A REVIEW OF ADVANCES IN ECONOMIC DISPATCH USING

ARTIFICIAL NEURAL NETWORKS

Engr. Tahir Nadeem Malik

Elect. Engg. Deptt. University of Engg.& Tech., Taxila (Pakistan) Student Member IEEE

Dr.Aftab Ahmad

National Power Control Centre WAPDA, Islamabad (Pakistan)

Engr.Aftab Ahmad

Elect. Engg. Deptt. University of Engg.& Tech., Taxila (Pakistan)

Key words: Economic Dispatch (ED), Artificial Neural Network (ANN), Hopfield Neural Network, Prohibited Zones,

Artificial Intelligence (AI).

Abstract:

Economic Dispatch Problem (EDP) has been discussed with reference to the developments based on

Artificial Neural Networks (ANN) approaches. A selected survey / overview on Economic Dispatch using

Artificial Neural Network within the IEE/IEEE publications frame work have been presented.

1 INTRODUCTION

Power Economic Dispatch (ED) is necessary and

vital step in power system operational planning. This

is non-linear constrained optimization problem and

is defined as the process of allocating generation

levels to the generating units in the mix, so that the

system load may be supplied entirely and most

economically (Happ, 1977). It is on-line function,

carried out after every 15-30 minutes or on request

in Power Control Centers. In this paper efforts have

been made to develop brief survey on the advances

in Economic Dispatch Problem based on Artificial

Neural Network Techniques. The paper is organized

as follows: The Economic Dispatch (ED) problem

formulation is introduced in Section II. Section III

addresses the selected survey on Economic Dispatch

Problem based on Artificial Neural Network (ANN)

Techniques. Lastly there is conclusion.

2 ECONOMIC DISPATCH

PROBLEM

The Economic Dispatch in its simplest form is

formulated as:

Minimize

(1)

∑

iiT

)P(CC

Where: C

: Total cost

: generator fuel cost of the ith generating

unit.

Subject to equality constraint:

(2)

∑

iLD

PPP

Where: P

: total load.

: transmission loss given by

∑

PPaP

Where: a

: transmission loss coefficient.

Inequality constraints:

i (min)

≤ P

i (max)

(3)

Where: P

i(min)

: the minimum generation power

i (max)

: the maximum generation power.

354

Nadeem Malik T., Ahmad A. and Ahmad A. (2004).

A REVIEW OF ADVANCES IN ECONOMIC DISPATCH USING ARTIFICIAL NEURAL NETWORKS.

In Proceedings of the First International Conference on Informatics in Control, Automation and Robotics, pages 354-357

DOI: 10.5220/0001137103540357

 SciTePress

Economic Dispatch Problem (EDP) has been solved

by optimization techniques (Chowdhury, 1990).

3 ECONOMIC DISPATCH

PROBLEM USING ANN

During the last decade the applications of ANN to

various problem of Power system have increased

considerably. Economic dispatch problem has been

solved by using the Hopfield Neural Network

architecture. In the discussion to follow a brief

survey / overview on ANN based techniques has

been given selectively within the framework of

IEE/IEEE publications.

(Matuda, 1989) presented the representation of

large numbers in neural networks and its application

to economical load dispatching of electric power.

This work represents one large number by one

neuron which converges finally to any real values by

using Hopfield Network. For the validation EDP has

been mapped in the proposed model and tested for

four node system supplying 30, 000 MW load.

(Park, 1993) presented Economic Load

Dispatch for Piecewise Quadratic Cost Function

using Hopfield Neural Network. EDP has been

mapped into Hopfield neural network with

differential synchronous transition model and

modified sigmoidal function. The proposed method

has been tested for Convex and Non convex

economic dispatch problems.

(Fukuyama, 1994) presented an application of

neural network to dynamic dispatch using multi

processors. Dynamic Economic Dispatch has been

carried out by using Gaussian Machine. The

modeling has been programmed in parallel C and

tested on transputer.

(King, 1995) presented Optimal Environmental

Dispatching of Electric Power System via an

Improved Hopfield Neural Network model. A

simulator has been developed and a criterion for

selecting its parameters has been discussed to obtain

an optimal dispatch in the minimum amount of

iterations.

(Djukanovic, 1996) presented a method for

resolving the real-time economic dispatch problem.

A suitable topology for the neural-net based on the

multi-layered perceptron has been developed and an

appropriate training method based on the back

propagation algorithm has been used. The proposed

method has been applied to two test systems. Hybrid

intelligent system consisting of GDR neural-net

architecture and Gold Works II based expert system

has been proposed for the solution of

environmental/economic dispatch problem

(Yalcinoz, 1997) presented Large Scale

Economic Dispatch using Improved Hopfield Neural

Network. Gee’s Hopfield Neural Networks (GHN)

has been modified to solve EDP. Proposed model

has been tested on 3, 20, 40, 80, 120,160 and 240

units. The execution time and number of iterations

have been reduced compared to classical methods.

(Su, 1997) presented a Fast-Computation

Hopfield Method to Economic Dispatch of Power

Systems. The method employs a linear input-output

model for neurons. EDP has been formulated in such

a way that direct computation instead of iterations

for solving the problems becomes possible. 3 and 13

bus systems have been tested.

(Yalcinoz, 1998) presented Neural Networks

Approach for solving Economic Dispatch Problem

with Transmission Capacity Constraints. Gee and

Prager’s (GP) method has been modified in order to

solve ED with transmission capacity constraints. The

proposed method (PHN) has achieved efficient and

accurate solutions for two-area power systems with

3, 4, 40 and 120 units.

(Lee, 1998) developed two different methods ---

the slope adjustment and bias adjustment methods,

in order to speed up the convergence of the Hopfield

Neural Network System. To guarantee and for faster

convergence, adaptive learning rates have also been

developed by using energy functions and applied to

the slope and bias adjustment methods. The results

of the traditional, fixed learning rate and adaptive

learning rate methods have been compared for

economic load dispatch problems.

(Walsh, 1999) presented Augmented Hopfield

Network for Constrained Generator Scheduling.

This paper presents an augmented Hopfield Neural

Network scheduling algorithm that unites the unit

commitment and generation dispatch functions. This

algorithm successfully considers ramp rate,

transmission and fuel constraints in addition to the

more common constraints. Model has been tested on

system consisting 17 thermal and 2 hydro units.

(Yalcinoze, 1999) presented Security Dispatch

using the Hopfield Neural Network. A mapping

process has been formulated and a computational

method for obtaining the weights and biases has

been described using a slack variable technique for

handling inequality constraints..

(Liang, 1999) developed re-dispatch approach

based on the Hopfield Neural Network considering

the dynamic dispatch problem that involve the

allocation of system generation optimally among

dispatchable generating units while tracking a load

curve and observing power ramping response rate

limits of the units, system spinning reserve

requirements. This method has been successfully

applied to utility system.

A REVIEW OF ADVANCES IN ECONOMIC DISPATCH USING ARTIFICIAL NEURAL NETWORKS

355

(Lee, 2000) presented Real Power Optimization

with Load Flow using Adaptive Hopfield Neural

Network. Instead of using the typical B-coefficient

method, actual load flow to compute the

transmission loss accurately.

(Su, 2000) presented New Approach with a

Hopfield Modeling Framework to Economic

Dispatch. The weighting factors associated with the

terms of the energy function can be either

appropriately selected or directly estimated in the

proposed model. The proposed method has been

tested on 3-bus and 13-bus system.

(Su, 2000) presented A Hopfield Model to

Economic Dispatch having Special Units. This paper

presents a Hopfield model with three strategies to

solve the economic dispatch (ED) problems having

prohibited operating zones. Application of the

proposed approach has been demonstrated using a

15-unit system with 4 units having prohibited zones.

(Altun, 2000) presented Constrained Economic

Dispatch with Prohibited Operating Zones: A

Hopfield Neural Network Approach. A new

mapping process has been used and a computational

method for obtaining the weights and biases is

described using a slack variable technique for

handling inequality constraints. The proposed

approach has been demonstrated on 18-unit system

with 4 units having prohibited zones.

(Hartati, 2000)

presents a summary of algorithms

that have been proposed for the application of the

Hopfield Neural Network to the Economic Load

Dispatch problem.

(Bastos, 2002) presented Modified Hopfield

Network in which internal parameters of the neural

network are computed using the valid-subspace

technique, which guarantees convergence to

equilibrium points that represent a solution for the

ED problem. Simulation results and a comparative

analysis involving a 3-bus test system have been

presented to illustrate efficiency of the proposed

approach.

4 CONCLUSIONS

Artificial Intelligence Tools are being used to solve

the EDP. These are gaining popularity over the

solution methods based on optimization theory due

to their strengths. The Hopfield Neural Network

architecture is dominating for the various aspects of

EDP. In (Park, 1993) a Hopfield neural network is

proposed to solve the classical economic dispatch

problem with non-convex cost function. The

computation effort for solving the problem is high

due to large number of iterations to obtain the

optimality. In (Su, 1997) an analytic Hopfield

method reducing considerably this computation

effort is proposed. However the method is not

applied to non-convex cost functions. In (Yalcinoze,

1998) a neural network approach for solving

Economic Dispatch with transmission capacity

constraints has been proposed. (Su, 2000) is the

extension of (Su, 1997) in the sense it incorporates

transmission loss. (King, 1995) and (Yalcinoze,

1999) incorporates the environmental and security

aspects in Hopfield model respectively. (Lee, 1998)

improves convergence and guarantees the

convergence. In (Lee, 2000) B coefficients have

been replaced by load flow equation. In (Altun,

2000) & (Hartati, 2000) there are Hopfield

approaches for dealing the prohibited zones problem

in ED. This selected review reveals that the

advancement in ANN approaches gradual,

systematic and gaining maturity for different aspects

of EDP and there is the potential for the use of ANN

to deal Economic dispatch problem.

REFERENCES

IEEE Working Group Report, 1981. Description and

bibliography of major economy security functions Part

I: Description. In IEEE Transactions PAS, pp.211-

214.

IEEE Working Group Report, 1981. Description and

bibliography of major economy security functions Part

II: Bibliography (1959-1972). In IEEE Transactions

PAS, pp.215-223.

IEEE Working Group Report, 1981. Description and

Bibliography of Major Economy Security Functions

Part III: Bibliography (1973-1979). IEEE

Transactions PAS, pp.224-235.

Happ H.H, 1997. Optimal power dispatch -- A

comprehensive survey. In IEEE Transactions on PAS,

vol.PAS-96, no.3, pp.841-854.

Chowdhury B.H. & Rahman Salfur 1990. A review of

recent advances in economic dispatch. In IEEE

Transactions on Power Systems,vol.5, no.4, pp.148-

1259.

Matuda, M. & Akimoto, Sand Y., 1989. The

representation of large numbers in neural networks

and its application to economical load dispatching of

electric power. In ICNN, vol. 1, pp.587-592.

Park, J. H., Kim, Y.S., Eom, I. K. & Lee, K. Y., 1993.

Economic load dispatch for piecewise quadratic cost

function using Hopfield neural network. In IEEE

Transactions on Power System, vol.8, no.3, pp.1030-

1038.

Fukuyama Yoshikazu & Ueki Yoshiteru, 1994. An

application of neural network to dynamic dispatch

using multi processors. In IEEE Transactions on

Power Systems, vol.9, no.4, pp1759-1765.

ICINCO 2004 - INTELLIGENT CONTROL SYSTEMS AND OPTIMIZATION

356

King Denise T., El-Hawary M. E. & El-Hawary Ferial

1995. Optimal environmental dispatching of electric

power systems via an improved Hopfield neural

network model. In IEEE Transactions on Power

Systems, vol.10, no.3, pp 1559-1565

Djukanovic Miodrag, Calovic Milan, Milogevic Borka &

Sobajic Dejan J., 1996. Neural net based real-time

economic dispatch for thermal .power plants. In IEEE

Transactions on Energy Conversion, vol.11, no.4,

pp.755-761.

Yalcinoz, T. & Short, M. J., 1997. Large-scale economic

dispatch using an improved Hopfield neural network.

In IEE Proceedings Generation, Transmission and

Distribution, vol.144 ,no.2, pp. 181-185.

Su Ching-Tzong & Chiou Gwo-Jen, 1997. A fast-

computation Hopfield method to economic dispatch of

power systems. In IEEE Transactions on Power

Systems, Vol.12, No.4, pp.1759-1764.

Yalcinoz T. & Short, M. J., 1998. Neural networks

approach for solving economic dispatch problem with

transmission capacity constraints. In IEEE

Transactions on Power System, vol.13, no.2, pp.307-

313.

Lee Kwang Y., Sode-Yome Arthit & Park June Ho, 1998.

Adaptive Hopfield neural networks for economic load

dispatch. In IEEE Transactions on Power Systems,

vol. 13, no.2, pp. 519-526.

Walsh, M. P. & O’Malley, M. J., 1999. Augmented

Hopfield network for constrained generator

scheduling. In IEEE Transactions on Power System

vol.14, no.2, pp.765-771.

Yalcinoz T., Short, M.J. & Cory, B. J., 1999. Security

dispatch using the Hopfield neural network. In IEE

Proceedings on Generation, Transmission and

Distribution, vol.146, no.5, pp.465 -470.

Liang Ruey-Hsun, 1999. A neural-based re-dispatch

approach to dynamic generation allocation. In IEEE

Transactions on Power Systems, vol.14, no.4,

pp.1388-1393.

Lee, K.Y., Nurogul, Faith, M., & Sode-Yome Arthit,

2000. Real power optimization with load flow using

adaptive Hopfield neural network. In Engineering

Intelligent Systems, vol.8, no.1, pp. 53-58.

Su Ching-Tzong & Lin Chien-Tung, 2000. New approach

with a Hopfield modeling framework to economic

dispatch. In IEEE Transactions on Power Systems,

vol.15, no.2, pp. 541 -545.

Su Ching Tzong, Lee Chu-Sheng & Chiou Gwo-Jen,

2000. A Hopfield model to economic dispatch of

power systems having special units. In Proceedings of

IEEE International Conference on Industrial

Technology, vol.1, pp.562 -566.

Altun, H., Hasan, U. & Yalcinoz, T., 2000. Constrained

economic dispatch with prohibited operating zones: a

Hopfield neural network approach. In Electrotechnical

Conference.2000, MELECON 2000, Vol.2, pp. 570 -

573.

Hartati Rukmi Sari & El-Hawary M.E., 2000. A summary

of Hopfield neural network to economic load dispatch.

In Electrical and Computer Engineering, 2000

Canadian Conference, vol.2, pp.707-711.

de Silva I. N., Nepomuceno, L. & Bastos T. M., 2002.

Designing a modified Hopfield network to solve an

economic dispatch problem with nonlinear cost

function. In Proceedings of the 2002 International

Joint Conference on Neural Network, Vol.2 , pp. 1160

-1165.

A REVIEW OF ADVANCES IN ECONOMIC DISPATCH USING ARTIFICIAL NEURAL NETWORKS

357