基于支路交换和声搜索算法的配网重构  

Distribution Network Reconfiguration Based on Branch Exchange Harmony Search Algorithm

作　　者：荣德生[1] 胡举爽 韩淑敏[1] RONG Desheng;HU Jushuang;HAN Shumin(Faculty of Electrical and Control Engineering,Liaoning Technical University,Huludao 125105,China)

机构地区：[1]辽宁工程技术大学电气与控制工程学院,葫芦岛125105

出　　处：《电源学报》2018年第2期159-164,共6页Journal of Power Supply

摘　　要：采用标准和声搜索算法HSA(harmony search algorithm)进行配网重构时,在搜索最优方案的过程中容易陷入局部最优,且搜索结果受初始参数值的影响。为此,引入结合启发式规则的支路交换法对新和声进行局部搜索,生成第2个新和声,增加了和声记忆库的多样性,使得算法能够逃离局部最优,进而提高了算法的全局搜索能力;采用基于环路分解的编码方式对和声编码,保证初始解集中每个和声都是可行解。采用IEEE 33节点系统进行算例分析,与标准和声搜索算法、遗传算法进行对比分析,结果表明该算法的收敛更迅速,可以快速逃离局部寻优,且寻优性能不受初始参数的影响,验证了改进算法满足可实现性较高的寻优性能。In the distribution network reconfiguration using a standard harmony search algorithm(HSA),the algorithm will easily be trapped by local optimums,and the searching results are also affected by the initial parameter values.To solve this problem,branch exchange method is introduced to the local search of a new harmony and further generate a second one,which increases the diversity of harmony memory and helps the proposed method jump out of local optimums,thus improving the global searching ability of the algorithm;the coding of harmony is realized using the encoding method based on loop decomposition,which ensures that every harmony in the initial solution set is a feasible solution.An IEEE 33-node system is adopted as a numerical example.Compared with the standard HSA and genetic algorithm,the proposed algorithm has a faster convergence speed,and it can quickly avoid falling into local optimums.Moreover,the optimization performance is not affected by the initial parameters.It is verified that the improved algorithm can realize a higher searching performance.

关 键 词：配网重构 和声搜索算法 环路分解 支路交换法 

分 类 号：TM73[电气工程—电力系统及自动化]