基于改进一维卷积神经网络的滚动轴承故障诊断分析

Research on Fault Diagnosis of Rolling Bearing Based on Improved One-dimensional Convolutional Neural Network

针对滚动轴承故障特征提取困难,导致故障诊断模型训练次数比较多且诊断准确率不够高的问题,提出基于改进樽海鞘群算法(ISSA)优化CBAM-1DCNN结构参数的滚动轴承故障诊断模型。将一维卷积神经网络(1DCNN)与注意力机制(CBAM)相结合,且在1DCNN中添加BN层,利用改进樽海鞘群算法对CBAM-1DCNN网络结构参数进行优化。结果表明:经改进樽海鞘群算法优化的CBAM-1DCNN滚动轴承故障诊断模型在较少的训练次数下达到最好的拟合效果和更高的故障识别精度,且具有良好的泛化能力。

Aiming at the difficulty of extracting fault features of rolling bearings,which leads to prolonged training of fault diagnosis models and insufficient diagnosis accuracy,a rolling bearing fault diagnosis model based on the improved salp swarm algorithm(ISSA)to optimize the structural parameters of CBAM-IDCNN is proposed.The fault diagnosis model combines a one-dimensional convolutional neural network(IDCNN)with a convolutional block attention module(CBAM),and the batch normalization(BN)layer is added to the 1DCNN network.The improved salp swarm algorithm is used to optimize the CBAM-1DCNN structural parameters.The results show that the CBAM-IDCNN rolling bearing fault diagnosis model optimized by the improved salp swarm algorithm can effectively achieve the best fitting performance and higher fault recognition accuracy with less training,and has good generalization ability.