LARGE-SCALE OPTIMAL VAR PLANNING BY HYBRID SIMULATED ANNEALING GENETIC ALGORITHM/

This paper presents a fast near-global optimization technique called h ybrid simulated annealing/genetic algorithm (HSAGA) for solving optima l VAR planning problems. The HSAGA incorporates simulated annealing in to genetic algorithms to improve both performances at the same time. T herefore, it has the ability to find the near-global optimal solution in a finite time. Moreover, the solution time is much less than that o f the conventional simulated annealing method. The HSAGA was applied t o a practical power system, Taiwan Power System (Tai-Power System), wi th satisfactory results, and a comparison between HSAGA, simulated ann ealing, genetic algorithms and the hybrid partial gradient descent/sim ulated annealing method was also presented. (C) 1998 Elsevier Science Ltd. All rights reserved.