雷电定位计算的粒子群优化方法  

Method of particle swarm optimization in lightning location estimation

作　　者：胡志祥[1] 赵文光[1,2] 文银平[1] 朱宏平[1,2] 

机构地区：[1]华中科技大学土木工程与力学学院,湖北武汉430074 [2]华中科技大学湖北省控制结构重点实验室,湖北武汉430074

出　　处：《华中科技大学学报（自然科学版）》2011年第4期129-132,共4页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：'十一五'国家科技支撑计划重点项目(2008BAC36B00)

摘　　要：针对雷电定位问题,引入粒子群优化(PSO)算法用于雷电定位计算.给出了利用这种算法进行雷电定位的计算步骤,并提出用PSO算法和传统迭代算法协作计算雷电位置的方法.通过数值仿真分析和电网雷击事故定位计算分析了PSO算法的性能.结果表明:该算法能克服传统迭代方法易于发散的缺点,稳定并精确地求解出雷电发生位置;该算法的计算量大于迭代方法,但比网格搜索法要大幅减小;利用PSO算法给出雷电定位初始值,再用迭代方法求解可保证计算稳定并减少计算量.The concept of particle swarm optimization （PSO） for lightning location estimation was presented after current lightning location problems were reviewed. The algorithm imitated a bird flock searching for food, which constantly approached to the optimum solution. The procedure of the PSO algorithm to find the lightning location was introduced, as well as a hybrid method which used PSO algorithm providing a good initial value for iterative method. The performance of the proposed method was confirmed by computer generated data and practical transmission-line lighting faults data of a lighting location system for power industry. The PSO algorithm was proved to perform better than conventional iterative method for its stability and grid search method for its efficiency. The PSO method can provide a good initial value for iterative method for convergence and computational tractability.

关 键 词：雷电定位 粒子群优化 到达方向 到达时间差 网格搜索 初始值 

分 类 号：TN953.7[电子电信—信号与信息处理] TM863[电子电信—信息与通信工程]