摘要
文章以线圈电流波形的时间和电流值为特征量,断路器5种典型故障为输出量,采用改进黑猩猩算法(Improved Chimp Optimization Algorithm,ICOA)对长短时记忆(Long Short Term Memory,LSTM)神经网络的三个关键参数进行优化,构建了基于ICOA-LSTM的高压断路器故障诊断模型。采用断路器故障数据进行仿真,并与现有断路器故障诊断模型进行对比分析。对比测试结果表明,ICOA-LSTM模型的诊断精度更高,计算时间更短,验证了ICOA-LSTM模型的优越性和有效性。
This paper takes the time and current value of the coil current waveform as characteristic variables,and five typical faults of circuit breakers as output variables.The Improved Chimp Optimization Algorithm(ICOA)is used to optimize the three key parameters of long short-term memory(LSTM)neural network,and a high-voltage circuit breaker fault diagnosis model based on ICOA-LSTM is constructed.Simulations were conducted using circuit breaker fault data and compared with existing circuit breaker fault diagnosis models.The comparative test results show that the ICOA-LSTM model has higher diagnostic accuracy and shorter calculation time,verifying the superiority and effectiveness of the ICOA-LSTM model.
作者
金枝洁
方艳
吴卓伦
JIN Zhijie;FANG Yan;WU Zhuolun(State Grid Huangshi Power Supply Company,Huangshi 435000,China)
出处
《安徽电气工程职业技术学院学报》
2024年第3期45-52,共8页
Journal of Anhui Electrical Engineering Professional Technique College
关键词
高压断路器
故障诊断
改进黑猩猩算法
长短时记忆神经网络
正确率
high-voltage circuit breaker
fault diagnosis
improved chimp optimization algorithm
long short-term memory neural network
accuracy
作者简介
金枝洁(1993-),女,湖北黄石人,本科,助理工程师,从事变电运维工作;通信作者:吴卓伦(1992-),男,湖北钟祥人,本科,工程师,从事变电运维及检修工作。
