Parallel tabu search for capacitor placement in radial distribution systems

Abstract

This paper proposes a new method for determining capacitor placement in distribution systems. The proposed method makes use of parallel tabu search to evaluate better solution in terms of computational effort and solution accuracy. The capacitor placement is hard to solve in a sense of global optimization due to the high nonlinear and mixed integer problem. To solve the problem efficiently, this paper focuses on tabu search that is one of efficient meta-heuristics. Tabu search is better than the conventional methods due to the adaptive memory that allows to escape from local minimum of the cost function. In this paper, parallel tabu search is proposed to improve the performance of tabu search in large-scale systems. Parallel tabu search considers both decomposition of solution neighborhood and multiplicity of tabu lengths. The proposed method is successfully applied in sample systems. A comparison is made between the proposed and conventional methods.