基于深度学习的电机故障诊断

Motor Fault Diagnosis Based on Deep Learning

故障诊断在保证电机的稳定运行中占据着非常重要的地位,因此,故障诊断在当前的研究中是一个热点。该研究利用短时傅里叶变换把一维的振动信号转换成二维的时频图,进而解决电机轴承的振动信号的非线性和不稳定性问题,并且作为卷积神经网络的输入,通过对故障特征信号的直接提取,来形成样本数据集,通过卷积神经网络与softmax多分类器来建立故障诊断模型,在Python中验证该算法优化的准确性,证明了该算法可以提高电机故障诊断的准确率。

Fault diagnosis plays a very important role in ensuring the stable operation of motor.Therefore,fault diagnosis is a hot topic in current research.In this study,the short-time Fourier transform is used to transform the one-dimensional vibration sig-nal into a two-dimensional time-frequency diagram,so as to solve the nonlinear and instability problems of the vibration signal of the motor bearing.As the input of the convolutional neural network,the sample data set is formed through the direct extraction of the fault feature signal.The fault diagnosis model is established by convolution neural network and softmax multi-classifier,and the ac-curacy of the algorithm optimization is verified in Python,which proves that the algorithm can improve the accuracy of motor fault di-agnosis.