SOLVING NON-CONVEX ECONOMIC DISPATCH USING WATER CYCLE

ALGORITHM

Sadaqat Naveed1, Shaikh Saaqib Haroon2, Naveed Ahmed Khan3

ABSTRACT

Economic dispatch (ED) is an extremely challenging part of power system operational planning.
ED is defined as a complex constrained engineering optimisation problem that aims to calculate
power generation of the generating units in an integrated power system for minimum generation
cost. Although ED has been expressed traditionally as a convex optimisation problem, it has become
highly complex due in modern power systems to non-smooth (or non-convex) objective functions
and various newly added constraints. This paper presents implementation of water cycle algorithm
(WCA) on ED problem. WCA is a meta-heuristic population based algorithm which has been recently
introduced for solving constrained engineering optimisation problems. WCA has been tested on
both non-convex and convex test systems in the presented paper. The non-convex test systems consist
of six and thirteen units whereas convex system consists of six units. The effectiveness of WCA has
been verified by comparing the simulation results with other algorithms presented in the literature.

Keywords: economic dispatch; water cycle; convex optimisation; transmission losses; power
generation; steam turbine.

1. INTRODUCTION

Economic dispatch (ED) is an extremely challenging constrained engineering optimisation problem

that aims to calculate power generation of the generating units in a power system at a minimum
generation cost. In the past, conventional optimisation techniques such as equal incremental cost
criteria, base load method, lambda iteration method, gradient search method, dynamic programming
were used to solve ED problems [1-4]. Practically, the cost curves of thermal machines are non-
convex due to valve point loading effect, multi-fuel options and prohibited operating zones. Owing
to these complications, conventional optimisation techniques can easily converge into local minima
[5]. Therefore, the conventional techniques are not considered as an appropriate tool for non-convex
ED problems. To solve such type of ED problems, many stochastic and hybrid techniques have been
proposed by the researchers such as genetic algorithm (GA) [6], simulated annealing (SA), evolutionary
programming (EP) [7], particle swarm optimisation (PSO) [8].
1 Post-graduate student, Department of Electrical Engineering, Bahria University, Islamabad, Pakistan, Ph. +923005500461,
Email: sadaqatnaveed@yahoo.com.
2 Assistant Professor, Department of Electrical Engineering, University of Engineering and Technology, Taxila, Pakistan, Ph. +923335221287,
Email: saaqib.haroon@uettaxila.edu.pk.
3 PhD student, Department of Electrical Engineering, COMSATS Institute of Information Technology, Islamabad, Pakistan,
Ph. +923144990664, Email: naveedahmed512@yahoo.com.
Manuscript received on 3rd December 2015, reviewed and accepted on 4th March 2016 as per publication
policies of NED University Journal of Research. Pertinent discussion including authors’ closure will be
published in March 2017 issue of the Journal if the discussion is received by 31st August 2016.
NED UNIVERSITY JOURNAL OF RESEARCH - APPLIED SCIENCES, VOL XIII, NO. 2, 2016 31