Impact of different inertia weight functions on particle swarm optimization algorithm to resolve economic load dispatch problems

Abstract

Dispatching generation units plays a valuable role in the economic operation of the plant. The Economic Load Dispatch Problems (ELDPs) deal with such economic operation of the plant. The Particle swarm optimization (PSO) technique becomes much more popular than other to resolve ELDPs. To control explosion phenomenon, Inertia Weight (IW) is used in the PSO algorithm. The IW may be a positive constant or a time varying function. In this article six different IW namely constant, linearly decreasing, natural exponent strategy 1 and 2, random and simulated annealing are used in PSO algorithm to resolve ELDPs of IEEE 5, 14 and 30 bus systems. The best, worst, average and their standard deviation costs are calculated for 20 trial runs using MATLAB programming. The average numbers of iteration and average computational time have been also examined. The analysis of results shows that the use of simulated annealing IW for IEEE-5 bus system and the use of natural exponent IW strategy 2 for IEEE-14 bus system and IEEE-30 bus systems in PSO algorithm provide better result with less computational time.