基于CNN-SVM的滚动轴承故障诊断

Bearing Fault Diagnosis Method Based on CNN-SVM

针对滚动轴承故障诊断,对其故障特征进行准确提取是提高诊断准确率的关键步骤。以原始时域振动信号为对象,提出一种基于卷积神经网络(CNN)和支持向量机(SVM)的轴承故障诊断模型。首先,将原始时域振动信号输入二维卷积神经网络进行自适应特征提取;其次,取全连接层结果作为支持向量机的输入,利用支持向量机对其进行分类;最后,经过实验验证,相比于传统神经网络故障诊断方法,该模型以原始时域数据为对象进行故障诊断的准确率能达到96.7%,且无需复杂的网络结构,具有一定现实意义。

For the fault diagnosis of rolling bearings,accurate extraction of the fault features is a key step to improve the accuracy of diagnosis.With the original time-domain vibration signals,a novel model based on a convolutional neural network(CNN)and support vector machine(SVM)is proposed for fault diagnosis of rolling bearings.First,the original time-domain vibration signals are input into a two-dimensional convolutional neural network for adaptive feature extraction.Then the results of the fully connected layer are taken as the input of the support vector machine,and the results of the fully connected layer are classified by the support vector machine.Experimental verification shows that compared with the traditional neural network fault diagnosis method,the accuracy of this model for fault diagnosis based on the original time-domain data can reach 96.7%without the need for a complex network structure.It has a certain practical significance for fault diagnosis.