Analysis of Economic Load Dispatch using improved partical swarm optimization

Particle Swarm Optimization is a stochastic optimization algorithm, but it converges to local optima, especially in some complex issue like optimization of high dimension function. It has been observed that the traditional particle swarm optimization algorithms converses rapidly during the initial stage of a search, but in course of time becomes steady considerable and gets trapped in a local optima. But this research paper presents four evolutionary optimisation models (IPSO 1, 2, 3, 4) based on the particle swarm optimization algorithms for Economic Load Dispatch considering cost of generation. Comparative analysis suggests that IPSO (Improved Particle Swarm Optimization) significantly improves the performance with less no of iteration. In the last version of IPSO, we have moved acceleration coefficient for personal factor Cp and global factor Cg in opposite direction (i.e. Cp maximum to minimum and Cg minimum to maximum), while keeping other parameter with some constant value, which shows that there is tremendous reduction in no of iteration. All different IPSO has been implemented to ECONOMIC LOAD DISPATCH to get optimum value of cost with less no of iteration.