Distribution System Reliability Analysis
Distribution System Reliability Analysis
To request enhancements,
A Project
MASTER OF SCIENCE
in
by
Gracious Nyati
FALL
2016
ANALYSIS OF A DISTRIBUTION SYSTEM RELIABILITY
A Project
by
Gracious Nyati
Approved by:
____________________________
Date
ii
Student: Gracious Nyati
I certify that this student has met the requirements for format contained in the University format
manual, and that this project is suitable for shelving in the Library and credit is to be awarded for
the project.
iii
Abstract
of
by
Gracious Nyati
Reliability analysis is believed to be one of the key features in power system design and
planning. Reliability defines the ability of a system or component to function under stated
conditions for a stated period of time. With the promotion of the concept of smart grid in
the past few years, more and more Distributed Generation (DG) units are being introduced
into the power system, which in turn may affect the operation of the networks. DGs play
an important part in many ways, the introduction of DG can significantly impact the flow
of voltage and power conditions of the system. There are many kinds of DG units classified
into two technologies: conventional and renewables. The main purpose of this project is to
study and analyze power system reliability due to the engagement of DG units in a
distribution system. A reliability analysis technique for a large scale system, with respect
to substation, reconfigurable, and electrical distribution is presented. The analysis will take
into account the average failure rate and the annual outage duration of the distribution
iv
system with and without distributed generation. Also the placement of distributed
generation and its effects on the reliability of the system is investigated and analyzed.
_______________________
Date
v
TABLE OF CONTENTS
Page
Chapter
1. INTRODUCTION .....................................................................................................1
4. SIMULATION.........................................................................................................11
4.4. Simulation with Wind Turbine Generators on Three Sub Buses ........... 16
5. CONCLUSION........................................................................................................20
Appendix ......................................................................................................................22
References ....................................................................................................................31
vii
LIST OF TABLES
Page
Table 4.1. Bus Input Data for the Simulated Distribution System …………………….12
Table 4.5. Results of Simulation with Wind Turbine Generators on three buses ……...17
viii
LIST OF FIGURES
Page
Figure 4.4. Simulation with Wind Turbine Generators on three buses ........................... 17
ix
1
1. INTRODUCTION
1.1. Introduction
The primary function of a power system is to provide satisfactory electrical supply to the
customers with reasonable level of pricing and reliability. With the growing demand and
the increasing dependence on electricity, reliability of the electric power system is believed
to be critical to the economic strength and the well-being of the society. In today’s society,
the reliability of power systems is being challenged, as the substructure is aging, the
demand is always growing, and also as new operational challenges emerge which include
the integration of irregular renewable generation [1]. The monetary and social
consequences of power loss in an area can have considerable impacts on both the customers
and the utility company supplying the service. The cost of a key power outage limited to
one state can be in the range of millions of dollars, and if the outage was to affect multiple
states, the cost could be in the range of hundreds of millions to billions of dollars [2].
Maintaining reliability is vital in any power system design and operation, as the system is
susceptible to irregularities that may include failures, disturbances such as human errors or
natural cause effects like lighting, tornadoes and also situations like protection or
communication system failures. With the development of smart grid, there has also been
an influx of Distributed Generations (DGs) into the power systems, which to some degree
has had some positive and negative effects on the system [3]. This project is aimed at
analyzing and predicting the effects on reliability of a distribution system with distributed
generation.
2
Distributed Generation (DG) also known as distributed energy or on-site generation (OSG)
is a technique that uses small-scale technologies to produce electric power close to the
energy) generators that offer a number of possible benefits [1]. In most cases, distributed
generators are said to provide lower-cost electricity and higher power reliability and
security to the power system with fewer environmental concerns compared to the
traditional power generators. The negative side of distributed generation is that sometimes
the power quality of the system can be affected, also islanding operation and voltage
control problems may be experienced due to the inclusion of distributed generation. Even
though the concept of distributed generation is not new, the increase of power demand and
the development of smart grid applications have resulted in a large growth of distributed
generators as well as any type of electrical inverter capable of producing A/C power, these
include small hydro, biomass, biogas, solar power, wind power and geothermal power.
The primary objective of an electric power system is to supply and meet the satisfaction of
the customers (consumers) and the suppliers (utility companies) of electricity with
reasonable guarantee of continuity and quality. In a quick short definition, power system
power supply under normal and abnormal conditions [3]. The view of power system
3
reliability is somewhat wide-ranging and covers the ability of the system to fulfill the
consumer and utilities requirements. System reliability can be subdivided into “system
SYSTEM RELIABILITY
System adequacy is associated with the existence of necessary facilities within the system
to fulfil the consumer load demand, these include the capabilities necessary to produce
sufficient energy and the related transmission and distribution facilities necessary to
transport the energy to the actual consumer load points. System security correlates to the
Regular electric consumers believe they are supposed to have electric power at all times,
however in reality utility companies are only able to meet that requirement to about 99.99
percent of the time [4]. Electric power is known to be one of the essential elements of the
modern economy, the availability of reliable power at reasonable cost is therefore crucial
for the economic growth and development of a country. Power utilities and governments
4
throughout the world strive to meet customer demands as economically as possible with
Electric Reliability Corporation (NERC) was established in the United States with a
mission to assure the reliability of the bulk power system in North America [4]. NERC
develops and enforces Reliability Standards [4]. Prior to the 1960s, reliability of a projected
electric power system was often predicted by generalizing the knowledge obtained from
the existing systems and employing some rule of thumb methods to project what the
the reliability indices [5]. Distributed generation can improve the capacity of some utility
systems to operate peak load on a feeder, and can also help supply loads during
There are several indices used to compute system’s reliability. The acknowledged
reliability indices are defined in the Institute of Electrical and Electronic Engineering
(IEEE) standard number P1366 [6]. A wide range of performance indices have been
suggested. The most common distribution indices are discussed in the following section.
or
= the total number of customers served for the area being indexed
SAIDI is the most commonly used performance index for a sustained interruption. It
measures the total duration of an interruption for the average customer on a given period
N
= Totail number of customer interruptions
SAIFI =
SAIFI
N Ti NTT otal number of customers served
6
Σr Ni
S
SA DI= i Customer
AIIDI interruption durations
ri TNota
T l number of customers served
Once some outage occurs, the average time to restore service is found from CAIDI. CAIDI
is the index used to calculate the average time required to restore service to the average
or
of electric power systems. This chapter briefly describes and defines the basics of electric
power distribution system design. A distribution system may be classified on the basis of:
DC distribution system.
underground system.
When a distribution system is connected to substation on one end only with the help of
feeder, the system is called “radial distribution system”. The feeders, distributors and
service mains are radiating away from the substation therefore the name radial system is
given to this type of system. There are combinations of one distributor and one feeder,
8
connecting the distributor to the substation. In a radial system like the one shown in figure
3.1, if the fault occurs either on a feeder or a distributor, all consumers connected to that
b. When the load on the distributor changes, the consumers at the distant end of the
c. As consumers are dependent on single feeder and distributor, a fault on any of these
The loop distribution system loops through the service area and returns to the point of
origin. This type of placement of switches allows the distribution system to supply power
to its customers from either direction. If one power source fails, switches are opened or
closed to obtain power source from the other source in the system. Loop systems provide
better service continuity than radial systems, with only short service interruptions during
9
switching [3]. However, they are more expensive than radial systems because of the
additional switching equipment requirements. As a result, loop systems are usually built to
serve commercial, light industrial buildings and shopping malls, where power outages are
more likely to endanger human lives or result in property losses. Reliability and service
quality can be considerably improved at even higher cost with a multiple parallel circuit
arrangement [2]. In these systems, two or more circuits are tapped at each substation. The
circuits can be radial or end in a second bulk power source and these interconnections allow
In order to improve system reliability, distributed generation unit can be added into the
system as shown in figure 3.3 below. In such a system, if a failure occurs between Source
#1 and Switch #1, the operator can open Switch #1 to isolate the failure and close Switch
10
#4 to incorporate the Distributed Generator to compensate for Source #1. Without the
Distributed Generator, the whole system would be drawing all the power from Source #2
which might cause system constraints violations or lower the power quality of the system.
If this happens during peak time, the system might experience high demand with less
supply which could be detrimental to the whole system and the customer.
Source #1 DG Source #2
4. SIMULATION
Distribution system reliability assessment is needed to evaluate the availability and quality
of power supply at each customer service entrance. Analysis of customer failure statistics
has shown that, compared to other parts of electrical power system, distribution system
failures are known to contribute as much as 90% towards the unavailability of power
supply to a load [7]. These statistics show the importance of reliability evaluation of
distribution systems. The basic reliability indices used to predict or measure the reliability
Indices consisting of the Expected Energy Not Supplied (EENS), Expected Interruption
Cost (ECOST), and Interruption Energy Assessment Rule (IEAR) can be used particularly
for each load point or for the whole system. These indices can be used to approximate the
concerning improvements or design of new distribution system [8]. Also, in order to study
the sensitivity of a reliability index EENS or ECOST with respect to failure rate of different
elements, element contribution to that index and their standings can be used.
The indices and rankings given above can be evaluated using the ETAP Reliability
Analysis module. This module provides the best tool to efficiently model various power
system elements and devices to include their effects on the distribution system reliability.
12
configurations. By using this module, we can evaluate the distribution system reliability,
and the values of different reinforcement patterns available to the planner that can be
quantitatively assessed to warrant that limited capital resources are used to achieve the
greatest possible improvement in system reliability [8]. Below in table 4.1 is the bus input
The system is simulated without any distributed generation, the system is radial and the
source is at the utility grid point through the main bus. As seen in figure 4.1, we see the
average failure rate and the annual outage duration at buses and load points as listed. We
focus on the 5 buses shown in the figure 4.1 for which the results are shown in table 4.2
Table 4.2 shows the results of the average rate of failure and the average outage duration
of the system at 5 bus stations selected for analysis without any distributed generation in
the system.
In this section, the simulation of the system was performed with DG on Sub2B, the logic
here was to analyze the system when the DG is placed in the middle of system. The
14
objective is to evaluate the effects of the DG to the surrounding bus loads in the whole
BUS ID Av. Failure Rate Av. Outage Duration Ave. Rate Failure
(f/yr) (hr) Improvement (%)
Main Bus 0.01 180.2 28.57
Sub2A 0.062 193.0 6.06
Sub2B 0.013 184.8 81.16
Sub 3 0.055 196.4 23.61
Bus 1 0.098 196.0 14.78
Table 4.3. Results of Simulation with a DG in Sub2B
Table 4.3 shows the results of the average failure rate, the average outage duration and the
average failure rate improvement percentage of the system compared to previous case
(without any distributed generation). Only the average failure rate is compared for different
15
scenarios to give an idea on the reliability improvement. The results above also indicate
that when a DG unit is placed under one bus, it has some effects on the surrounding loads
but most effect is felt on the bus or load closest to it. In this particular simulation we can
see that we get the most improvement on Sub2B with highest improvements percentage
For this section of the simulation, three conventional synchronous generators were added
to the distribution system on the three sub buses Sub2A, Sub2B and Sub3. The results of
the average rate failure and the average outage duration are shown in table 4.4
BUS ID Av. Failure Rate Av. Outage Duration Ave. Rate Failure
(f/yr) (hr) Improvement (%)
Main Bus 0.010 180.2 28.57
Sub2A 0.011 193.0 83.33
Sub2B 0.013 184.8 81.16
Sub 3 0.014 196.4 80.55
Bus 1 0.057 196.0 50.43
Table 4.4. Results of Simulation with Synchronous DG on Three Buses
Table 4.4 shows the results of the average failure rate, the average outage duration and the
average failure rate improvement percentage of the system compared to the system without
any distributed generation. In this system three conventional synchronous generators are
placed on Sub2A, Sub2B and Sub 3, based on the results of table 4.4 we can see that adding
more DGs to the system results in higher reliability improvement. The DGs turn to share
the responsibility of the loads on the system and thereby improving the reliability of the
system.
For this part of the simulation, we incorporated the wind turbine generators as the
distribution generators to analyze and compare the effects of different types of DGs. In
order to compare and analyze the results, the three wind turbine generators are connected
into the three buses similar to the previous simulation in section 4.3. Wind Turbine
BUS ID Av. Failure Rate Av. Outage Duration Ave. Failure Rate
(f/yr) (hr) Improvement (%)
Main Bus 0.010 180.2 28.57
Sub2A 0.013 193.0 83.33
Sub2B 0.013 184.8 81.16
Sub 3 0.016 196.4 77.78
Bus 1 0.059 196.0 48.70
Table 4.5. Results of Simulation with Wind Turbine Generators on Three Buses
The results on table 4.5 show the average failure rate, the average outage duration and the
average failure rate improvement percentage of the system compared to the system without
any distributed generation. In this system, three turbine generators are connected in three
buses similar to section 4.3. The results show that the system had high improvements
compared to a system without DGs, however slightly lower results when compared to
18
system with conventional generators. Comparing the average failure rate improvement of
renewable generator to the conventional generators in same position shows there was very
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
Main Bus Sub2A Sub2B Sub3 Bus 1
Without DG With 3 Sych. Gen With 3 Wind Turbines
The chart in figure 4.5 gives a comparison of three scenarios simulated in this distribution
system without any DG in the system, with three conventional synchronous DGs connected
in buses Sub2A, Sub2B and Sub3, and finally with three Wind Turbine Generators
connected in the same buses. Looking at the average rate of failure for the system, we can
see that without DGs, the rate is higher making it less reliable compared to the ones with
DGs. Also comparing the system with added conventional synchronous DGs with the one
with Wind Turbine Generators, we can conclude that the system with added conventional
19
synchronous generators is slightly more reliable than the wind turbine one. Below is a table
for the average performance indices, which gives us the results of the three scenarios
discussed above.
Table 4.6 presents a summary of performance indices for three scenarios in order to
compare the effects of the DG in different scenarios. In the first scenario without DG in
the system, we can see that reliability is the lowest, however in the second scenario, (the
system with three conventional DGs) the reliability of the system is the highest compared
to all three scenarios analyzed. The third scenario with three wind turbine generators we
can realize that the reliability is higher than without any DG but less than with conventional
DGs.
20
5. CONCLUSION
deregulation of power systems are some of the factors that have accelerated the applications
of distributed generation in the past few decades. Distributed generators have specific
characteristics which distinguish them from traditional generating units. The use of
the system [9]. In this project, using ETAP software we analyzed four different ways of
using distributed generation and their effects on the average failure rate and the annual
First we analyzed the system without any distributed generation to give us the reliability
base of the system and then we added one DG to see what kind of improvement we could
achieve, and as shown in section 4 there was some improvement. We then analyzed the
effects caused by putting more DGs in the system and we proved that reliability went up
significantly with more DGs. Also we analyzed and compared the concept of using
conventional generators versus using renewables in the form of wind turbine generators.
Comparing renewable and conventional DGs we realized that there was much stable higher
improvement on the conventional DGs than renewable ones, this is contributed by the
power except at times when they have problems that need to be fixed. On the other hand,
renewables have different effects associated with their supply, for example wind turbine
DGs used in this simulation have to take into account the weather conditions. When there
21
is less wind, the generators will not be producing to their maximum capabilities and this
would have an effect on the system’s reliability. With more wind, the generators are able
Incorporating distributed generation in power systems is not the only way to improve
reliability, there are also some steps that we can take into consideration for example proper
maintenance of the power system. In this project we have shown how distributed
generation can improve the reliability of a system. Also in this project it is assumed that
the load in the system remains unchanged in all scenarios, however in real life loads
incorporate more load modeling approaches and consider the variation of the renewables.
22
Appendix
Reliability Analysis
Total
Number of Buses: 11
Number of Branches: 6 1 0 2 0 1 0 10
Synchronous Power
Grid Inverter Total
Generator
Number of Sources: 1 1 1 3
Number of Loads: 2 2 3 13
6
System Frequency: 60 Hz
Unit System: English
Project Filename: EXAMPLE
23
Replacement
Switch
Bus A MTTR Time Avail. Time
ID kV f / yr hour hour Yes/No hour
Branch Connections
LUMP5 Govt. & Inst. Sub23 0.0730 153.48 11.2040 4.7617 22796.26
T2 Xfmr2 95036.68
CB11 PD 19007.34
CB2 PD 19007.34
Fuse1 PD 19007.34
CB12 PD 14437.03
CB21 PD 14073.97
CB27 PD 9866.73
T1 Xfmr3 187.50
T1 Xfmr3 187.50
T3 Xfmr2 187.50
T3 Xfmr2 187.50
CB44 PD 37.50
CB18 PD 37.50
CB43 PD 37.50
CB41 PD 37.50
CB13 PD 37.50
CB42 PD 37.50
T2 Xfmr2 24666.83
T2 Xfmr2 24666.83
CB21 PD 4933.37
CB10 PD 4933.37
CB11 PD 4933.37
CB11 PD 4933.37
CB2 PD 4933.37
CB6 PD 1225.50
CB5 PD 1225.50
CB26 PD 1225.50
CB26 PD 1225.50
CB24 PD 1225.50
CB24 PD 1225.50
CB4 PD 1225.50
CB6 PD 1225.50
CB5 PD 1225.50
CB26 PD 1225.50
CB26 PD 1225.50
CB24 PD 1225.50
CB24 PD 1225.50
CB4 PD 1225.50
29
T1 Xfmr3 187.50
T1 Xfmr3 187.50
T3 Xfmr2 187.50
T3 Xfmr2 187.50
CB44 PD 37.50
CB18 PD 37.50
CB43 PD 37.50
CB41 PD 37.50
CB41 PD 37.50
T2 Xfmr2 22851.51
T2 Xfmr2 22851.51
CB21 PD 4570.30
CB21 PD 4570.30
CB10 PD 4570.30
CB11 PD 4570.30
CB11 PD 4570.30
CB2 PD 4570.30
CB2 PD 4570.30
CB18 PD 0.00
Fuse2 PD 0.00
Fuse2 PD 0.00
CB13 PD 0.00
CB16 PD 0.00
CB7 PD 0.00
CB15 PD 0.00
CB17 PD 0.00
CB8 PD 0.00
CB8 PD 0.00
SUMMARY
System Indexes
ASAI 0.9977 pu
ASUI 0.00232 pu
EENS 130.226 MW hr / yr
ECOST 286,465.10 $ / yr
IEAR 2.200 $ / kW hr
References
systems," in IEEE Power Engineering Society Summer Meeting, 16-20 July 2000.
[2] R. B. Standler, "Liability of Electric Utility in the USA for Outage and Blackout,"
[Link]/outage, 2011.
[5] T. Gonen, Electrical Power Distribution System Engineering, CRC Press, 2008.
[7] W. X. Lin Jikeng, "Reliability Evaluation for Distribution System with Distributed