精英克隆局部搜索的多目标动态环境经济调度差分进化算法

Multiobjective dynamic economic emission dispatch differential evolution algorithm based on elites cloning local search

为有效解决复杂多目标动态环境经济调度问题,提出一种基于精英克隆局部搜索的多目标动态环境经济调度差分进化算法。以传统的差分进化(differential evolution, DE)算法为框架,为了提高DE算法的开采和探索能力,增设精英群的克隆和突变机制,采用动态选择方式确定精英群,有效增强算法的全局搜索能力。数值试验以IEEE-30的10机、15机系统为测试实例,并将提出的算法与三种代表性算法比较。结果表明,新算法所获的Pareto前沿具有较好的收敛性和延展性,可为电力系统调度人员提供更灵活的决策方案。

An efficient multiobjective differential evolution algorithm based on elites cloning local search scheme was proposed to solve complex dynamic economic emission dispatch. The conventional differential evolution(DE) algorithm was used as the framework of the proposed algorithm. A cloning operator was developed to enhance the exploration and exploitation ability of elites in the DE algorithm. The elite population to be cloned was established by a dynamic selection mechanism for enhancing the global search ability of the proposed algorithm. To validate the effectiveness of the proposed algorithm, the IEEE 30 bus 10-generator and 15-generator systems were studies as test cases in numerical experiments. The simulation results indicated that the Pareto-optimal front obtained by the proposed algorithm presented a superior performance in convergence and extension over the other reported results recently. As a result, the results were able to provide decision solutions more extensively for decision-makers in power system dispatch.