Distribution System

Planning

1
 Objectives

♦ Introducing the main concepts, planning criteria, and procedure for

distribution system planning.

♦ Introducing how the planning decision is made and the tools and
the algorithms required.

♦ Presenting the main factor and sub-factors affecting the planning

process.

♦ Explaining the procedure for distribution system planning.

2
 Objectives (Cont.)

♦ Presenting the main classification for distribution system planning.

♦ Reviewing different existing models for the distribution planning

process.

♦ Reviewing traditional and advanced planning approaches and

techniques.

♦ Identifying the reasons for the deviations between the theoretical

(academic) distribution planning techniques and the practical ones.

3
 Outlines
♦ Definitions and Terms.
♦ Factors Affecting Distribution System Planning.
♦ Distribution System Planning Process .
– General.
– Distribution System Planning Procedure.

♦ Planning Process Classification.

– Short Range Planning Process.
– Long Range Planning Process.

♦ Planning Models, Techniques, and Approaches.

– Traditional Operational Research Based Approaches.
– Algorithm-Based Techniques.
– AI-Based Techniques.

♦ Deviations between Research and Practical Techniques.

4
 Definitions and Terms
Objective
♦ The objective of distribution system planning is to assure that the
growing demand for electricity, in terms of increasing growth rates and
high load densities, can be satisfied in an optimum way by additional
distribution systems, from secondary conductors through the bulk power
substations, which are both technically adequate and reasonably
economical .
Procedure
♦ The planner must determine the load magnitude and its geographic
location. Then the distribution substation must be placed and sized in
such a way to serve the load at maximum cost effectiveness by
minimizing losses and construction costs, while considering the
constraints of service reliability.
5
 Criteria
The planning criteria represent the companies’ policies and obligations
to the consumers. They can be summarized in the following criterion:
• The service continuity must be ensured.
• The maximum allowable peak load voltage drop at the most remote
customers’ locations on the secondary network must be within the acceptable
limits as defined by the appropriate governing body.
• The maximum allowable voltage dip occasioned by the starting of a motor of
specified starting current characteristics at the most remote customers’
locations on the secondary must be within the acceptable limits as defined by
the appropriate governing body.
• The service Reliability as high as economically possible.
• The power losses must be minimum.
• The maximum allowable peak load must be met.

6
 Decision
♦ The decisions made by the planner includes one or more items from the
following list:
1. Optimum substation expansions.
2. Optimum substation locations.
3. Optimum substation sizes.
4. Optimum transformers number and sizes.
5. Optimum load transfers between substations and demand centers.
6. Optimum feeder routes, design, and sizes to supply the given loads.
7. Optimum switchgear for the proposed distribution system .

♦ In order to make such decisions some factors have to be included in the

planning process such as; Kirchhoff’s current and voltage laws, feeders
and substations capacities, voltage drop constraints, and variable costs.

7
 Tools and Programs
Group 1

♦ Load flow programs.

♦ Radial or loop load flow programs.

♦ Short circuit and fault current calculation programs.

♦ Voltage drop calculation programs.

♦ Total system impedance calculation programs

8
 Tools and Programs
Group 2

♦ Load forecasting programs.

♦ Voltage regulation and regulator setting programs.

♦ Capacitor planning programs.

♦ Reliability programs.

♦ Optimal sizing and siting algorithms.

9
 Long Range Planning (6 to 20 years)
Steps for long range planning

1. To perform Load forecast.

2. To consider Demand side management requirements.
3. To perform reliability studies.
4. To identify service standards (steady state voltages and transient voltage).
5. To select distribution substations (sizes and locations).
6. To select number of transformers in each substations.
7. To select feeders (number and routing).
8. To formulate the problem.
9. To set the constraints.
10. To use appropriate software to solve the problem.
10
 Short Range Planning (1 to 5 years)
Steps for short range planning

1. Load analysis and validation.

2. Load modeling.
3. Peak load modeling.
4. Single contingency analysis.
5. Voltage limits.
6. Equipment loading limits.
7. Reactive power analysis.
8. Problem analysis and solution.
9. Voltage analysis.

11
 Factors Affecting Distribution
System Planning
♦ Load Forecasting.
♦ Substation Expansion.
♦ Substation Site Selection.
♦ Total Cost.
♦ Voltage Selection.
♦ Feeder Route Selection.
♦ Number of Feeders Selection.
♦ Conductor Size Selection.
12
 Distribution System Planning
Process
The planning process can be considered consisting of five
main stages.

♦ Stage 1: Problem Identification

♦ Stage 2: Goals Identifications
♦ Stage 3: Alternatives Identifications
♦ Stage 4: Alternatives Evaluation
♦ Stage 5: Best Alternative Selection
13
 Errors in the Alternative Evaluation Stage

This stage might face some error resulting form the fact that:

♦ Planners didn’t consider all alternatives.

♦ Planners didn’t consider all criteria, factors, and attributes.
♦ Planners didn’t use all available resources to evaluate all
alternatives.
♦ Planners didn’t evaluate the alternatives on equal judgment
bases.

14
Distribution System Planning
Procedure
Spatial Load
Forecast

Substation
Planning

Authorized Distribution Coordinated Long

Projects System Planning Range Plan

Feeder Planning

Customer Level
Planning
Distribution system planning Process Elements
15
Distribution System Planning Decision
Flow Process
Load Forecast

Is the System No Is it Required to No

Performance Build Additional
Good? Substation?

Yes

Substation Site
Selection

Present System
No Feedback
Expansion

Design New
System
Yes

Yes
Is the Total Cost
Acceptable?
This is the Required
System

16
 Procedure for Substation Site Selection

Service Region

Candidate Areas
Identification

Sites for Future

Unsuitable Sites Candidate Sites
Evaluation

Proposed Sites

17
 Distribution System Planning Stages
Type of Coordinated
Period Results Tools
Planning Levels

Customer level Simulation and trending forecast

Small areas load forecast for
Spatial Load planning, marketing methods, end-use analysis of load /
---- 1 – 25 years peak demand, annual kWh,
Forecast customer types, ..etc.
plans, and customer value, and corporate
distribution plans. forecast database.

Substation project schedule Substation selection optimization

Sub-transmission
Substation Short and Short: 3 – 7 years coordinated with sub- applications, combined sub-
and distribution
Planning long range Long: 5 – 20 years transmission and distribution
levels planning.
transmission-substation-feeder
plans, and long range plans system optimization applications.

Evaluation of feeder level cost Multi-feeder optimization

Substation and
Feeder System and performance impact, long- programs, combined substation-
Long range 5 – 20 years customer levels
Planning range feeder level cost estimation
planning.
feeder system optimization
for budget. applications.

feeders route maps, pole

Substation, feeder Feeder design CAD systems, feeder
Short Range location, construction
Short range 1 – 5 years system and customer optimization programs (single or
Feeder Planning equipments details, and required
levels planning. multi-feeder).
authorizations and schedules

End-use load models, integrated

Schedules for customer-side
Customer Level Short and resource optimization programs,
1 – 25 years resource projects or programs, All planning levels
Planning long range
long range plans.
customer response models, valued-
based planning analysis.

Multi-level plan that covers

essential elements of long-range
Coordinated Attentions to details and good
----- ----- sub-transmission, substation, All planning levels
Multi-Level Plan distribution, and customer level
communication.
plans. 18
 Distribution System Planning
Classification

The planning process can be classified into two main types

♦ Short Range Planning Process.

♦ Long Range Planning Process.

19
 Short-range and long-range planning
periods for transmission and distribution
planning
System Level Capacity Short Range Long Range
Planning Planning Period
Period
Generation > 150 MVA 10 30
Generation < 50 MVA 7 20
EHV Transmission >= 500 kV 10 30
Transmission To 345 KV 8 25
Sub-Transmission 34 kV – 99 KV 6 20
Distribution Substation - 6 20
Feeder System - 6 20
Primary Feeders - 4 12
Lateral and Small Feeders - 1 4
Services Transformers and - 0.5 2
Secondary
Customer Level - 5 20

20
 Traditional Distribution System
Planning Techniques
There are two common traditional Operational Research-based
techniques that have been widely applied in network flow
optimization, supply chain planning, and energy system
logistics for decades:

♦ Transportation Approach.
♦ Transshipment Approach.

21
 Transportation Approach

This technique by Crawford et al solves the problems of substation

location and sizing

∑ ∑ (MFD ) X
m n
♦ Mathematical Model: Minimize i =1 j =1 ij

♦ Subject to

∑
n
X ij = S i i= 1, 2, …….., m
j =1

∑
m
X ij = Di j= 1, 2, …….., n
i =1

♦ Where: Si is the supply at node i, Dj is the demand at node j, Xij is the amount
of power flow from node i to node j, MFD is the minimum feasible distance, n
is the number of demand nodes, and m is the number of substation nodes
22
 Transshipment Approach
This technique was proposed by Wall et al and was used for
solving the problems regarding distribution feeder routing

∑ ∑
m n
♦ Mathematical Model: Minimize i =1 j =1
C ij X ij

♦ Subject to
i= 1, 2, …….., m
∑ j =1 X ij = S i + k
n

∑
m
X ij = Di + k j= 1, 2, …….., n
i =1

L ≤ X ij ≤ U i= 1, 2, …….., m, and j= 1, 2, …….., n

♦ Where: k is the number larger than the sum of supplies and the sum of demand,
N is the total number of nodes, Cij is the Cost of sending power from node i to
node j, L is the lower capacity limit, and U is the upper capacity limit.
23
 Modern Distribution System Planning Techniques
(Multiple Criteria Decision Making (MCDM) Technique)
1
Distribution Network

Develop Feasible Alternative Solution

Input Specification Evaluation Criteria Selection

- Preliminary Budget - Capital Cost - Supply Availability
- Planning Period - System Security - Annual Energy Losses
- Environmental Impact - Capacity Constraints

Performance Analysis
Initial Network Analysis & Problem Set Requirement
- Network Electrical Characteristic
- Load Forecasting - Load curves
- Equipment Failure Rates
- New Network Connection - Equipment Restoration Times
- Environmental Constraints
- Distributed Generation Location
- Overloaded Equipment
Solution Characteristics
- Annual MWh Losses - Capacity Constraints MWh
- Supply Availability - Environmental Impact
- Supply Security - Investment Cost

Develop Problem Set Options

No All Feasible Yes
solutions MCDM Analysis
been Assessed?

Review of Results

1 Final Design Components

24
Comprehensive Optimization Model
for Distribution System Planning

25
Comprehensive Optimization Model
for Distribution System Planning

26
Comprehensive Optimization Model
for Distribution System Planning

27
Comprehensive Optimization Model
for Distribution System Planning

28
Comprehensive Optimization Model
for Distribution System Planning

29
Comprehensive Optimization Model for
Distribution System Planning

30
 Comprehensive Optimization Model
for Distribution System Planning
Solution Approach:

♦ Linear Programming (LP) or Mixed-Integer Linear Programming

(MILP) for a simplified version.
♦ Nonlinear Programming (NLP) or Mixed-Integer Nonlinear
Programming (MINLP) for more detailed power flow models.
♦ Heuristic methods (e.g., Expert System-Based Techniques, Genetic
Algorithms, Particle Swarm Optimization, or Ant Colony approach)
for large-scale systems with complex constraints.

31
 Expert System-Based Techniques
Structure of the developed expert system

Expert System
Users

Ruled Based Expert System

Reliability
Computation Inference Engine Knowledge Base
Algorithm

Expert Knowledge

Minimum Cost Minimum Loss Interview with

Algorithm Algorithm Utility Planners

32
Expert System-Based Techniques
Load allocation algorithm
Identify all Feasible Load
Switching Pattern

Select One Possible Load

Switching Pattern

Minimum Cost Minimum Power Loss

Algorithm Algorithm

Feasible Yes Yes Feasible

Load Relocation Load Relocation
Plan? Plan?

No No

Compute Reliability Indices,

Cost and Loss

Store Load Allocation Plan,

Loss, Cost, and
Reliability Indices

More Yes
Load Switching
Patterns?

No

Produce Results 33
Expert System-Based Techniques
Structure of knowledge-based expert system
User

User Interface

Knowledge Explanatory
Network Data
Acquisition Subsystem

Knowledge Base
Inference Machine
Simulation Method

Knowledge
Acquisition
Subsystem

Expert Result
34
Ant Colony System (ACS) Approach
Start

Ant Colony (k) sends ants for exploration until all

nodes are connected and generates a candidate
grid

Process is repeated independently for a

predetermined number of ant colonies (m)
generating an equal number of candidates grids
Process Repeated for the set of expeditions

Globe Revision Rule

Convergence Achieved?
No Or
Maximum Number of
Expeditions Sets Reached?

Yes

Produce Results
35
 Deviations between Research and Practical
Techniques
The reality is that these sophisticated techniques are not commonly used
in real life situation and the reason for that can be attributed to the
following reasons:
1. There is a time lag between research and the application of this research.

2. Distribution planning depends to a great extent on the planner’s experience and his prior knowledge
of the system.

3. In the new era of the deregulated market the distribution planning problem has changed
tremendously rendering the most sophisticated techniques obsolete before they are put into practice.

4. The new trend now in planning has very short term objectives such as reliability enhancement and
new interties for peak load shedding, which is very different from the theoretical planning methods’
goal.
36
 Deviations between Research and Practical
Techniques (Cont.)
In general the main differences between theoretical planning
methods and practical methods are:

1. Theoretical planning methods are asset management-based methods which

depend on database of the system and numerical/ statistical models for the
distribution system.

2. Practical methods are budget-based methods which depend on promoting higher

performance of the system in order to increase revenues and profit within the
limits set by the regulatory body.

37
 Deviations between Research and Practical
Techniques (Cont.)
The main deficiencies associated with the theoretical approaches for
planning are:
• Inaccurate load forecasting technique results.
• Inaccurate estimations for cost, land availability, and construction times.
• Failure to handle sudden changes in industry.
• Failure to produce good economic measures using least cost plans.
• Failure to use adequate cost plans results in appropriate system operating cost and higher
investments.
• Failure to include generation units not owned by the utility.
• Developed planning techniques are not capable of dealing with uncertainties in the system
behavior.
• Failure to incorporate other energy sector planning activities.
• Ignoring customers’ reactions.
38
 Deviations between Research and Practical
Techniques (Cont.)
Foote and McDonald summarize the main reasons for these
deficiencies as follows:
♦ The formulation of the problem is analytical and can’t take into consideration the
stochastic nature of the planning problem.

♦ These methods are not capable to integrate the new electric power system
structure into their equations.

♦ These methods are not capable to handle distributed generation in the system and
as such their results don’t reflect the economical benefit or loss due to the
distributed generation existence on the system.
39
 Deviations between Research and Practical
Techniques (Cont.)
The following requirements and features are required to
enhance future planning techniques in order to make them
suitable for practical application:
♦ More ability to deal with multiple criteria.
♦ More flexibility in providing multiple and diverse solutions.
♦ Ability to produce whole system solutions.
♦ Ability to adapt to the new power system structure and to interact with it.
♦ Make the whole process less user-dependent and yet incorporate more of the
experts’ experience
♦ Ability to deal with uncertainties.
♦ Ability to provide weight to future activities.
♦ Usage of developed advanced simulation and optimization techniques.
40
 References
♦ T. Gonen, Electric Power Distribution System Engineering, McGraw-Hill, 1986.
♦ S.K. Khator, L.C. Leung, “Power Distribution Planning: A Review of Models and Issues”. IEEE
Transaction on Power Systems, Vol. 12, No. 3, August, 1997, pp. 1151 -1159.
♦ H.L. Willis, “Power Distribution Planning” Reference Book, Marcel Dekker Inc., New York, 1997.
♦ B.M. Crawford and S.B. Holt, “A Mathematical Optimization Technique for Locating and Sizing
Distribution Substations and Deriving their Optimal Service Areas”, IEEE Transaction on Power System
Apparatus, Vol. 94, No. 2, March/April, 1975, pp. 230 – 235.
♦ D.L. Wall, G.L. Thompson, and J.E.D. Northcote-Green “An Optimization Model for Planning Radial
Distribution Network”, IEEE Transaction on Power System Apparatus, Vol. 98, No. 3, 1979, pp. 1061 -
1068.
♦ G.L. Thompson, and D.L. Wall, “A Branch and Bound Model for Choosing Optimal Substation Location”,
IEEE Transaction on Power System Apparatus, Vol. 100, No. 5, 1981, pp. 2683 - 2688.
♦ T. Gonen, and B.L. Foote, “Distribution System Planning Using Mixed-Integer Programming”, Proceedings
of the IEE, part C, Vol. 128, No.2, 1981, pp. 70-79.
♦ M. Ponnavaikko, K.S. Prashasa, and S.S. Venkata, “Distribution System Planning through a Quadratic
Mixed Integer Programming Approach”, ”, IEEE Transaction on Power Delivery, Vol. 2, No. 4, 1987, pp.
1157 – 1163.
♦ K. Aoki, K. Nora, T. Satoh, M. Kitagawa, and K. Yamanaka, “New Approximate Optimization Method for
Distribution System Planning”, IEEE Transaction on Power Systems, Vol. 5, No. 1, 1990, pp. 126 - 132.
♦ K. Nora, T. Satoh, K. Aoki, M. Kitagawa, and K. Yamanaka, “Multi-Year Expansion Planning for
Distribution System”, IEEE Transaction on Power Systems, Vol. 6, No. 3, 1991, pp. 952 - 958.

41
 References (Cont.)

♦ K. Nora, T. Satoh, H. Kuwabara, K. Aoki, M. Kitagawa, and T. Ishihara, “Distribution Systems Expansion
Planning by Multi-Stage Branch Exchange”, IEEE Transaction on Power Systems, Vol. 7, No. 1, 1992, pp.
208 - 214.
♦ S.K. Goswami, “Distribution Systems Planning Using Branch Exchange Technique”, IEEE Transaction on
Power Systems, Vol. 12, No. 2, 1997, pp. 718 - 723.
♦ G.J. Peponis, and M.P. Papadopoulos, “New Dynamic, Branch Exchange Method for Optimal Distribution
System Planning”, IEE Proceedings on Generation, Transmission and Distribution, Vol. 144, No. 3, 1997, pp.
333 – 339.
♦ E. Masud, “An Interactive Procedure for Sizing and Timing Distribution Substation Using Optimization
Techniques”, IEEE Transaction on Power System Apparatus, Vol. 93,No. 5, 1974, pp. 1281 - 1286.
♦ M.A. Abu El-Magd et al., 14 Annual Modeling and Simulation Conference, Pittsburgh, PA, USA, 1983.
♦ D.I. Sun, D.R. Farris, P.J. Cote, R.R. Shoults, and S.S. Chen, “Optimal Distribution Substation and Primary
Feeder Planning via the Fixed Charge Network Formulation”, IEEE Transaction on Power System Apparatus,
Vol. 101,No. 3, 1982, pp. 602 - 608.
♦ M.A. EL-Kady, “Computer Aided Planning of Distribution Substation and Primary Feeders”, IEEE
Transaction on Power System Apparatus, Vol. 103, No. 6, 1984, pp. 1183 - 1189.
♦ W.H. Quintana, H.K. Temraz, and K.W. Hipel, “Two-Stage Power System Distribution Planning Algorithm”,
Proceedings of the IEE, part C, Vol. 104, No. 1, 1993, pp. 17 – 29.
♦ S.K. Khator, L.C. Leung, “Power Distribution Planning: A Review of Models and Issues”, IEEE Transaction
on Power Systems, Vol. 12, No. 3, 1997, pp. 1151 - 1159.

42
 References (Cont.)

♦ J. Chen, and Y. Hsu, “An Expert System for Load Allocation in Distribution Expansion Planning”, IEEE
Transaction on Power Delivery, Vol. 4, No. 3, 1989, pp. 1910 - 1917.
♦ Y. Hsu, and J. Chen, “Distribution Planning using a Knowledge-Based Expert System”, IEEE Transaction on
Power Delivery, Vol. 5, No. 3, 1990, pp. 1514 - 1519.
♦ G. Brauner, and M. Zobel, “Knowledge Based Planning of Distribution Networks”, IEEE Transaction on
Power Systems, Vol. 9, No. 2, 1994, pp. 942 - 948.
♦ W.M. Lin, Y.S. Su, and M.T. Tsay, “Genetic Algorithm for Optimal Distribution System Planning”,
Proceedings. of the International Conference on Power System Technology, 1998. POWERCON '98, Vol.
1, 18 - 21 Aug., 1998, pp. 241 – 245.
♦ M. Skok, D. Skrlec, and S. Krajcar, “Genetic Algorithm and GIS Enhanced Long Term Planning of Large
Link Structured Distribution System”, Large Engineering Systems Conference on Power Engineering 2002,
LESCOPE 02, 26-28 June, pp. 55 – 60.
♦ P. Espie, G.W. Ault, G.M. Burt, and J.R. McDonald, “Multiple Criteria Decision Making Technique Applied
to Electricity Distribution System Planning”, IEE Proceedings on Generation, Transmission and Distribution,
Vol. 150, No. 5, 2003, pp. 527 – 535.
♦ J.F. Gomez, H,M. Khodr, P.M. DeOliveira, L. Ocque, J.M. Yusta, R. Villasana, and A.J. Urdaneta, “Ant
Colony System Algorithm for the Planning of Primary Disetribution Circuits”, IEEE Transaction on Power
Systems, Vol. 19, No. 2, 2004, pp. 996 - 1004.
♦ R.E. Brown, “Web-Based Distribution System Planning”, IEEE Power Engineering Society Summer
Meeting, 2002, Volume: 1 , 21-25 July 2002, pp. 362 - 365.
♦ G.W. Ault, C.E.T. Foote, J.R. McDonald, “Distribution System Planning in Focus” IEEE Power Engineering
Review, January, 2002, pp.60 -63.
43