Economic dispatch of power systems using an adaptive charged system search algorithm

Abstract

In this article, an adaptive charged system search (ACSS) algorithm is developed for the solution of the economic dispatch problems. The proposed ACSS is based on the charged system search (CSS) which is a meta-heuristic algorithm utilizing the governing Coulomb law from electrostatics and the Newtonian laws of mechanics. Here, two effective strategies are considered to present the new ACSS. The first one is an improved initialization based on opposite based learning and subspacing techniques. The second one is Levy flight random walk for enriching updating process of the algorithm. Many types of economic dispatch cases comprising 6, 13, 15, 40, 160 and 640 units generation systems are testified as benchmarks ranging from small to large scale problems. These problems entail different constraints consisting of power balance, ramp rate limits, prohibited operating zones and valve point load effects. Additionally, multiple fuel options and transmission losses are included for some test cases. Moreover, simple constraint handling functions are developed in terms of penalty approach which can readily be incorporated into any other meta-heuristic algorithm. Results indicate that the ACSS either outperform or perform well in comparison to the CSS and other optimizers in finding optimized fuel costs.