首页|基于FFT-LSTM的抽水蓄能发电机定子匝间短路故障诊断方法

基于FFT-LSTM的抽水蓄能发电机定子匝间短路故障诊断方法

扫码查看
定子短路故障是抽水蓄能发电机常见的故障之一,其会对发电机的性能和安全性产生严重影响,为确保抽水蓄能发电机安全稳定运行,提出基于FFT-LSTM的抽水蓄能发电机定子短路故障诊断方法.建立抽水蓄能发电机定子绕组匝间短路故障模型,分析定子绕组匝间短路故障时,发电机定子电、磁相关状态.以抽水蓄能发电机定子绕组匝间短路故障时的三相电流信号为依据,基于磁势相等原理将三相电流变换成两相电流后,利用FFT转换定子两相电流的时域信号为频域信号,获取故障电流频谱图输入LSTM网络中进行处理,输出抽水蓄能发电机定子绕组匝间短路故障诊断结果.实验结果表明,该方法可以更好地区分抽水蓄能发电机正常与故障状态,实现抽水蓄能发电机定子绕组匝间短路故障诊断,且故障诊断的交叉熵损失低.
Diagnosis Method of Stator Short Circuit Fault of Pumped Storage Generator Based on FFT-LSTM
Stator short circuit fault is one of the common faults in pumped storage generators,which can have a serious impact on the performance and safety of the generator.To ensure the safe and stable operation of pumped storage generators,a FFT-LSTM based stator short circuit fault diagnosis method is proposed.Establish a stator winding inter turn short circuit fault model for pumped storage generators,and analyze the electrical and magnetic related states of the generator stator during short-circuit faults.Based on the three-phase current signal during the stator winding inter turn short circuit fault of a pumped storage generator,the three-phase current is transformed into two-phase current based on the principle of magnetic potential equality.The time-domain signal of the stator two-phase current is converted into a frequency-domain signal using FFT,and the fault current spectrum is obtained and input into the LSTM network for processing.The diagnosis result of the stator winding inter turn short circuit fault of the pumped storage generator is output.The experimental results show that this method can better distinguish between normal and fault states of pumped storage generators,achieve stator winding inter turn short circuit fault diagnosis of pumped storage generators,and have low cross entropy loss in fault diagnosis.

pumped storagea generatorstator short circuitFFTLSTMfault diagnosis

李树峰、林文峰、李甲骏、张斌、罗全兵、李国宾、苏毅、屠黎明

展开 >

北京四方继保自动化股份有限公司,北京市 100085

河北张河湾蓄能发电有限责任公司,河北省石家庄市 050026

国网新源集团有限公司,北京市 100052

抽水蓄能 发电机 定子短路 FFT LSTM 故障诊断

国网新源控股有限公司科技项目资助

SGXYKJ-2022-007

2024

水电与抽水蓄能
国网电力科学研究院

水电与抽水蓄能

影响因子:0.247
ISSN:2096-093X
年,卷(期):2024.10(1)
  • 15