PSO-ELM算法的井下漏电故障选线方法  

Line selection method for underground leakage fault based on PSO-ELM algorithm

作　　者：胥良 何士刚 Xu Liang;He Shigang(School of Electrical&Control Engineering,Heilongjiang University of Science&Technology,Harbin 150022,China)

机构地区：[1]黑龙江科技大学电气与控制工程学院,哈尔滨150022

出　　处：《黑龙江科技大学学报》2025年第1期140-146,共7页Journal of Heilongjiang University of Science And Technology

摘　　要：为了研究井下漏电故障选线多判据融合选线,提出粒子群算法优化极限学习机(ELM)的故障选线方法。利用Mablab/Simulink仿真软件搭建1140 V的井下供电系统模型,获取零序电流信号的无功、基波和暂态特征,经过故障测度函数计算得到故障测度数据经粒子群算法优化后输入至ELM神经网络模型,通过训练输出选线结果。结果表明,文中漏电选线准确率高达98.33%。该方法判断精度高,速度快,可以满足井下漏电故障选线对可靠性与速动性的要求。This paper seeks to investigate the method for the line selection underground leakage fault with the multi criteria fusion,and proposes a fault line selection method based on the particle swarm optimization algorithm optimizing the extreme learning machine(ELM).The study involves building a 1140 V underground power supply system model by using Matlab/Simulink simulation software to obtain the reactive power characteristics,fundamental amplitude characteristics,and transient characteristics of the zero-sequence current signal for obtaining the fault measurement data by calculating the fault measurement function;inputting the ELM neural network model optimized by particle swarm optimization algorithm,and outputting the line selection results by training.The results show that this method has high accuracy and fast speed,and meets the requirements of reliability and speed for underground leakage fault line selection.

关 键 词：漏电故障 粒子群算法 极限学习机 故障测度 

分 类 号：TD68[矿业工程—矿山机电]