Power System Reactive Power Optimization Based on an Improved Immune Algorithm

Chuang-xin GUO,Cheng-zhi ZHU,Bo ZHAO,Yi-jia CAO

DOI: https://doi.org/10.3321/j.issn:1000-1026.2005.15.005

2005-01-01

Abstract:Based on clonal selection principle, an improved immune algorithm (ⅡA) is proposed to seek power system reactive power optimization. This algorithm generates the next population under the guidance of the previous superior antibodies in a small and a large neighbor domains respectively, in order to realize the parallel global and local search capabilities. Through analysis of Markov chain, the proposed method is proved to be global convergent. In order to handle discrete variables in reactive power optimization, the paper proposes a simple truncation technology. The method is only performed in evaluating the fitness function. The proposed algorithm applied for optimal power system reactive power is evaluated on both an IEEE 30-bus and a practical 118-bus power system. Simulation results show that the ⅡA method is more efficient than some earlier reported approaches, and it can offer higher quality solutions to reactive power optimization within reasonable computing time.