摘要
三相异步电机定子发生匝间短路故障时,定子电流信号的频谱中会增加特定频率的故障特征。本文针对三相异步电机建立有限元模型并植入故障,采集定子电流信号并用FFT(Fast Fourier Transformation,快速傅里叶变换)提取电流中的故障特征,使用LSTM网络(Long Short-Term Memory,长短时记忆网络)进行预测表明,此方法针对定子匝间短路故障的识别准确率达93%,为该故障诊断提供了一种有效的方案。
When an inter-turn short circuit fault occurs in the stator of a three-phase asynchronous motor,fault characteristics with specific frequencies will be added to the frequency spectrum of the stator current signal.Thus,a finite element model of the three-phase asynchronous motor is established firstly and the inter-turn short circuit fault is simulated.The stator current signal is then collected and the fault characteristics are extracted with Fast Fourier Transformation(FFT).The Long Short Term Memory(LSTM)Network is adopted for prediction.Experimental simulation shows that the accuracy of the proposed method in identifying the inter-turn short circuit fault can reach 93%.It provides an effective solution for the diagnosis of such fault.
作者
李昂
王彤
张健
LI Ang;WANG Tong;ZHANG Jian(CHN Energy Dadu River Maintenance and Installation Co.,Ltd.,Leshan 614000,China)
出处
《水电与新能源》
2024年第4期22-25,共4页
Hydropower and New Energy
关键词
三相异步电机
定子匝间短路
LSTM网络
three-phase asynchronous motor
inter-turn short circuit fault of stator
LSTM network
