改进仿电磁学算法在多目标电网规划中的应用

提出基于仿电磁学(electromagnetism-like mechanism,ELM)算法的高容错性多目标电网规划方法。为提升算法的性能,引入被动聚集思想对基本模型的全局寻优能力进行了改善。采用自适应权重、自适应变异和精英策略等措施来改善仿电磁学算法的收敛性。针对电源规划、变电站布点和负荷需求均已知的电网规划,将反映电网经济性和可靠性指标的多目标函数转化为求电网规划最小耗费的单目标模型。对一个18节点系统进行了十进制编码,计算结果证明了该算法能有效地解决电网规划这类含离散变量的大规模组合优化问题和提高规划方案的综合满意度。通过与遗传算法、基本ELM的仿真对比,改进的ELM模型在寻优效率和容错性方面具有明显优势。

Application of Improved Electromagnetism-Like Mechanism in Multi-Objective Power Network Planning

Based on electromagnetism-like mechanism(ELM) algorithm,a high fault-tolerant method for multi-objective power network planning is proposed.To enhance the performance of ELM algorithm,its global search ability of the basic model is improved by introducing the idea of passive congregation in.The measures such as adaptive weight,adaptive variation and elitist strategy are adopted to improve the convergence of ELM algorithm.For the power network planning in which the generation planning,substation locations and load demand are known,the multi-objective function reflecting economy and reliability index of power network is turned into single objective model to solve the minimum cost of network planning.Applying decimal coding to a 18-bus system,calculation results show that the proposed algorithm can effectively solve such a large-scale combinatorial optimization problem containing discrete variables and improve the overall satisfaction of the planning scheme.Comparison of simulation results from genetic algorithm and basic ELM algorithm show that the improved ELM algorithm possesses obvious advantages in search efficiency and fault-tolerance.