基于BP神经网络的异步电机匝间短路程度研究  

Study on Extent of Inter-turn Short Circuit in AC Motor Based on BP Neural Network

作　　者：郭西进[1] 孙爱进[1] 许允之[1] 

机构地区：[1]中国矿业大学信息与电气工程学院,江苏徐州221008

出　　处：《煤矿机械》2011年第10期276-278,共3页Coal Mine Machinery

摘　　要：提出了一种针对异步电机匝间短路故障程度的诊断方法。把从异步电机定子端采集的失电残余电压进行频谱分析,提取其中部分高次谐波成分并与其中的基波成分对比,比值作为不同程度的匝间短路故障的特征向量,再利用BP神经网络进行仿真,得出诊断结果。实验表明,该方法能有效、准确地做出电机匝间短路故障程度的诊断。A method for diagnosing extent of inter-turn short circuit in AC motor was proposed.The residual voltage,which was collected form stator winding of AC motor,was analysed by frequency spectrum.Their specific value was used to indicate eigenvalue vector of different inter-turn short circuit fault.Some higher harmonics and fundamental was extracted from frequency spectrum.Then eigenvalue vector was emulated by BP neural network.The result of diagnosing was obtained.The experiment demonstrated effectiveness and high accuracy rate of method,which was used to diagnose extent of inter-turn short circuit in AC motor.

关 键 词：失电残余电压 BP神经网络 异步电机 匝间短路 故障程度 

分 类 号：TM343[电气工程—电机]