.01 1 5
43
 Table 3.7 Frequency of Interruptions from 2011 E.C to 2014 E.C

Feeders 2011 E.C 2012 E.C 2013 E.C 2014E.C

Tota unPlan
pla unPlan Total plan unPlann l pla n Total
Unplann
nned ned ned ed Nned Ed planned ed Total
15kv F1 70 111 181 4 69 73 17 244 261 48 300 348
15kF2 84 105 189 8 77 85 17 226 243 38 219 257
15KVF3 62 59 121 7 41 48 7 165 172 30 166 196
15KVF4 104 94 198 6 87 93 7 154 161 34 218 252
177 295
33KVF1 45 132 3 43 46 5 154 159 18 313
573
33KVF2 51 522 15 241 256 17 226 243 20 382 402
372
33KVF3 59 313 6 164 170 12 190 202 21 403 424
33KVF 211 5 161 166 11 217 14 446 460
4 30 270 228
Total 1547 54 883 937 93 1576 223 2429
505 2052 1669 2652

44
 Table 3.8 Duration of Interruptions (hr/yr) from 2011 E.C to 2014 E.C

Feeders 2011 E.C 2012 E.C 2013 E.C 2014E.C

Tota unPlan
pla unPlan l plan unplanned Total pla n Total
Unplann
nned ned ned ed Nned Ed planned ed Total

122.91 336. 99.74 380.44 472.91

15kv F1 213.76 67 4.85 94.89 37.956 342.49 6 91.63 564.54
294. 120. 191.633
15kF2 221.5 73.35 85 22.75 98.13 88 22.41 200.33 222.74 49.16 240.793
129.3
252. 67.9
15KVF3 208.11 43.97 08 4.6 63.3 6.6 88.85 95.45 77.86 207.16
161.03
349. 101.46
15KVF4 243.39 106.43 8 5.5 95.96 16.6 116.76 133.36 38.35 199.38

693. 121.42 451.976

33KVF1 153.12 540.77 89 3.07 118.35 6.1 264.79 270.89 25.55 477.526

911. 505
33KVF2 82.84 828.56 4 60 445 15.46 403.11 418.57 29.53 611.46 640.99

585. 257.22
33KVF3 169.26 416.1 4 7.67 249.55 26.06 229.77 255.83 27.21 481.96 509.15
33KV 6.07 229.59 15.5 15.98 684.1 700.0
F4 390.12 336.6 223.524 288.5 8
138.02 7 273.2 2
Total 114. 1388.7 146.68 1919. 2065. 3184. 3539.
2522.21 3952. 51 1503.21 12 806 355.27 349 619
1430 21

45
 Table 3. 9 The Percentage of average duration and frequency of interruptions of different
faults

Planned TSC PSC TEF PEF

Name
Freq(n) Dur(hr) Freq(n) Dur(hr) Freq(n) Dur(hr) Freq(n) Dur(hr) Freq( Dur(hr)
n)

15KV F1 28.1
16.10% 25.20% 6.72% 0.288% 36.96% 40.65% 12.055 0.71% 5% 33.13%

15K F2 21.1
18.99% 35.92% 6.45% 0.47% 44.83% 44.78% 8.52% 0.95% 8% 17.86%

15KV F3 18.2
19.73% 47.73% 23.09% 1.18% 27.56% 30.03% 11.35% 1.06% 4% 19.98%

15KV F4 24.8
21.44% 38.75% 9.51% 0.36% 30.39% 31.50% 13.77% 0.82% 5% 28.5%

33KV F1 52.8
10.21% 12.01% 4.89% 0.2% 17.84% 21.98% 14.24% 3.6% % 62.19%

33KV F2 39.8
6.98% 7.58% 7.66% 0.75% 28.56% 34.09% 16.96% 3.35% % 54.20%

33KV F3 40.9
8.39% 14.31% 9.76% 1.15% 23.28% 35.37% 17.63% 4.27% % 44.87%

33KV F4 49.2
5.47% 10.05% 4.29% 0.38% 24.84% 32% 16.16% 2.88% % 54.67%
36.2
Average 11.96% 18.50% 8.30% 0.59% 28.96% 33.49% 14.50% 2.62% % 44.78%

Percentage average frequency of interruption each type of fault

Causes Operation TSC PSC TEF PEF

%Frequency(Int/ 11.96% 8.3% 28.96% 14.5% 36.2 %
Year)
Another fault includes, overload, blackout (total loss of power to an area), under frequency,
Transmission line fault etc

Fig 3.7 Percentage average frequency of interruption each type of fault

46
Percentage averageduration of interruption each type of fault

Causes Operation TSC PSC TEF PEF

%duration (Hr) 18.5% 0.59% 33.46% 2.62% 44.72%

18.19
0.58 operation
TSC
45.7
PSC
TEF

32.94 PEF

2.57

Figure 3.8 Percentage average duration of interruption each type of fault

. Percentage of four years duration interruptions of overall system of jigjiga Substation From the
above pie chart (Figure 3.8), 18.5% (Hr.) of the interruption duration of the overall system are due
to operation and maintenance, 44.72 % is due to a distribution permanent earth fault, 33.46 % is due
to distribution permanent short circuit, 2.62 % is due to distribution temporary earth fault and 0.59%
is due to temporary short circuit
The reliability indices of the existing substation can be calculated using equations (2.7) to (2.14),
which are given in chapter two. Based on the data given in Tables 3.9, 3.10 and 3.11, we can
calculate the reliability indices for 2011 to 2014 and the average of the four years

47
Table 3.10 Calculated reliability indices of 2011 jigjiga Substation

Feeders SAIFI SAIDI CAIDI ASAI ASUI EENS AENS

Int/Cust/Yr Hr/cust/Yr Hr/inter % % (MWh) (MWh/
Co)
15kv 181 336.67 1.86 0.961 0.039 572.3
0.20388315
feeder1
15kv 189 294.85 1.56 0.966 0.034 1002.49
0.30104805
feeder2
15kv `121 252.08 2.08 0.971 0.029 839
0.30453721
feeder3
15kv 198 349.825 1.766 0.960 0.04 1329
0.1570922
feeder4
33kv 177 693.389 3.917 0.921 0.079 554
0.65717675
feeder1
33kv 573 911 1.589 0.89 0.11 1343
0.68316327
feeder2
33kv 372 585.36 1.573 0.933 0.067 468.2
0.29916933
feeder3
33kv 241 529 2.19 0.939 0.061 740.6
0.59342949
feeder4

Table 3.11 Calculated reliability indices of 2012 jigjiga Substation

Feeders SAIFI SAIDI CAIDI ASAI ASUI EENS AENS

Int/Cust/Yr Hr/cust/Yr Hr/inter % % (MWh) (MWh/
Co)
15kv 73 99.74 1.366 0.988 0.012
feeder1 197.4852 0.061236
15kv 85 120.68 1.419 0.986 0.014
feeder2 458.584 0.119206
15kv 48 67.79 1.412 0.992 0.008
feeder3 250.823 0.079124
15kv 93 101.6 1.09 0.988 0.012
feeder4 457.2 0.046965
33kv 46 2.63 0.986 0.014
feeder1 121.42 121.42 0.125175
33kv 256 505 1.97 0.942 0.058
feeder2 898.9 0.397743
33kv 170 257.2 1.512 0.970 0.03
feeder3 282.92 0.156656
33kv 166 229.59 1.38 0.97 0.03
feeder4 378.8235 0.263989

48
 Table 3.12 Calculated reliability indices of 2013 jigjiga Substation

Feeders SAIFI SAIDI CAIDI ASAI ASUI EENS AENS

Int/Cust/Yr Hr/cust/Yr Hr/inter % % (MWh) (MWh/
Co)
15kv 261 380.446 1.457 0.956 0.044
feeder1 802.7411 0.232409
15kv 243 222.74 0.916 0.974 0.026
feeder2 890.96 0.21864
15kv 172 95.45 0.55 0.989 0.011
feeder3 362.71 0.106679
15kv 161 133.36 0.82 0.98 0.02
feeder4 637.4608 0.064002
33kv 159 270.89 1.703 0.969 0.031
feeder1 297.979 0.249773
33kv 243 418.57 1.722 0.952 0.048
feeder2 795.283 0.319391
33kv 202 255.83 1.266 0.97 0.03
feeder3 306.996 0.150858
33kv 225 388.52 1.726 0.955 0.045
feeder4 691.5656 0.416606

Table 3.13 Calculated reliability indices of 2014 jigjiga Substation

Feeders SAIFI SAIDI CAIDI ASAI ASUI EENS AENS

Int/Cust/Yr Hr/cust/Yr Hr/inter % % (MWh) (MWh/
Co)
15kv 348 1.622 0.935 0.065
feeder1 564.54 1467.804 0.352668
15kv 257 0.936 0.972 0.028
feeder2 240.793 1155.806 0.2367
15kv 196 1.056 0.976 0.024
feeder3 207.16 986.9102 0.207247
15kv 252 0.79 0.977 0.023
feeder4 199.38 1136.466 0.089486
33kv 313 1.525 0.945 0.055
feeder1 477.526 620.7838 0.430204
33kv 402 1.594 0.926 0.074
feeder2 640.99 1281.98 0.427327
33kv 424 1.2 0.941 0.059
feeder3 509.15 712.81 0.292495
33kv 460 700 1.52 0.92 0.08
feeder4 1330 0.665

49
Table 3.14 Calculated reliability indices the average of four years at jigjiga Substation

Feeders SAIFI SAIDI CAIDI ASAI ASUI EENS AENS

Int/Cust/Yr Hr/cust/Yr Hr/inter % % (MWh) (MWh/
Co)
15kv
feeder1 215.75 345.349 1.57625 0.96 0.04 760.0826 0.212549
15kv
feeder2 193.5 219.7658 1.20775 0.9745 0.0255 876.96 0.218899
15kv
feeder3 134.25 155.62 1.2745 0.982 0.018 609.8608 0.174397
15kv
feeder4 176 196.0413 1.1165 0.97625 0.02375 890.0317 0.089386
33kv
feeder1 173.75 390.8063 2.44375 0.95525 0.04475 398.5457 0.365582
33kv
feeder2 368.5 618.89 1.71875 0.9275 0.0725 1079.791 0.456906
33kv
feeder3 292 401.885 1.38775 0.9535 0.0465 442.7315 0.224795
33kv
feeder4 273 461.7775 1.704 0.946 0.054 785.2473 0.484756

SAIFI Int/Cust/Yr
400
350
300
250
200
150
100 SAIFI Int/Cust/Yr
50
0

Fig 3.9 Four years average value of SAIFI for jigjiga substation feeders

50
 SAIDI Hr/cust/Yr
700
600
500
400
300
200 SAIDI Hr/cust/Yr
100
0

Figure 3.10 Four years average value of SAIDI for jigjiga substation feeders
In this thesis, a special emphasis is given only to feeder two 33kv (Togowuchale) of jigjiga
distribution system. Because as per the data obtained from jigjiga substation and shown in figure 3.9
and 3.10 the mentioned city feeder has more frequency and duration interruptions than the other
feeders in the existing substation.
Table 3.15: Calculated reliability indices of jigjiga city feeder for year 2011-2014 E.C

Feeder Reliability 2011E.C 2012E.C 2013E.C 2014E.C Average

Name Indices
Jigjiga SAIFI 573 256 243 402 368.5
substation (Int/Cust/Yr)
33kv SAIDI 911 505 418.57 640.99 618.89
Feeder2 (Hr/cust/Yr)
CAIDI 1.589 1.97 1.722 1.594 1.718
(Hr/inter)
ASAI % 0.89 0.942 0.952 0.926 0.9275
ASUI % 0.11 0.058 0.048 0.074 0.0725
EENS 1343 898.9 795.23 1281.98 1079.77
(MWr)
AENS 0.447 0.299 0.265 0.427 0.35
(MWr)

51
 3.9.3 Comparison of the calculated values of Reliability Indices with
benchmarks [23,24]
In the Table 3.16 bellow shows the calculated Reliability indices values of jigjiga city feeder line
L2 compared with Ethiopia Electric Agency standard and other country benchmark values most
commonly used reliability indices. This thesis also focuses on the customer oriented reliability
indices and energy oriented indices.
Table 3.16: Comparisons of SAIFI and SAIDI values with different countries
Countries SAIFI (Int./Year/Customer) SAIDI (Hr./Year/Customer)
United States 1.5 4
Austria 0.9 1.2
Denmark 0.5 0.4
France 1 1.03
Germany 0.5 0.383
Italy 2.2 0.967
Netherlands 0.3 0.55
Spain 2.2 1.73
UK 0.8 1.5

52
 Table 3.17 Average and Peak Load of the jigjiga Substation in 2011 E.C
2011 peak 2011 minimum average Number of connected
feeder load(MW) Load(MW) load(MW) customer
1 1.7 0.17 0.935 2807
2 3.4 0.66 2.03 3330
3 3.33 0.39 1.86 2755
4 4 0.61 2.305 8460
1 0.8 0.066 0.433 843
2 1.47 0.074 0.837 1960
3 0.8 0.075 0.4375 1565
4 1.4 0.05 0.725 1248

Table 3.18 Average and Peak Load of the jigjiga Substation in 2012 E.C

2012 peak minimum average Number of connected

feeder load(MW) load(MW) load(MW) customer
1 1.98 0.198 1.089 3225
2 3.8 0.7 2.25 3847
3 3.7 0.43 2.065 3170
4 4.5 0.69 2.595 9735
1 1 0.07 0.535 970
2 1.78 0.08 0.93 2260
3 1.1 0.088 0.594 1806
4 1.65 0.059 0.8545 1435

Table 3.19 Average and Peak Load of the jigjiga Substation in 2013 E.C

2013 peak load minimum load average

feeder (MW) (MW) load(MW)
1 2.11 0.2 1.155 3454
2 4 0.8 2.4 4075
3 3.8 0.5 2.15 3400
4 4.78 0.78 2.78 9960
1 1.1 0.09 0.595 1193
2 1.9 0.097 0.9985 2490
3 1.2 0.098 0.649 2035
4 1.78 0.063 0.9215 1660

53
 Table 3.20 Average and Peak Load of the jigjiga Substation in 2014 E.C

2014 Peak minimum load average load no of customer at

feeder load(MW) (MW) (MW) feeder
1 2.6 0.493 1.5465 4162
2 4.8 0.9 2.85 4883
3 4.764 0.6 2.682 4762
4 5.7 0.9 3.3 12700
1 1.3 0.1 0.7 1443
2 2 0.11 1.055 3000
3 1.4 0.1 0.75 2437
4 1.9 0.082 0.991 2000

3.9.4 Outage Cost Evaluation of Jigjiga substation 33kv Feeder2 Line

Power interruption cost is both the utility and the customer sides; the size of the economic losses
due to interruption depends largely on the composition of the customers that experience
interruptions. customers in the jigjiga city are divided in to three categories: residential, commercial
and small industrial customers, on the utility side power interruption cost is estimated based on
customer data, interruption data, cost per outage data avariety of methods can be utilized to
evaluate customer impacts due to interruption. These methods can be groped,based on
methodological approach used, in to three broad categories, namely ,various analytical
evaluations,case studies of black outs, and customers surveys, even if a single approach has not
been universally applied , customer surveys are the main approach to determine specific information
for outage cost evaluation purposes. In practice, the cost of interruption from the customer
perspective is related to the nature and degree to which the activities interrupted are dependant on
electrical supply, in turn, this dependency is a function of both customer and interruption
characterstics include duration, frequency, and time of occurrence of interruptions; whether an
interruptions is complete or partial , if advance warning or duration information is supplied by the
utility; and whether the area affected by outage is localized.

54
 Table 3.21: Ethiopia electricity tariff (Birr/KWh) according to consumer class
Monthly As of As of As of Since Max
Residential Consumption dec 11/2011 dec 12/2012 dec 13/2013 Dec 14/2014
(KWh) on ward on ward on ward on ward
0-50 0.273 0.273 0.273 0.273 0.273
51-100 0.4591 0.567 0.6644 0.767 0.767
101-200 0.7807 1.0622 1.3436 1.625 1.625
201-300 0.9125 1.2750 1.6375 2 2
301-400 0.9750 1.3833 1.7917 2.2 2.2
401-500 1.0423 1.4965 1.9508 2.4 2.4
Above 500 1.1410 1.5877 2.0343 2.481 2.481
General Single phase 1.0352 1.3982 1.7611 2.124 2.124
tariff tariff(0-50kwh)
Low Three 0.8161 1.0544 1.2927 1.5310 1.5310
voltage phase tariff
industry Demand 50 100 150 200 200
tariff charge
Straight Single phase 1.0352 1.3982 1.7611 2.124 2.124
light

Based on Table 3.21, the interruption cost in the study area that is n jigjiga substation can be
calculated using the formula below; Interruption
Cost=Not supplied energy (in kWh)* ETB/Kwh
Where, the tariff 1.1985 is taken from the average of the first block tariffs of residential and
commercial customers, which most Ethiopian electricity customers are assumed to be grouped.
For jigjiga Substation Togowuchale feeder (L2) in the year 2011 has 1343.269 MWh unsupplied
Energy as seen in Table 3.18; then, Interruption cost is calculated as
Interruption cost=1343.269 *1000*1.1985=1,609,907.89 ETB

55
Table 3.2 Summary of estimated interruption cost from 2011 to 2014 E.C in jigjiga city
Toguwuchale feeder line.
Years EENS (MWh)
2011 1343
2012 898.9
2013 795.28
2014 1281.98
Max 1343

3.9.5 Method of fuzzy logic based Distribution Network Reconfiguration

From different network reconfiguration method this thesis paper used fuzzy logic method. it is the
strategy that allows determining the optimal network reconfiguration in radial distribution
networks, in correspondence to technical, regulatory and economic aspects, a formulation that
takes into account the investment, maintenance and power interruption cost has been developed
seeking for a reduction in total costs while takinng technical expert
In the jigjiga distribution system one of the major problems is the radial network is long, not
reconfigured with self or with parallel feeder and improper placement sectionalize switch on each
feeder so the reliability of electric power in our zone is at risk so, inorder to solve this problem.
The new proposed method is using fuzzy logic optimization technique for reconfiguration
distribution network through sectionalizing switch and tie switch By considering four linguistic
input such as line segment or length between buses, total number of branch line b/n two bus, Total
load capacity installed between bus and Environmental condition of the region e.g wind speed is
considering these geographical exposure inputs on the fuzzy logic Method.

56
Fig 3.11 fuzzy logic control for the benefit of switching allocation

57
 Table 3.23 set rule of switching allocation

Rule Linguistic
number Inputs Linguistic output
Wind speed SIB CLS SL Benefit of switch allocation
1 Very large Very Small Very Small Very Small Very small
2 Very large Very Small Very Small Very large Very large
3 Very large Very Small Small Very small Very small
4 Very large Very Small Large Medium medium
5 Very large Very Small Large Very large large
6 Very large Small Very small Very small Small
7 Very large Small Very Small Small Very small
8 Very large Small Small large large
9 Very large Small Small Very Large Very large
10 Very large Small Medium Very large Very large
11 Very large Small Small Very large Large
12 Very large Medium Small Medium Medium
13 Very large medium Medium Medium Medium
14 Very large small Medium Medium Medium
15 Very large Medium Large medium Medium
16 Very large large Large large large
17 Very large Medium Large Medium Large
18 Very large large Large Very large Very Large
19 Very large Very Large Very small Large Large
20 Very large Medium Large Very Large Very large
21 Very large Very small Large Large Large
22 Very large small Medium Large large
23 Very large Very large Very large Very Large Very Large
24 Very large Very large large Very Large Very Large
25 Very large Very small Very Large Very smll small
Very
26 large small Very Large small small
27 Very large Large Large Very Large Very Large

58
 Fig 3.12 fuzzy logic control for the benefits of tie switch allocation

Table 3.24 The Sets of Rules Used for Allocation of tie switches

Linguistic
Rule number Inputs Linguistic output

DHF LMF TCLF Benefit of tie switch allocation

1 very small small small small

2 very small very small very small very small
3 small small small small
4 medium small small small
5 very small large large large
6 large large large small
7 large medium medium small

8 medium large large medium

9 very small medium medium medium

10 medium medium medium medium

11 small large large medium

12 very small very large very large very large

59
Due to the nature of the study, it starts by reviewing literature related to the investigation of the
reliability problems of power distribution systems. Recent and unpublished important historical
outage data related to power reliability will be collected from Ethiopian electric Utility (EEU) at
jigjiga district. Interviews with the respective professional workers at jigjiga substations would be
contracted. Previously historical outage data would have been gathered for detail assessment and
investigation to come up with a clear solution of the problems at hand.
Generally the following methodology would have been followed in conducting the theis work.
1. Collection and analysis of data for radial distribution network from EEU of jigjiga region
office
2. Applying artificial intelligence for performing optimal placement of switches for the given
distribution network or the switching placement problems based on fuzzy logic technique
which was implemented using matlab software
3. Based on 2 reliability indices simulation carried out with etap software to compare the
initial and optimal configuration of distribution network
4. Develop a predictive reliability model of the study area using distribution analysis
softwares. The model is calibrated to represent the area‟s existing reliability
5. In general the following figure 3.13 summarizes the methodology used in this thesis.

60
 Data collection
and Literature
Reivew

System
Modeling

Optimization
problem
formulation

Solving
optimization
problem

Discussion
results

Fig 3.13 over all work Flow Methodology

61
 4 .CHAPTER FOUR
4.1 Result and Discussion
This chapter presents the explanations of the modeling and simulation of the existing system
with different mitigation alternative techniques to improve the system reliability of the jigjiga city
feeder line at a reasonable cost. Different techniques have been analyzed by ETAP 19.0.1 software
simulation, the simulation focuses on evaluating the impact of using different techniques such as
tie switch, sectionalizing switch and other techniques,. The reliability indices SAIFI, SAIDI and
cost benefits would be the main drivers for comparations of the alternatives using Fuzzy logic
techinques,. The reliability analysis is conducted by performing the following procedures in ETAP
19.0.1 software.
 Modeling of one-line diagram of the existing system using the software.
 Identify and input relevant parameters for the simulation
.  Specify type and operation characteristics of the protection devices.
 Specify protection coordination.
 Conduct Load flow and reliability assessment simulation

4.2 System Modeling

In this thesis, electrical transient Analysis program (ETAP 19.0.1 version) software has been used
as a design, simulation and reliability assessment analysis has been calculated.
To predict the reliability indices of jigjiga distribution system, values of failure rates and mean
time to repair for each component of static loads are necessary, to estimate the failure rate of the
line per kilometer, the total number of outages should be divided by the feeder length (kilometers)
as indicated in equation (4.1). The average mean time to repair (MTTR) each failure is computed
using equation (4.2)

𝛌A = ( ) 4.1
∗

MTTR= 4.2

By using equation (4.1) and equation (4.2), the basic reliability parameters used in ETAP
software for reliability analysis are calculated as follows:

𝛌A= =4.09 int/km.yr

∗

MTTR= =1.68 hr/int

62
 To estimate the failure rate of a component ETAP19 uses a combination of active and
passive failure rate use commensally.

The active failure rate is the number of failure rate per year per unit length. The passive
failure rate is associated with the component failure mode that does not cause the
operation of protection breakers and therefore does not have an impact on the remaining
healthy components, repairing or replacing the failed component will restore service
(ETAP software Library) ,as there is no means of isolating faulty area in the system is
assumed as zero in the model

The Togowuchale feeder (33 kV) has been constructed by 150 mm2 14 stranded, Aluminum
Conductor (ALC)

In order to modeling the existing city distribution line and to analyze the current distribution
reliability status, the following values have been considered as an input to ETAP 19.0.1
software.

 The city line data,

 Cable length,
 Number of customers,
 The connected load
 Number of transformers with their rating,
 Active failure rate
 Duration of interruption in hours
 Frequency of interruption

63
Figure 4.1 System modeling of the Togowuchale (line, L2)

64
Figure 4.2: Window of ETAP 19 software existing system of reliability page opened
Simulation result of reliability indices for existing system

65
The above simulation shows that the designed system has given approximated reliability indices
values as found from the average of four years historical interruption data of jigjiga city. The two
basic reliability indices, system average interruption frequency index and system average
interruption duration index are 621.39hr/cu.yr and 367.62f/cu.yr respectively.

4.3. Simulation and Result Analysis of jigjiga substation Togowuchale Feeder

This part shows the simulation analysis of the distribution system with different reliability
improvement techniques.
4.3.1 Optimal placement of sectionalize switch
The fuzzy logic system (FLS) was developed to solve specialist human task, inside a specific
knowledge domain. The heuristic knowledge about a system can be used to help to build a good
project. in this thesis a FES is used to evaluate the benefit of switch installation in the given
feeder section. The FLS applied to solve this problem has the following in put variables.
 Wind speed/geographical condition of the region/.
 Number of branches line (NBR)
 Total connected load (TCL)
 Section line length (SLL)
Number of branches connected with each section of the given feeder (NBR): This is a normalized
representation of the branching at the start of the section under consideration. It is the number of
load points connected to the main buses under each section.
Total connected load to the given section of the feeder (TCL): This is a standard representation of
the amount of load that is applied to the section under consideration.
Fuzzy logic rules are used in the design of the proposed fuzzy logic controller inference engine.
The rules created are derived from the common sense. Every linguistic input and output has a set of
membership functions (MF), and the associated MF used for all inputs and outputs are triangular
MF. Five fuzzy sets – very small, small, medium, large and very large - are employed in the
representation of the input.

66
Fig 4.3 Fuzzy Set Used To Output Variables

67
 Fig 4.4 for sectionalizes switch allocation problem
The rules for the FIS must be generated once the fuzzy sets for the input and output variables have
been defined. The output value, which is a number in the interval between 0 and 1, is determined
through fuzzy inference. This value is evaluated for every section of the network in order to
determine whether or not a switch needs to be installed over there. Every input variable (which was
initially a non-fuzzy input) is transformed into a fuzzy number during the inference process based
on the fuzzy sets that represents the input variables.
The number of branches, total connected load, and section line length are depicted in various
shapes in Figure 4.4. Each curve represents a membership function for various fuzzy (linguistic)
variables, such as very small (VS), small (S), medium (M), Improvement of Reliability of
Distribution System Using Network Reconfiguration large (L), and very large (VL). Using these
variables, rules are created as in figure 4.5 for determining the optimal number and location of
sectionalizes switches using these variables.

68
Figure 4.5 Rules Created for Optimal Placement of sectionalizes switch
The fuzzy logic designer tool consists of several interactive interfaces for creating a fuzzy
inference system (FIS) . Depending on the type of fuzzy rules developed, there are four dimensions
of the fuzzy inputs can be seen in different ways. I.e The correlation between the high wind speed
and total connected load of each line section, high wind speed and length of the line section, high
wind speed and total connected load is shown in figures 4.5 and 4.6, 4.7 respectively. Since there
are four inputs and there must be three possible three-dimensional surface viewers; as a result, the
surface viewers are generated using the previously developed fuzzy rules.
The higher flat surface on the surface viewer in figures 4.5 4.6 and 4.7 indicates that the section
needs a sectionalizes switch as it has been generated from the fuzzy rule.

69
Figure 4.6 Surface high wind speed and Number of Branches

70
Figure 4.7 Surface high wind speed and Total connected load

Fig 4.8 surface high wind speed and section line length

71
 The fuzzy rule viewer shown in figure 4.8 is used to view the inference process of fuzzy
inference system. It can be used to adjust the input values and view the corresponding output of
each fuzzy rule, the aggregated output fuzzy set, and the defuzzified output value. To view the
inference process, we must specify the input and output variables of fuzzy inference system, their
corresponding membership functions, and the fuzzy rules.

Figure 4.9 Rule Viewer of Fuzzy Optimization for Optimal Placement of sectionalize switch
The fuzzy inputs which are indicated in table 4.1 are the main factors that determine the

72
 performance of feeder2 33kv of the distribution system. They are obtained by normalizing the
input variables in appendix B.1. Normalization, i.e., input and output scaling factor, is used for
scaling and normalizing input and output variables in intervals between 0 and 1. These inputs are
derived from the characteristics of the chosen feeder, and the fuzzy inputs have a direct impact on
the optimization result.
Table 4.1 Optimization Index for sectionalizer switch Making Decision
Section No of branch Normalization Length Normalized Connected Normalized BEN index
NO (km) value load(KVA) value
1 1 0.2 0.5 0.06 0 0 0.189
2 1 0.2 0.8 0.09 630 1 0.505
3 1 0.2 1 0.12 25 0.039 0.38
4 1 0.2 3.5 0.42 25 0.039 0.264
5 2 0.4 1.5 0.18 400 0.634 0.546
6 1 0.2 0.7 0.08 200 0.317 0.369
7 5 1 0.07 0.008 125 0.198 0.5
8 1 0.2 0.2 0.024 50 0.079 0.358
9 1 0.2 0.5 0.06 100 0.158 0.393
10 2 0.4 2.7 0.32 250 0.396 0.571
11 1 0.2 0.5 0.06 100 0.158 0.442
12 1 0.2 1.3 0.156 25 0.039 0.399
13 1 0.2 2.5 0.301 25 0.039 0.278
14 1 0.2 0.1 0.012 200 0.317 0.215
15 1 0.2 0.5 0.06 100 0.158 0.393
16 1 0.2 0.2 0.024 100 0.158 0.275
17 1 0.2 0.2 0.024 315 0.5 0.235
18 1 0.2 0.15 0.018 315 0.5 0.215
19 1 0.2 2.4 0.289 125 0.198 0.442
20 1 0.2 2.5 0.301 150 0.238 0.46
21 1 0.2 1.3 0.156 100 0.158 0.471
22 1 0.2 3.65 0.439 25 0.039 0.264
23 1 0.2 0.25 0.0301 100 0.158 0.264
24 1 0.2 3.9 0.469 100 0.158 0.51
25 1 0.2 5.7 0.686 100 0.158 0.704
26 1 0.2 5.6 0.674 100 0.158 0.704
27 1 0.2 0.5 0.06 50 0.079 0.0893
28 1 0.2 4.6 0.554 50 0.079 0.701
29 1 0.2 2.1 0.253 50 0.079 0.089
30 1 0.2 1.05 0.126 100 0.158 0.0981
31 1 0.2 0.15 0.018 200 0.317 0.109
32 2 0.4 1.95 0.234 50 0.079 0.0989
33 1 0.2 8.3 1 50 0.079 0.857
34 1 0.2 0.2 0.024 250 0.396 0.109
35 1 0.2 7.6 0.915 50 0.079 0.735
36 1 0.2 2.2 0.265 100 0.158 0.0981
37 1 0.2 1 0.12 200 0.317 0.109
38 1 0.2 0.05 0.006 200 0.317 0.0924

73
39 1 0.2 0.15 0.018 225 0.357 0.109
40 1 0.2 0.1 0.012 50 0.079 0.0893
41 1 0.2 0.3 0.036 50 0.079 0.0893
42 1 0.2 0.3 0.036 200 0.317 0.109
43 2 0.4 0.1 0.012 300 0.476 0.5
44 1 0.2 5 0.602 100 0.158 0.703
45 1 0.2 0.5 0.06 100 0.158 0.09
46 1 0.2 0.56 0.067 350 0.55 0.5
47 1 0.2 0.4 0.048 50 0.07 0.5
48 1 0.2 0.2 0.024 200 0.317 0.109
49 3 0.6 0.25 0.0301 250 0.396 0.5
50 1 0.2 0.3 0.036 25 0.039 0.0893
51 2 0.4 0.26 0.031 200 0.317 0.0989

4.3.2 Optimal placement of tie switch

Distribution system network reconfiguration using interconnections system is highly recommended
by distribution system experts. In order to provide power in alternative ways, tie switches are used
to connect the supply side with that of the branches that are far from the supply. The feeder is
reconfigured utilizing tie switches connecting different nearby buses.
Normally open tie switches in ring design divide the feeder load according to the distance
between the buses. Improvement of Reliability of Distribution System Using Network
Reconfiguration and the capacity of the load connected to them. Ring feeder architecture is crucial
for power sharing to minimize power outages during peak load conditions. The buses of feeder2
may be able to employ tie switches to connect with the other buses based on fuzzy decisions.

74
Fig 4.10 Fuzzy Input-Output Variables and the FL Designer for Tie Switch Allocation

Figure 4.11 Rules Created for Optimal Placement of tie switch

75
The surface viewer shown in figure 4.11 indicates the higher flat surface requires placement of tie
switch.

Figure 4.12 Surface Viewer of DHF and LMF for Tie-Switch allocation
The values greater than or equal to 0.74 are chosen for the placement of tie-switches between buses
based on the fuzzy output

Figure 4.13 Surface Viewer of DHF and TCLF for Tie-Switch allocation
The higher value shows the need of the placement of tie-switches between buses based on the
fuzzy output Table 4.1 and 4.2, respectively; list the input variables for the fuzzy system and the
corresponding normalized values for fuzzy decisions for the placement of tie-switches.

76
 The values greater than or equal to 0.74 are chosen for the placement of tie-switches between buses
based on the fuzzy output. Accordingly, table 4.3 shows that there would be one tie-switch BF2
39 TO BF439
Table 4.2 Length between the Buses, Total Connected Load to the Bus and nearby healthy feeder
Benefit of tie switch installation index
Buses Distance of nearby Length of Total connected load
healthy feeder the model of modeling feeder
For 33kv feeder2 For 33kv feeder buses (DHF in KM) feeder(LMF) (TCLf in kva)
and 33kv feeder4 ( only modeling feeder) in km from
buses sub
BF22 TO BF42, Bf22 to Bf21 0.3 0.5 0
BF23 TO BF43, Bf2 3 to Bf21 1.3 1.6 630
BF24 TO BF44, B f2 4 to Bf21 3.4 2.6 655
BF25 TO BF45, BF25 to Bf21 7.85 6.1 680
BF26 TO BF46, BF26 to Bf21 11.35 8.1 1080
BF27 TO BF47, BF27 to Bf21 14.07 8.8 1280
BF28 TO BF48, BF28 to Bf21 15.14 8.87 1920
BF29 TO BF49, BF29 to Bf21 16.6 9.07 1970
BF210 TO BF410, BF210 to Bf21 19.1 9.57 2070
BF211 TO BF411, BF211 to Bf21 28.8 12.27 2320
BF212 TO BF412, BF212 to Bf21 36.3 12.77 2420
BF213 TO BF413, BF213 to Bf21 42.6 14.07 2445
BF214 TO BF415, BF214 to Bf21 45.6 16.57 2470
BF215 TO BF415, BF215 to Bf21 45.8 17.57 2670
BF216 TO BF416, BF216 to Bf21 51.3 18.07 2770
BF217 TO BF417, BF217 to Bf21 52.5 18.27 2870
BF218 TO BF418, BF218 to Bf21 54.7 18.47 3185
BF219 TO BF419, BF219 to Bf21 55.15 18.62 3500

BF220 TO BF420, BF220 to Bf21 58.05 21.02 3650

BF221 TO BF421, BF221 to Bf21 60.5 22.52 3650

BF222 TO BF422, BF222 to Bf21

65 25.02 3800

77
BF223 TO BF423, BF223 to Bf21 26.32 3900
66.6
BF224 TO BF424, BF224 to Bf21 29.97 4000
71.25
BF225 TO BF425, BF225 to Bf21 81.5 30.22 4100

BF226TO BF426, BF226 to Bf21 86.69 34.12 4200

BF227 TO BF427, BF227 to Bf21 92.59 39.82 4300

BF228 TO BF428, BF228 to Bf21 108.29 45.52 4400
BF229 TO BF429, BF229to Bf21 109.09 46.02 4450

BF230 TO BF430, BF230 to Bf21 116.84 50.62 4500

BF231 TO BF431 BF231 to Bf21 112.74 52.72 4550

BF232 TO BF432, BF232 to Bf21 85 53.77 4650
BF233 TO BF433, BF233 to Bf21 60 53.92 4850
BF234 TO BF434, BF234 to Bf21 56 54.07 4900
BF235 TO BF435 BF235 to Bf21 46.7 62.37 4950
BF236 TO BF436 BF236 to Bf21 43 62.57 5200
BF237 TO BF437 BF237 to Bf21 35 70.17 5250
BF238 TO BF438 BF238 to Bf21 15 85 5250
BF239 TO BF439 BF239 to Bf21 10 90 5350
BF240 TO BF440 BF240 to Bf21 13 90.5 5450
BF241 TO BF442 BF241 to Bf21 17 92 5650
The rule viewer shown in figure 4.12 is used to make decision where to place tie-switches between
the main buses

78
Figure 4.14 Rule Viewers for Tie-Switch Placement Decisions

79
 Table 4.3 Normalized Values of pharameters
Buses Normalization Normalization
For 33kv feeder2 For 33kv feeder of nearby Length of the Normalizatio Ben index
healthy feeder model n of Total
and 33kv feeder4 buses ( only (DHF in KM) feeder(LMF) in connected
km from sub load of
buses modeling feeder) modeling
feeder (TCLf
in kva)
BF22 TO BF42 Bf22 to Bf21 0.0025 0.005 0 0.138
BF23 TO BF43 Bf2 3 to Bf21 0.011 0.016 0.11 0.144
BF24 TO BF44 B f2 4 to Bf21 0.029 0.026 0.115 0.222
BF25 TO BF45 BF25 to Bf21 0.067 0.0628 0.12 0.291
BF26 TO BF46 BF26 to Bf21 0.097 0.083 0.19 0.342
BF27 TO BF47 BF27 to Bf21 0.12 0.09 0.226 0.469
BF28 TO BF48 BF28 to Bf21 0.129 0.09 0.339 0.469
BF29 TO BF49 BF29 to Bf21 0.142 0.09 0.348 0.476
BF210 TO BF410 BF210 to Bf21 0.16 0.098 0.366 0.431
BF211 TO BF411 BF211 to Bf21 0.246 0.126 0.41 0.33
BF212 TO BF412 BF212 to Bf21 0.31 0.13 0.428 0.306

BF213 TO BF413 BF213 to Bf21 0.364 0.145 0.432 0.294

BF214 TO BF415 BF214 to Bf21 0.39 0.17 0.437 0.246
BF215 TO BF415 BF215 to Bf21 0.39 0.18 0.472 0.249
BF216 TO BF416 BF216 to Bf21 0.439 0.18 0.49 0.138
BF217 TO BF417 BF217 to Bf21 0.449 0.188 0.507 0.156
BF218 TO BF418 BF218 to Bf21 0.469 0.19 0.56 0.156
BF219 TO BF419 BF219 to Bf21 0.47 0.19 0.619 0.156

BF220 TO BF420 BF220 to Bf21 0.498 0.216 0.646 0.17

BF221 TO BF421 BF221 to Bf21 0.519 0.232 0.646 0.164

BF222 TO BF422 BF222 to Bf21 0.169

0.558 0.257 0.67

80
BF223 TO BF423 BF223 to Bf21 0.27 0.69 0.172
0.57

BF224 TO BF424 BF224 to Bf21 0.308 0.707 0.18

0.61

BF225 TO BF425 BF225 to Bf21 0.697 0.311 0.725 0.18

BF226TO BF426 BF226 to Bf21 0.74 0.35 0.743 0.19

BF227 TO BF427 BF227 to Bf21 0.79 0.41 0.76 0.205

BF228 TO BF428, BF228 to Bf21 0.926 0.469 0.778 0.267
BF229 TO BF429, BF229to Bf21 0.93 0.474 0.787 0.267

BF230 TO BF430, BF230 to Bf21 1 0.52 0.796 0.5

BF231 TO BF431 BF231 to Bf21 0.96 0.54 0.805 0.265

BF232 TO BF432, BF232 to Bf21 0.72 0.55 0.823 0.5
BF233 TO BF433, BF233 to Bf21 0.51 0.55 0.858 0.42
BF234 TO BF434, BF234 to Bf21 0.479 0.557 0.867 0.446
BF235 TO BF435 BF235 to Bf21 0.399 0.64 0.876 0.478
BF236 TO BF436 BF236 to Bf21 0.368 0.645 0.92 0.476
BF237 TO BF437 BF237 to Bf21 0.299 0.72 0.929 0.476

BF238 TO BF438 BF238 to Bf21 0.128 0.92 0.929 0.473

BF239 TO BF439 BF239 to Bf21 0.085 0.97 0.946 0.741
BF240 TO BF440 BF240 to Bf21 0.111 0.98 0.964 0.472
BF241 TO BF442 BF241 to Bf21 0.145 1 1 0.472

81
 Every necessary parameter (connected load, number of branch line, cable length and active failure
rate are the major ones).

Figure 4.15 placing one sectionalize switch on 33kv F2 line for Scenario- 1

82
Simulation result reliability of indices for Scenario- 1

83
 Simulation result Scenario- 1, shows the reliability improvement with one sectionalizes switch in
the feeder. As can be seen from the table, sectionalize switch can significantly improves the
reliability of the system as follows
1. The expected number of outages per year has been reduced from 367.62 to 346.08 (4.49%
system reliability has improved),
2. The annual outage duration has been reduced from 621.39 to 585.37 hours (5.8% reduction
in outage duration)
3. The annual EENS improved from 1343.269 MWhr/yr to 1289 MWhr/yr (4.04%) as
compared to the existing system
Scenario - 2: Reconfigured Network with using Two sectionalizes switch in Existing System

Figure 4.16 placing two sectionalizer switch on 33kv F2 line for Scenario- 2

84
Simulation result reliability of indices for Scenario- 2

85
 Simulation result Scenario- 2, shows the reliability improvement with two sectionalizes switch
in the feeder. As can be seen from the table, the two sectionalize switch can significantly improves
the reliability of the system.as follows
1. The expected number of outages per year has been reduced from 367.62 to 343.09 (6.67%
system reliability has improved),
2. The annual outage duration has been reduced from 621.39 to 580.418 hours (6.59% reduction in
outage duration)
3. The annual EENS improved from 1343.269 MWhr/yr to 1274.5 MWhr/yr (5.11%) as compared
with the existing system.
. Scenario- 3 Reconfiguring the System using three sectionalizer switches on the existing system.

Figure 4.17 placing three sectionalizer switches on Distribution line forScenario- 3

86
 Simulation result reliability of indices for Scenario- 3

Simulation result Scenario- 3 shows the Reliability improvement with three sectionalizes switch
in the feeder. As can be seen from the table, three sectionalize switch can significantly improves
the reliability of the system.as follows
1. The expected number of outages per year has been reduced from 367.62 to 331.68 (10.8 %
system reliability has improved),
2. The annual outage duration has been reduced from 621.39 to 561.76 hours (9.5% reduction in
outage duration)
3. The annual EENS improved from 1343.269 MWhr/yr to 1254.3 MWhr/yr (6.62%) as
compared with the existing system.

87
 Scenario- 4 Reconfiguring the System using three sectionalizes switches and one tie switch on the existing system with 33kv feeder4
(BF2 39 TO BF439 )

Figure 4.18 NR of 33kv F2 with 33kv f4

88
 33KV feeder 4 can be connected by tie switch without causing any problem on the 33kv feeder2
line without any problem by construction new 10km Medium line voltage.
Depending on this, the failure rate of the system or feeder is calculated as follows

𝛌A= =3.685 int/km.yr

∗

MTTR= =1.68 hr/int

Simulation result Scenario- 4

89
Simulation result Scenario- 4 shows the Reliability improvement with three sectionalizes switch in
the 33kv feeder2 and one tie switch connection through 33kv feeder4 line Improve the reliability of
the system as follows.
1. The expected number of outages per year has been reduced from 367.62 to 166.37 (54.74 %
system reliability has improved),
2. The annual outage duration has been reduced from 621.39 to 269.94 hours (56.55% reduction in
outage duration)
3. The annual EENS improved from 1343.269 MWhr/yr to 701 MWhr/yr (47.76%) as compared
with the existing system
In scenario-4 the reliability of jigjiga 33kv feeder2 is improved by 54.74%, 56.55% and 47.76 %
for SAIFI, SAIDI and EENS respectively.

90
 4.4 Economical Cost Analysis
In this thesis, an economical cost analysis has been done on the interruption cost only related to
utility side only by using expected energy not supplied (EENS) of each scenario .for scenario1 the
annual average unsold energy is reduced from 1343.269 MWhr/yr to 1289 MWhr/yr as shown
table 3.21 the revenue saved due to reliability improvement can be estimated as follows
The revenue loss before improvement due to unsold energy is 1343.269MWh*1000*1.1985
ETB/KWH=1,609,907.89ETB
For Scenario-1; The Revenue saved after improvement due to unsold energy is:
1289Mwh*1000*1.1985ETB/Kwh=1,544,866.5ETB then after scenario1 improvement saved
energy is calculated as 1,609,907.89ETB -1,544,866.5ETB =65041.39 ETB
The investment cost can be calculated as below table 4.4work authorization
Table 4.4 one sectionalazer switch erection cost estimation

91
From above table the total investment cost is 71691.86 ETB
The payback period is used as an important factor to evaluate the economics competiveness of the
respective scenarios
It is the ratio of the cost of the investment to the annual saving by using that investment

Payback period

Payback period= 71691.86ETB/65041.39=1.1022yr payback time.

The reliability indices SAIFI and SAIDI have been reduced by 4.49% and 5.8% respectively
For scenario2 -The Revenue saved after improvement due to unsold energy is
1274.6Mwh *1000*1.1985=1527608.1Birr
Then after scenario2 improvement saved energy is calculated as
1,609,907.89ETB-1527608.1Birr =82,299.79Birr
Table 4.5 two sectionalazer switch erection cost estimation

92
Payback period

Payback period= 𝑟 payback time

For scenario3 -The Revenue saved after improvement due to unsold energy is
1254.3Mwh*1000*1.1985=1503278.55 ETB
Then after scenario3 improvement saved energy is calculated as
1,609,907.89ETB-1503278.55Birr =106,629.34Birr
Table 4.6 Three sectionalazation switch errection cost estimation

Payback period= pay back time

93
For scenario4 -The Revenue saved after improvement due to unsold energy is
701Mwh*1000*1.1985= 840148.5 ETB
Then after scenario4 improvement saved energy is calculated as
1,609,907.89ETB- 840,145.5 ETB Birr =769,762.39 Birr
Table 4.7 10km mv line constraction cost estimation
Customer Name:-General customer

Address:- - Sheger jigjiga ethiopia

Type of work:-.10km mv line exstension project

10 km
Stock No Description of Materials Unit Quantity Unit Price Total Price
1000075460 Imp.Wooden poles 10mt. ea 31 5,753.51 178,358.81

9200000171 concrite wooden pole 300 dan ea 50 10,053.44 502,672.00

09-02-120 POLE WOOD IMPEREGNATED 10MT Ea 0 2,105.94 -

09-02-110 POLE WOOD IMPEREGNATED 9MT Ea 0 1,962.44 -

1000074219 33kv chain insulator ea 6 360.00 2,160.00
1000074572 Long bolt M16,30mm ea 81 243.43 19,717.83
1000074622 33 kv H.T stay insulators ea 5 86.00 430.00
1000074101 33KV Pin ea 198 88.5 17,523.00
1000073713 33kv Insulator. ea 198 516.61 102,288.78
1000074094 suspension cross arm 33kv ea 66 4,008.23 264,543.18
9200000212 Big collar ea 66 206.85 13,652.10
12-02-910 bolt & nut M10 ea 300 13.79 4,137.00
1000074097 Tay straip ea 112 232.66 26,057.92
1000054942 AAC AL CONDUCTOR 95MM MT 30000 42.63 1,278,900.00
Total material cost Other than pole 1,729,409.81

Total pole costs 681,030.81

Total Material cost 2,410,440.62

Labour 561,600.00

OH cost 93,046.50

GRAND TOTAL 3,065,087.12

94
 Payback period

Payback period

In scenario-4, very important thing is isolating the 33kv feeder2 power line of Togowuchale-
Awubere which goes to 20km and replacing with it 33kv feeder4 power line Sheder- to Awubere
it takes 10km ,isolating the togowuchale-Awubere line increase the reliability the feeder in
togowuchale city.
In scenario-4- compared to other scenario this option have high technical advantage because of
reduction of line transportation by 20km that means it reduced line power loss
Table 4.8 summary of the results of reliability improvements of all scenarios
Customer Oriented scenarios
Reliability Indices
Existing Scenario Scenario Scenario Scenario
System 1 2 3 4

SAIFI (f/customer/yr) 367.62 346.08 343.09 331.68 166.37

SAIDI 561.76 269.94
(hr/customer/yr) 621.39 585.37 580.41
CAIDI(hr/customer/I 1.69 1.69 1.694 1.622
nt.) 1.69
ASAI (Pu) 0.929 0.93 0.932 0.935 0.969
ASUI (Pu) 0.07094 0.066 0.066 0.064 0.0303
% Reduction in 0 10.8 % 54.74 %
SAIFI 4.49% 6.67%
% Reduction in 6.59% 9.5% 56.55%
SAIDI 0 5.8%

From above table 4.8 scenario four the percentage reduction for SAIFI and SAIDI is about 54.74
%, 56.55% respectively because of that this scenario has enhanced the reliability of the system
very well compared to the existing system.

95
Table 4.9: summary of the result of system energy oriented reliability improvement for all
scenarios.

Customer Oriented scenarios

Reliability Indices
Existing Scenario Scenario Scenario Scenario
System 1 2 3 4

EENS(MWhr/y
r) 1343.269 1289 1274.59 1254.3 701.7
ECOST(ETB/y
r) 1,609,907.89 1544866.5 1527596.115 1503278.55 840987.45

%Reduction in
EENS 0 4.04% 5.11% 6.623%
47.76%
%Reduction in
ECOST 0 4.04% 5.11% 6.623% 47.76%
In addition to the system configuration upgrading, scheduled preventive maintenance, additional
rehabilitation works on the entire of the system, enhancing manpower problems, control strategy
and other related problems will have a great role in the system reliability improvement.

96
CHAPTER FIVE
CONCLUSIONS AND RECOMMEDATIONS
5.1 CONCLUSIONS
The research paper presented in this thesis mainly focused on irreducible the system Reliability
indicator in jigjiga distribution system. The reliability of distribution system can be improved by
using different improvement techniques. To select the appropriate and relevant improvement
method for a particular system the first step was making reliability assessment, through the
assessment process using different approaches and reliability indices that shows the performance of
the system was evaluated. For identifying the improvement techinque which are relevant to the
system predictive reliability alternatives are suggested. Moreover, the major causes for power
interruption are identified and the cost incurred by customers and utility are evaluated. In this
thesis both the historical and predictive reliability analysis analysis of jigjiga town distribution
system was carried out. The most common reliability indices, which include SAIDI, SAIFI, EENS
and ASAI, are used to measure the performance of the distribution system. The results obtained for
the historical analysis of the study area for the years 2011, 2012, 2013 and 2014 E.C showed that
the system reliability indices of the study area indicate poor performance, related to other European
countries as [23,24]. The reliability improvement techinque suggested as predictive reliability can
minimize the power outage frequency and duration. This improvement has a great impact socially
and economically both for the utility and the consumers. This research seeks to model the impact
of the optimal switch to distribution reliability. The reliability assessment in this work is carried
out with analytical approach and ETAP software simulation. The analytical approach presents the
reliability measures like SAIFI and SAIDI during the course of an average year. Hence, the mean
values like SAIFI and SAIDI during the course of an average year. Hence the mean values of
SAIFI and SAIDI for distribution systems optimal placement are obtained. The reduction in
reliability indicators in distribution system is achievable through the use of optimal placement
distribution system switches. Optimal switches arrangement in the network provides a reduction of
unsupplied energy to consumers, and costs associated with expected energy not supplied. Selection
of their number may depend on cost change of unreliability power supply or funds amounts
allocated for investments by the distribution operator network.

97
To solve this problem optimal number of switch allocation, FES optimization method is selected.
to illustrate the performance of proposed algorithm, an actual togowuchale 33kv feeder over head
line of jigjiga distribution system was selected as the test system
In this thesis work, four different scenarios have been proposed and analyzed for the
implementation of reliability improvements in the study area. As the simulation result shows all
scenarios have different reliability improvement contribution on the city feeder with 1.1022 to 3.98
years pay back period investment cost. All scenarios are tolerable interms of the investment cost.
But to improve city power reliability indices more scenario-4 is highly proposed. in this scenario
replacing 33kv feeder2 power line which goes to awubere woreda by 33kv feeder4 power line have
done in this thesis work, due to this increase the reliability 33kv feeder2 power line by decreasing
line transportation by 20km, it should be effective for reliability enhancement and better solution
for permanent and temporary faults.
By implementing scenario-4 the overall reliability of the the jigjiga 33kv feeder2 is improved by
54.74%, 56.55% and 47.76% for SAIFI, SAIDI and EENS respectively
The proposed solution has the potential to save around 769,762.39 ETB per year from the
technology investment made to improve the reliability satisfaction of the society has also been
considered as the priceless advantage.

5.2 Recommendation
The following recommendation is suggested based on practical observation and analysis on jigjiga
distribution system.
The reliability indices of all feeders of the distribution system does not meet the reliability standard
requirement, it require it implement structural changes to the energy sector increased investment
infrastructure, introduce regulatory initiatives to improve over all distribution system reliability
There must be feeder expansion for further improvement of reliability and meet the reliability
standards of Ethiopian electric agency (EEA)
EEU and customers have to formulate service level agreement (SLA) between them also there
must be a competitive electrical power provider to minimize the risk of customers due to power
outages.

98
EEU distribution system operators must follow the best maintenance strategies which are discussed
in the thesis by avoiding default way of maintenance culture.
The reconfiguration of 33kv feeder1and 33kv feeder 4 lines can be easily reconfigured through
33kv feeder2 line for the best reliability improvement of jigjiga 33kv distribution system.

5.3 Future works

Amore thorough examination of potential techniques to take system reliability indices in to
account, may be with a weighted average of several indices like SAIFI, SAIDI, and THD as the
objective function to minimize, is another possible extension of the approach. Therefore using the
reliability, power losses, environment effect, and other factors in to account will be preferable.
Reliabilty of the selected distribution system should be analyzed and reconfiguration must be
carried considering time-varying load and , the thesis offers average values for reliability indices it
does not take in to account the various behaviors of components over time , therefore one can
provide probabilistic approach to the same objectives by creating a chronological sequence of
random events, which builds the history of each component over predetermined time, it can
produce a significant sample system behavior.
Reconfiguration more than two feeders is preferable to improve reliability of the jigjiga
distribution system.

99
References
[1] N. Rugthaicharoencheep and S. Sirisumranukul, “Optimal Feeder
Reconfiguration with Distributed Generation in Three-Phase Distribution
System by Fuzzy Multiobjective and Tabu Search.” [Online]. Available:
www.intechopen.com

[2] G. M. Stapleton, G. Nourbakhsh, and P. Queensland, “APPLIED

RELIABILITY ASSESSMENT OF TRANSMISSION SUBSTATIONS.”

[3] T. Dorji, “Master of Science in Electric Power Engineering Reliability

Assessment of Distribution Systems-Including a case study on Wangdue
Distribution System in Bhutan,” 2009.

[4] D. Zhu, Y. Liu, W. Cyre, and J. D. Arthur, “Electric Distribution Reliability

Analysis Considering Time-varying Load, Weather Conditions and
Reconfiguration with Distributed Generation,” 2007.

[5] A. Ngaopitakkul, C. Pothisarn, S. Bunjongjit, B. Suechoey, C. Thammart, and

A. Nawikavatan, “A Reliability Impact and Assessment of Distributed
Generation Integration to Distribution System,” Energy Power Eng., vol. 05,
no. 04, pp. 1043–1047, 2013, doi: 10.4236/epe.2013.54b199.

[6] B. Bordeerath, “RELIABILITY EVALUATION OF COMPOSITE SYSTEMS

WITH RENEWABLE ENERGY SOURCES.”

[7] M. Teshome and F. Mabrahtu, “Reliability Assessment and Study the Effect
of Substation Feeder Length on Failure Rate and Reliability Indices,” Am. J.
Electr. Power Energy Syst., vol. 9, no. 3, p. 41, 2020, doi:
10.11648/j.epes.20200903.11.

[8] T. A. Faculty, X. Yu, and I. P. Fulfillment, “Copyright 2011 By Xuebei Yu,” no.
August, 2011.

100
[9] B. Moradzadeh, “Optimal Distribution Reconfiguration and Demand
Management Optimal Distribution Reconfiguration and Demand
Management within Practical Operational Constraints within Practical
Operational Constraints.” [Online]. Available:
https://trace.tennessee.edu/utk_graddiss

[10] D. Tegegne Asradew, “Reliability Analysis and Improvement of Debre

Berhan Power Distribution Network by Using Fuzzy Logic Optimization
Technique,” Am. J. Electr. Comput. Eng., vol. 5, no. 2, p. 45, 2021, doi:
10.11648/j.ajece.20210502.12.

[11] E. Engineering, “LECTURE NOTES ON ELECTRICAL DISTRIBUTION

SYSTEMS (15A02701) IV B. Tech I Semester (JNTUA-R15) VEMU
INSTITUTE OF TECHNOLOGY:: P.KOTHAKOTA Department of Electrical
and Electronics Engineering,” vol. 517112, pp. 2019–2020, 2019.

[12] Y. J. Jeon, J. C. Kim, J. O. Kim, J. R. Shin, and K. Y. Lee, “An efficient simulated
annealing algorithm for network reconfiguration in large-scale
distribution systems,” IEEE Trans. Power Deliv., vol. 17, no. 4, pp. 1070–
1078, Oct. 2002, doi: 10.1109/TPWRD.2002.803823.

[13] “Electrical Power Systems Quality, Second Edition.”

[14] J. Douglas, “Electrical Power Quality,” Van Nostrand’s Sci. Encycl., 2005, doi:
10.1002/0471743984.vse2831.

[15] A. Esteban and A. Alberto, “Optimal Selection and Allocation of

Sectionalizers in Distribution Systems Using Fuzzy Dynamic
Programming,” Energy Power Eng., vol. 02, no. 04, pp. 283–290, 2010, doi:
10.4236/epe.2010.24040.

[16] T. J. T. Hashim and A. Mohamed, “Optimal coordinated voltage control in

101
 active distribution networks using backtracking search algorithm,” PLoS
One, vol. 12, no. 10, 2017, doi: 10.1371/journal.pone.0177507.

[17] T. J. Hashim and A. Mohamed, “for Distribution Network with Distributed

Generations,” orld Acad. Sci. Eng. Technol. Int. J. Electr. Comput. Energ.
Electron. Commun. Eng. Vol7, vol. 7, no. 7, pp. 806–811, 2013.

[18] J. Vieira, A. Mota, and F. Morgado Dias, “Neuro-Fuzzy Systems: A Survey.”

[Online]. Available:
https://www.researchgate.net/publication/310047437

[19] T. Thakur, “Tabu Search Based Algorithm for Multi-Objective Network

Reconfiguration Search Based Algorithm for Multi-Objective Network
Reconfiguration Problem,” no. January 2011, 2014.

[20] S. Satyanarayana and B. H. Rao, “Artificial Bee Colony Algorithm for

Distribution Feeder Reconfiguration With Distributed Generation,” Int. J.
Eng. Sci. Emerg. Technol., vol. 3, no. 2, pp. 50–59, 2012.

[21] T. A. (Tom A. . Short, Electric power distribution handbook. CRC Press,

2004.

[22] “www.ETAP.com operation of technology, Inc. (888) OTI-ETAP”.

[23] “Emerging Technologies’’ Contributions in Promoting” Dr. Gashawtena

Bayou et al., B. Defence and Industry Capabilities," in DEC “Techno-Fiesta-
2015,” and 2011. Ethiopia, July 15-16,

[24] D. on C. of O. E. Ethiopian Electric Agency (EEA), E. E. Line, and Quality of

Supply No. EEA/1/2013, Addis Ababa, and 2013. Electric Agency (EEA),

[25] G. Melesse, H. Melesse, and D. T, “Determining the reference basic wind

speed in Ethiopia and comparing with compulsory Ethiopian standard CES
145, 2015,” Results Eng., vol. 9, no. October 2020, p. 100209, 2021,
102
 Appendex A
Wind speed of ethiopian main city

no Name of town Annual max wind Reference

speedin mps
1 Asosa 31.88m/5 G. Melesse, H.
2 dire 38.5m/s Melesse, and D. T,
3 hawasa 34.34m/s “Determining the
4 gambela 26.46m/s reference basic
5 harer 39.1 m/s wind speed[25]
6 debirebirehan 38.41m/s
7 gonder 35.74m/s
8 debiretabor 33.11m/s
9 arbaminch 40.25m/s
11 jima 26.11m/s
12 jigjiga 54.65m/s
13 addisabeba 44.85m/s
14 adama 37.8m/s
15 bahirdar 30.71m/s

Appendex B
Component reliability data

Component type Failure rate (f/yr.) Repair time (hrs.) Switching time
Transformer
33/0.4 KV 1.35 1.68 1
Breaker
33 KV 1.35 1.68 1
Bus bar
33 KV 1.35 1.68
Feeder line(33KV) 4.09 1.68 1
Isolator(33 KV) 4.09 1.68 1
Fuse (00.4 KV) 1.35 1.68 1

103