摘要
针对电网规划的多目标性,提出一种改进的多种群遗传算法(Poly-PopulationGeneticAlgorithm,简称PPGA)。子种群对应独立的目标函数,父种群对应归一化的目标函数,精华种群采用最优个体保存法保存父子种群中的最优个体,并作为收敛依据,保证了算法全局收敛性。父子种群采用改进自适应SGA,进行独自寻优,并通过迁徙算子以一定频度进行信息交换,打破"封闭竞争"。在原有配电网络的基础上实现网络的有效扩展,满足当前配电网络规划的多目标多阶段的要求。算例结果表明,该方法是有效的。
This paper suggests an improved poly-population genetic algorithm(PPGA) with an aim at multi-object distribution electric network planning. Son population corresponding to the independent goal functions, and father population corresponding to the normalization goal function, quintessence population adopting optimum individual maintaining the law father and son optimum individual of population taken as convergence basis can guarante the overall convergence property of algorithms. The father and son population adopts the improved adaptive SGA, seeking for the excellence independently, and the operators carry on information to exchange with certain frequentness through migrating,whereby breaking 'sealing competition'. The valid expansion of the network on the basis of the original distribution network is realized in this paper. The requests of multi-stages and multi-goal in present distribution network planning are statisfied. The results from the computed examples indicate that this method is effective and feasible.
出处
《西安理工大学学报》
CAS
2005年第1期69-72,共4页
Journal of Xi'an University of Technology
关键词
电网规划
多种群遗传算法
多目标优化
electric network planning
poly-population genetic algorithm
multi-objective optimization
