基于遗传算法优化小波神经网络的井下电缆故障测距方法

摘要

According to the problems that difficult to position when single phase grounding fault happens at underground coal mine power cable and the low reliability and low distance measuring precious for existing distance measuring method,the genetic algorithm optimized wavelet neural network based on transient electric valve after fault is proposed,which realizes the method of underground coal mine cable single phase grounding fault distance measuring.Through different transition resistance,fault distance situation＇s simulation results show that the wavelet neural network based on genetic algorithm optimization can obtain accurate fault location results,and compared with the fault location methods using wavelet neural network based on BP algorithm,it performs better on location precision and convergence time.%针对煤矿井下电网发生单相接地故障后定位困难,现有测距方法可靠性及测距精度低等问题,提出基于故障后暂态电气量由遗传算法优化小波神经网络,实现井下电缆馈线单相接地故障测距的方法。通过不同过渡电阻,故障距离情况的仿真结果表明,基于遗传算法优化的小波神经网络能实现准确故障测距,且较之基于BP算法小波神经网络的测距方法,其在测距精度和收敛时间方面表现更优。