基于深度学习的电机轴承故障诊断研究

针对目前已有的电机轴承故障诊断算法对于人工干预和专家经验的依赖,以及故障诊断工作的复杂度逐渐的提高。文章提出了基于深度学习中卷积神经网络的故障诊断算法,使用原始振动数据作为网络模型的输入对其进行训练以发挥其强大的自学习能力。根据振动数据的特点和实验对比选择模型的结构和参数,进而通过深层次网络结构的卷积操作以实现对原始振动数据的特征提取,最终在输出端利用Softmax分类器输出分类结果。通过实验验证表明,该方法对于轴承故障分类准确率能够达到99.8%,对比其他方法具有很好的分类效果。

Research on Fault Diagnosis of Motor Bearing Based on Deep Learning

In view of the dependence of the existing motor bearing fault diagnosis algorithms on manual intervention and expert experience,the complexity of fault diagnosis is gradually increasing.In this paper,a fault diagnosis algorithm based on deep learning convolutional neural network is proposed.The original vibration data is used as the input of the network model to train it so as to exert its strong self-learning ability.The structure and parameters of the model were selected according to the characteristics of vibration data and the experimental comparison,and then the features of the original vibration data were extracted through the convolution operation of the deep network structure.Finally,Softmax classifier was used to output the classification results at the output end.The experimental results show that this method can achieve 99.8%accuracy in the classification of bearing faults.