基于小波变换和神经网络的电机综合保护

Integrated Protection for Motors Based on Wavelet Transform and Neural Network

本文在分析大型电动机故障诊断与保护理论的研究现状和发展趋势的基础上,讨论了以过流保护为基础的电动机常规保护、基于对称分量法的新型综合保护的大型电动机在线监测和故障诊断。提出一种利用小波变换和人工神经网络(ANN)实现自适应电流保护的方法。该方法充分利用了小波变换强大的时频分解能力、优异的奇异性检测能力和人工神经网络所具有的强大的自适应能力、学习能力和模式识别能力,实现时电动机中的各种故障情况的识别,对网络进行训练,结果表明,该方法具有可靠、优越性、可行性。

Based on the analysis of the actuality and development of fault diagnosis and protection theory for a large capacity motor, the paper discussed the conventional overcurrent protection for motor and on-line monitoring and fault diagnosis based on symmetrical components methods＇ new type integrated protection.The authors present a new method of the application of wavelet transform and artificial neural network in the adaptive current protection.It means that many kinds of faults identification of motor could be solved by using strong time-frequency decomposition capabilities and outstanding singularity detection capability of wavelet transform and strong adaptive ability and studing ability and pattern recognition of ANN.By network trained, the simulation result shows that the methods has high reliability and superiorities and feasibilities.