改进免疫算法在电力系统有功优化中的应用

Improved Immune Algorithm Applied to Active Power Optimization of Electric Power System

免疫算法(ImmuneAlgorithm,IA)是在免疫系统识别多样性的启发下所设计出的一种新的多峰值函数的寻优算法,它具有抗原识别、记忆、抗体的抑制和促进等显著特点,能实现精确控制群体多样性和特异性。IA 将目标函数和约束条件比作抗原,将问题的解比作抗体。通过亲和度的计算来评价抗体并促进或抑制抗体的产生,减小了进化过程陷入局部最优解的可能性;通过抗原记忆,提高了局部搜索能力,加快了计算速度。将 IA 用于 IEEE30节点系统的有功最优潮流计算,并与传统牛顿算法的计算结果进行了比较,结果表明 IA 能够以更快的速度得到最优解。

Immune Algorithm is a new optimization algorithm imitating the immune system to solve the multimodal function optimization problem. Comparing with the genetic algorithm, the proposed algorithm based on immune principle exists following salient features such as antigen recognition, memory mechanism, the boost or restriction of antibody generations, etc. In immune algorithm the objective function and the constraints are assimilated to the antigens and thc solution of the problem is assimilated to the antibody. Through the calculation of affinity the antibody is evaluated and the boost or restrain of its generation is determined, thus, the pos- sibility of the evolutionary process falls into local optima is decreased. Through the memory mechanism the ability of local search is improved; therefore, the calculation is speeded up. The IA is used to the calculation of active power optimization in an actual 30-bus system and the calculation results by IA is compared with that by traditional genetic algorithm （GA）. The comparison results show that the more optimal solution can be obtained by IA and the performance of IA is far better than that of GA.