DRPT2008 6-9 April 2008 Nanjing China

Comprehensive Method on Evaluation of

Distribution Network Planning
Jun Xiao, Yanyan Cui, Fengzhang Luo, Mingzhi Liu,

Jianmin Wang, Yinong Li , Yubo Gao and Saiyi Wang

evaluation mainly includes some single indices such as

Abstract-- Planning is vital for the development of power security [1]-[3], reliability [4] as well as power supply
system; and the quality of planning has direct influence on quality [5]-[6], etc. Reference [5] and [7] presented the
security, reliability and economy of the power system in the
objective and comprehensive evaluation indices by making
future. Therefore, it is urgently needed to study quantitative
evaluation method for distribution network planning. Firstly, use of the fuzzy comprehensive evaluation method, the
this paper proposes a practical comprehensive hierarchy and uncertainty of input data has been concerned in [6] while
method on evaluation of distribution network planning. evaluating the reliability. To a certain extent, all of the above
Secondly, the implementation steps for the evaluation of improved the assessment results and evaluated the technical
planning network are given, which include the determination level of distribution network from different aspects, but
of indices hierarchy and the evaluation of planning. In the end,
didn’t give effective guidance to the construction of network
this paper takes 10kV network distribution planning of a
because of the lack of comprehensive evaluation. Reference
district in China for example in order to illustrate the specific
evaluation steps and methods based on the hierarchy, at the [8] and [9] proposed a comprehensive decision-making
same time, the case shows that the evaluation method has method on evaluation of network planning, and Reference
strong feasibility and practicability. The planning evaluation [10] and [11] proposed a practical comprehensive hierarchy
hierarchy and method proposed in this paper can not only and method for urban distribution system evaluation on the
seize the key elements influencing the characteristics of
basis of systematic and quantitative comprehensive
network planning, but also describe the planning integrally
and quantitatively, and possess significant engineering evaluation theories. However, the hierarchy mainly focuses
practicability. on evaluating the current network and lacks the evaluation
to the planning network, so it can’t solve the practical
Index Terms--Comprehensive Evaluation Hierarchy; problems of network planning in the future effectively.
Distribution Network; Planning; Quantitative The evaluation hierarchy and method on network
planning proposed in the paper can be used to evaluate not
I. INTRODUCTION only the current network but also the planning network. A

P LANNING is vital for the development of power system; case study has proved that the planning evaluation hierarchy
and the quality of planning has direct influence on and method can describe the planning integrally and
security, reliability and economy of the power system in the quantitatively, and possess significant practicability.
future. Distribution network planning in China has improved
greatly, however, there are some problems needed to be II. PRINCIPLE OF E VALUATION HIERARCHY
solved urgently, such as how to evaluate the results after ESTABLISHMENT
distribution network planning; whether the planning For one thing, the selection of indicators must reflect the
alternative is optimized; how to prove and quantitate the actual situation of network as completely as possible during
effectiveness of a planning after its implementation in the establishment of evaluation index hierarchy. On the
practical network. The major cause leading to the problems other hand, the validity of the information, the difficulty of
above is the lack of comprehensive knowledge and data collection, as well as the quantity of computation
understanding of the planning and also the lack of specific should also be concerned. Therefore, the establishment of
criteria for measuring planning alternatives. Therefore, it is evaluation index hierarchy should meet certain principles as
urgently needed to study quantitative evaluation method for follows:
distribution network planning. (1) Consistency with the purpose of evaluation
The traditional method for urban distribution system (2) Direct measurability
(3) Comparability
Jun Xiao is with Key Laboratory of Power System Simulation and
Control of Ministry of Education, Tianjin University, Tianjin, 300072,
(4) Independence with each other.
China (e-mail: xiaojun@tju.edu.cn). (5) Integral completeness
Yanyan Cui is with Key Laboratory of Power System Simulation and In addition to meeting the above principles, the
Control of Ministry of Education, Tianjin University, Tianjin, 300072, establishment of planning evaluation hierarchy has its own
China (e-mail: cyy_yz@sina.com).
Fengzhang Luo is with Key Laboratory of Power System Simulation characteristics as follows:
and Control of Ministry of Education, Tianjin University, Tianjin, 300072, (1) The hierarchy of planning network pays more attention
China (e-mail: luofengzhang@tju.edu.cn). to the characteristics of the planning and evaluates the
Mingzhi Liu is with Tianjin Electric Power Corporation, Tianjin,
300200, China (e-mail: mingzhi.liu@tepco.com.cn).
network more integrally, the establishment of which is
Jianmin Wang is with Urban Power Supply Company, SMEPC, more concise related to the complexity of the evaluation
Shanghai, 200122, China (e-mail: wangjm@sq.smepc.com).

978-7-900714-13-8/08/ ©2008DRPT 1249

 DRPT2008 6-9 April 2008 Nanjing China
hierarchy of current network.
(2) The establishment of evaluation hierarchy of
distribution network planning ought to be suitable to the
characteristics of the urban development and load
characteristics.
(3) The evaluation of planning network pays more
attention to the relative changes of index value in addition
to the absolute quantity.
III. EVALUATION INDEX HIERARCHY OF DISTRIBUTION
NETWORK PLANNING
Analytic Hierarchy Process (AHP) is one key technique
in the evaluation, which is used in the analysis of the
complex issues and the establishment of evaluation
hierarchy. The sticking point of AHP is to establish clear
level structure to decompose a complex problem and it has
been used in the planning [11]. According to the principle of
the hierarchy establishment, combining the features and
purposes of network planning, applying the basic principle
of AHP, this paper proposes evaluation index hierarchy of
network planning.
Fig. 1. Structure of evaluation hierarchy of distribution network planning
B. Introduction of Index
A detailed introduction to the five primary indices in the
evaluation hierarchy proposed above will be made.
(1) Security
The security of distribution network means the ability to
keep continuous power supply for load against a group of
potential accidents at any time. Just the security for the
maximum load of the year is concerned in the planning.
" 'N-1' analysis" and "Capability of against interruption
of a large area" are used to quantitatively evaluate the
ability above, "N" of which means the total number of
certain important equipments (including power transformer
and feeder .etc), "1"means the number of elements with
fault.
(2) Reliability
The distribution system reliability is the capability of
uninterrupted power supply under the consideration of the
capability of the component, the voltage limitation of the
bus, the limitation of frequency, the scheduling
maintenance outage and the reasonable failure outage.
1250
A. Structure of Index Hierarchy
Security, reliability and economy is the basic requirement
of power supply; the network planning alternative is
generally determined according to the load forecasting with
uncertainty, so planning network should meet the demand
for load of different growth patterns in a certain stage. At
the same time, it is needed to quantitatively analyze the
coordination considering the integrity and consistency of the
network. To sum up, combining the views of electrical
experts, determine the evaluation index hierarchy of
distribution network planning given in Fig. 1.
The hierarchy contains five primary indices, such as
security, reliability, economy, adaptability and coordination.
Each index includes a number of subordinate indicators in
order to quantify the evaluation from different aspects. The
five primary indices constitute a whole which can evaluate
the planning efficiently.
"System average interruption duration", "System average
interruption frequency", "Customer average interruption
duration", "Average service availability" and "Interruption
cost" can deficit the reliability.
(3) Economy
The economic evaluation of network planning
alternatives is an important means to make scientific
decision to the power construction project, reduce and
avoid mistakes in decision, as well as enhance economic
efficiency.
"Operation economy" of network analyzes the power
system from the loss rate of network and the utilization rate
of equipment. "Construction economy" analyzes the
economic benefit brought by the investment in power
network from the aspects of investment and income of
power supply. "Expenses of year" is calculated according to
the network investment, operation and maintenance costs,
as well as the expense of loss.
(4) Adaptability
The network planning is based on the load forecasting
whose uncertainty requires that there is broad space for
 DRPT2008 6-9 April 2008 Nanjing China

development in the network; therefore, it is needed to the same level.

evaluate the ability of network adapting to the load
development.
(5) Coordination
Network is an indivisible whole, partial overloading or
underloading will have a great impact on the security,
reliability and economy of network. The coordination
between the high-voltage network and the medium-voltage
network is necessary; otherwise, the weaker network will
weaken the power supply of the stronger ones.
IV. IMPLEMENTATION STEPS FOR EVALUATION OF
PLANNING NETWORK
Fig. 2 describes the implementation steps for evaluation
of planning network, which include the determination of
indices hierarchy and the evaluation of planning.
It is also needed to give synthetic consideration to the
characteristics of the urban development and load
characteristics while setting scoring standard and
establishing the weight values for all indices. For instance,
the urban district in the development of mid-term with rapid
load growth requires a higher system margin, so it is
appropriate to raise the scoring standard of each index in
"adaptability" and to increase the weight values of them.
V. CASE STUDY
This paper takes 10kV network distribution planning of a
district in China for example in order to illustrate the
specific evaluation steps and methods based on the
hierarchy and test the feasibility and practicability of the
planning evaluation hierarchy and method.
A. Basic Data of the Planning Evaluation
The district area is 348.4 hectares and contains three
35kV substations; its 10kV network is composed mainly
of cables, secondarily of overhead lines. The area is in the
development of the mid-term whose function is mainly
business, education, scientific research and living. 30
appropriate evaluation indices are selected from the
hierarchy according to the characteristic of the case. First
of all, the collection of basic data in planning evaluation is
needed before calculating the indices. Table 1 and table 2
show the date needed in the case.

TABLE І
PARTIAL D ATA OF CURRENT NETWORK

No Classification Name of Date Value Unit

T hrough rate of single power transformer
Fig. 2. Implementation steps for evaluation of planning network 100 %
int errupt ion in 110(35) kV substat ion
T hrough rate of single feeder interruption of
100 %
In the stage of the indices hierarchy determination, it is 10 kV
needed to select appropriate indices, to set suitable 1 Security
Maximum load loss rate of "N-1" 0 %
evaluation criteria and scoring standard, and to establish the Average load loss rate of "N-1" 0 %
T hrough rate of two power transformers
weight values for all indices. 0 %
int errupt ion in 110(35) kV substat ion
The planning evaluation process includes three important Maximum load loss rate of "N-2" 18.85 %
parts: for one thing, to collect the original data such as the Average load loss rate of "N-2" 9.55 %
general situation of the network, planning report and so on; Margin of substation expansion 109.8 %
for another, to evaluate each single index on the current Margin of feeder space 3.92 %
2 Adaptability Margin of channel 3.21 %
network and planning one; in the end, to provide
Margin of power supply 3.42 %
supplementary analysis for the decision-making of Margin of net work expansion 29.58 %
distribution network planning through the comparison of … … … … …
the evaluation results on different planning schemes with
each other. There are 30 basic data of the planning evaluation in
Scores of indices are gained by converting the original Table 1, 6 of which are got from the planning report of the
data of index to a score according to a specific scoring area, 24 of which are calculated according to the report.
standard. 100 basis, 10 basis and 5 basis are usually used in
the scoring standard. There are many methods to determine
the scoring standard, such as fuzzy membership degree
method [11]. Weight values of indices which are
represented by normalized vector depict the importance
among indices of the same level.
Assume that there are two indicators in comparison, if
the former is more important than the latter, the weight of
the former is larger than the weight of the latter, vector (0.6, TABLE П
0.4) can be used to depict it. We can use the method PARTIAL DATE OF PLANNING NETWORK
comparing two indictors with each other or experts deciding
directly to determine the weight values of the attributes of

1251
 DRPT2008 6-9 April 2008 Nanjing China

No Classification Name of Date Value Unit degree method and synthesizing distribution system experts'
T hrough rate of single power transformer
100 %
experience, according to the characteristics of network and
interruption in 110(35) kV substation the purposes of planning evaluation. The same index may
T hrough rate of single feeder interruption
100 % have different evaluation criteria and scoring standards in
of 10 kV
Maximum load loss rate of "N-1" 0 %
the different stages of network development, so it is also
1 Security needed to give synthetic consideration to the characteristics
Average load loss rate of "N-1" 0 %
T hrough rate of two power transformers of the urban development and load characteristics. Table3
0 %
interruption in 110(35) kV substation summarizes the partial scoring standard of the case.
Maximum load loss rate of "N-2" 8.65 % Take "Through rate of single power transformer
Average load loss rate of "N-2" 3.07 %
interruption in 110(35) kV substation" for example to
Margin of substation expansion 109.79 %
clarify Table 3. The urban district requires a higher power
Margin of feeder space 12 %
2 Adaptability Margin of channel 11 %
supply security because it is in the development of
Margin of power supply 4.5 % mid-term, so it can be appropriate to raise the scoring
Margin of network expansion 25.1 % standard of "Through rate of single power transformer
… … … … … interruption in 110(35) kV substation". While the value of
the index is 90%, 60% and 40%, the score of the index is 80,
There are 35 basic data of the planning evaluation in 60 and 40. If the value is 100% and 0%, the score is 100
Table 2, 6 of which are got from the planning report of the and 0.
area, 29 of which are calculated according to the report.
B. Establishment of Scoring Standard
Suitable evaluation criteria and scoring standard for the
selected indices are set by means of fuzzy membership
TABLE Ш
PARTIAL SCORING STANDARD OF COMPREHENSIVE EVALUATION INDEX

Scoring Standard Score 100 95 90 80 75 70 60 40 0

T hrough rate of single power transformer
Proportion(%) 100 90 80 60 40 0
interruption in 110(35) kV subst ation
"N-1"
T hrough rate of single feeder interruption of 10 kV Proportion(%) 100 90 80 60 40 0
analysis
Maximum load loss rate of "N-1" Proportion(%) 0 20 40 60 80
Security Average load loss rate of "N-1" Proportion(%) 0 20 40 60 80
Capability of T hrough rate of two power transformers
Proportion(%) 100 90 80 60 40 0
against interruption in 110(35) kV subst ation
interruption Maximum load loss rate of "N-2" Proportion(%) 0 20 40 60 80
of a large area Average load loss rate of "N-2" Proportion(%) 0 20 40 60 80
… … … … … … … … … … … … …

C. Establishment of Weight Values
The method of expert group decision-making is applied
to determine the weight values of the attributes of the same
level, in order to quantitatively depict the importance
among indices. While in the process of establishment, the
importance of index and the rate of incidence must be
concerned, and the sum of weight values which belong to
indices of the same level must be 1. In addition, we should
consider these influence of the factors such as the
development stage of the urban, load characteristics, the
composition of consumer and so on. For example,
compared with the initial stage of development, the urban
district in the period of rapid growth requires a higher
system margin, so the weight value of "adaptability" should
evaluation results with each other in order to measure the
effect when the planning is virtually implemented.
Firstly, the basic data of the planning evaluation
collected will be converted into scores ranging from 0 to
100 according to evaluation scoring standard which is
determined by the preceding method.
Different scores describe the network evaluated from
different aspects, but it is insufficient in presenting its
overall conditions. Therefore, a calculation must be
literately invoked to get the score of upper level in the
hierarchy by AHP [11], until the overall score is finally
gained. It may be defined as
n
S ( k +1) = ∑S
j =1
(k )
j Wj
(k )
（1）

be larger.
Table 4 summarizes the weight values of partial indices
in the corresponding levels of the case, data in "()" stands
for the weight value. For instance, "(0.4)" in "Maximum
load loss rate of 'N-1' (0.4)" means that the weight value of
"Maximum load loss rate of 'N-1' " is 0.4.
D. Comprehensive Evaluation and Comparison
The current network and planning network in the near
future are comprehensively evaluated by comparing the
TABLE IV
S(k+1) stands for attribute A(k+1)’s score in level k+1, n for
attributes’ number in lever k of A(k+1), Sj(k) for attribute j’s
score in level k of A(k+1), and Wj(k) for j’s weight here.
The higher overall score indicts that the distribution
network’s performances have achieved a higher level in
every aspect such as technical reasonableness, system
security, as well as economics, etc. Table 4 summarizes the
partial results of the distribution network comprehensive
evaluation.

1252
 DRPT2008 6-9 April 2008 Nanjing China

PARTIAL COMPREHENSIVE EVALUATION OF CURRENT AND PLANNING NETWORK

Comprehensive Evaluation Hierarchy Evaluation to the

Classification
Evaluation Index Date Score Effectiveness of a Planning
T hrough rate of single power Number(Seat) 3/3
T hrough rate transformer interruption in 100/100
of "N-1" 110(35) kV substation(0.5) Proportion(%) 100/100
"N-1"
Nnalysis (0.2)100/100 T hrough rate of single feeder Number(Strip) 71/56
100/100
(0.5)100/100 interruption of 10 kV(0.5) Proportion(%) 100/100
Security Maximum load loss rate of "N-1" (0.4) P roportion(%) 0/0 100/100
(0.2)84.2/87.6 Average load loss rate of "N-1" (0.4) P roportion(%) 0/0 100/100
Capability of T hrough rate T hrough rate of two power Number(Seat) 0/0
against of "N-2" transformers interruption in 0/0 Unemproved
(0.2)0/0 110(35) kV substation(1) P roportion(%) 0/0
interruption
of a large area Maximum load loss rate of "N-2" (0.4) Proportion(%) 18.85/8.65 81/91 Improved slightly
(0.5)68.4/75.2 Average load loss rate of "N-2" (0.4) Proportion(%) 9.55/3.07 90/97 Improved slightly
… … … … … … …
Comprehensive
76.42/81.48
Evaluation
Notes: for each index in Tab.5, data in front of "/" stands for the value and score of current network, data in back of "/" stands for the value and
score of planning network; data in "()" stands for the weight value; score increasing more than 20 percent is taken for "improved obviously",
score increasing to 100 is taken for "completely resolved".

The conclusion can be obtained from the evaluation
results compared with the current network ， that the
planning network in the near future has been improved in
some aspects as follows:
(1) Capability of against interruption of a large area: the
capability of transferring load in the planning network
increases nearly by 20 percent, the load loss rate of ’N-2’
analysis is cut down, so the capability of against
interruption of a large area is improved.
(2) Utilization rate of equipment: increases 17 percent,
from 47.12% in current network to 54.8% in planning
network.
(3) Margin of feeder space: 6 positions are reserved in the
planning network, 4 of which are new.
(4) Matching degree of substation capacity: the index is
improved obviously, the score of which increases from 58
to 84.
(5) Load balancing: there is a difference of 41 percent
between the maximum load rate and the minimum load
rate of substation in current network whose load
balancing is poor, while the score of the index in planning
network reaches 100.
To sum up, the analysis of the results has shown that the
planning network improves the primary problems of the
current network, the overall score increases from 76.42 to
81.48, especially the adaptability is markedly improved and
the score of the coordination increases from 64.65 to 80.4,
so the planning alternative is feasible.
VI. CONCLUSION
This paper proposes the evaluation index hierarchy of
distribution network planning on the basis of the analysis of
principle of evaluation hierarchy establishment. Then the
implementation steps for the evaluation of planning
network are given. In the end, this paper takes 10kV
network distribution planning of a district in China for
example in order to illustrate the specific evaluation steps
and methods based on the hierarchy, at the same time, the
case shows that the evaluation method proposed in this
paper has strong feasibility and practicability.
The case study has proved that the planning evaluation
hierarchy and method can not only seize the key elements
influencing the characteristics of network planning, but also
describe the planning integrally and quantitatively, and
possesses significant engineering practicability.
VII. REFERENCES
[1] Wu Kaigui，Wang Shao，Zhang Anbang．The study of reliability
assessment of electric power system using RBF neural network
[J]．Proceedings of the CSEE，2000，20(6)：9-12．
[2] Xie Yinghua，Wang Chengshan．Reliability evaluation of medium
voltage distribution system based on feeder partition method
[J]．Proceedings of the CSEE，2004，24(5)：35-39．
[3] Li Weixing，Wang Peng，Li Zhimin et al．Reliability evaluation of
complex radial distribution systems considering restoration sequence
and network constrains[J]．IEEE Transactions on Power Deliver，
2004，19(2)：753-758．
[4] Liu Wei，Guo Zhizhong．Research on security indices of distribution
networks[J]．Proceedings of the CSEE，2003，23(8)：85-90．
[5] Tang Huizhi，Peng Jianchun．Research on synthetic and quantificated
appraisal index of power quality based on fuzzy theory[J]．Power
System technology，2003，27(12)：85-88．
[6] Jovan Nahman，Dragoslav Peric．Distribution System Performance
Evaluation Accounting for Data Uncertainty[J]．IEEE Transactions
on Power Delivery，2003，18(3)：694-700．
[7] Farghal S A，Kandil M S，Elmitwally A．Quantifying electric power
quality via fuzzy modeling and analytic hierarchy processing[J].
IEEE Proceedings-Generation，Transmission and Distribution，2002，
149(1)：44-49．
[8] Xiao Jun, Luo Fengzhang, Wang Chengshan. Design and Application
of a Multi-Attribute Decision-Making System for Power System
Planning. Power System Technology, 2005, 29(2): 9-13.
[9] Xiao Jun，Wang Chengshan，Zhou Min．An IAHP-based MADM
method in urban power system planning[J]．Proceedings of the
CSEE，2004，24(4)：50-57．
[10] Xiao Jun，Gao Haixia，Ge Shaoyun． Evaluation Method and Case
Study of Urban Medium Voltage Distribution Network [J]．Power
System technology，2005，29(20)：77－81．
[11] Xiao Jun，Gao Haixia，Lin Kai．Comprehensive Hierarchy and
Method for Urban Distribution System Evaluation．2006 China
International Conference on Electricity Distribution

1253
 DRPT2008 6-9 April 2008 Nanjing China

VIII. BIOGRAPHIES

Jun Xiao (1971- ), male, associate professor.

His research interests are in power system
planning, evaluation and distribution automation.
School of Electrical and Automation Engineering
of Tianjin University.
Tel: 86-22-83718291.
Fax: 86-22-83718290.
Email: xiaojun@tju.edu.cn.

Yanyan Cui (1983- ), was born in Baoding,

China, on July 11, 1983. Female, graduate
student in Tianjin University.
Her research interests are in power system
planning and evaluation. School of Electrical and
Automation Engineering of Tianjin University.
Tel: 8613662142323.
Email: cyy_yz@sina.com

Fengzhang Luo (1980- ), male, received the

B.Eng. degree and the M.Eng. degree from
Tianjin University, Tianjin, China, where
he is currently pursuing the Ph.D. degree. His
research interests are distribution system
planning and decision-making, artificial
intelligence application in power systems.
Tel: 8613920902432
Email: luofengzhang@tju.edu.cn

Mingzhi Liu (1973- ), male, Engineer. He

graduated from Tianjin University and worked at
Tianjin Electric Power Corporation, Tianjin,
China.
His research interests are in power system
planning.
Tel: 8613302087601
Email: mingzhi.liu@tepco.com.cn

Jianmin Wang (1964- ), male. He is now a senior engineer in power

system planning at Urban Power Supply Company, SMEPC, Shanghai,
China. His research interest is distribution system planning.

Yinong Li (1968- ), male. He is now a senior engineer in power system

planning at Urban Power Supply Company, SMEPC, Shanghai, China. His
research interest is distribution system planning.

Yubo Gao (1979- ), male. He is now an engineer in power system planning

at Urban Power Supply Company, SMEPC, Shanghai, China. His research
interest is distribution system planning.

Saiyi Wang (1978- ), male. He is now an engineer in power system

planning at Urban Power Supply Company, SMEPC, Shanghai, China,
after getting his PHD degree from Tianjin University, Tianjin, China. His
research interest is distribution system planning.

1254