摘  要：为进一步提高异步电动机故障检测的准确性，将人工神经网络应用于异步电动机故障检测。通过提出一种基于BP神经网络的电机故障检测方法，设计了适合该检测系统的网络结构。仿真结果表明：相对于其他算法，该网络结构具有更快的学习速度和更高的学习精度，完全适用于电动机故障检测。   关键词：人工神经网络；异步电动机；故障检测；模式识别；模式分类   中图分类号：O242.1；TP751        文献标识码：A          文章编号：1672-4984（2008）03-0135-03   Fault diagnosis system for asynchronous electromotor based on artificial neural network   LIU Zhao-you， QIU Shi-hui， WANG Qi   （Department of Electrical Engineering，Chengdu Electromechanical College，Chengdu 610031，China）   Abstract： Artificial neural network was applied to detect the fault of asynchronous electromotor in order to improve the accuracy of asynchronous electromotor fault diagnosis. One fault diagnosis method for electromotor based on BP neural network was proposed，and the network structure was designed for this diagnosis system. The simulation result indicates that the net structure has mush faster learning speed and more superior learning precision compared with other algorithm. It is entirely practical for the fault diagnosis system of the electromotor.   Key words： Artificial neural network； Electromotor； Fault diagnosis system； Pattern recognition； Pattern classification   Editor：liyan