基于遗传小波网络的电动机故障诊断

对基本的遗传算法进行了改进,并将其与BP算法相结合构成混合算法应用到电动机故障诊断的小波神经网络训练中。仿真结果表明,该算法有效地解决了小波网络初值设置不合理,极易进入局部极小的区域,以致网络振荡增大、不收敛,而GA算法独立训练神经网络速度缓慢等缺点。用训练过的小波神经网络模型对5组电动机故障进行验证和诊断的仿真结果也表明了,该算法具有较快的收敛速度和较高的计算精度,证实了该算法应用于电动机故障诊断的有效性。

Motor Fault Diagnosis Based on Genetic Wavelet Network

Improves the simple genetic algorithm and combines that genetic algorithm with BP algorithm to the wavelet neural network in the motor fault diagnosis.Simulation result shows the problem was solved that wavelet network settles into local small extremum so easily that the network surging will increase and the network will not be convergent if the initialization is unreasonable,and overcomes the shortcoming that the speed is too slow if use genetic algorithm to train neural network independently.Simulation result of 5 groups' motor fault diagnosis with trained wavelet network indicates the train speed and the diagnosis accuracy improves to some extent all.That shows the validity of that method in motor fault diagnosis.