基于改进WNN分析功率曲线的S700K转辙机故障诊断

Fault diagnosis of S700K switch machine based on improved WNN analyses power curve

基于S700K转辙机常见故障下的功率曲线提出一种将小波变换、改进型遗传算法与神经网络相结合的故障诊断方法。用相应故障模式下的功率信号进行正交小波分解,把结果作为神经网络的输入特征向量,利用改进的遗传算法优化BP神经网络的参数,用训练好的BP神经网络进行故障诊断。研究结果表明:该方法可以有效的运用到S700K转辙机的故障诊断中,并提高转辙机故障诊断的精度与速度。

In this paper,based on the power curve under the common fault of S700K switch machine,a fault diagnosis method based on wavelet transform,improved genetic algorithm and neural network was proposed.Orthogonal wavelet decomposition was carried out by using the power signal in the corresponding fault mode.The result was taken as the input eigenvector of the neural network,and then the improved genetic algorithm was used to optimize the parameters of the BP neural network.Finally,the trained BP neural network was used for fault diagnosis.The results show that the method can be effectively applied to the fault diagnosis of S700K switch machine and improve the accuracy and speed of fault diagnosis of switch machine.