Scholarlink

This article proposes a model for rolling bearing fault diagnosis based on variational mode decomposition combined with a convolutional neural network and a bidirectional long short-term memory network fusion (VMD-CNN-BiLSTM). In order to better extract the features of the original vibration signal, the Osprey Cauchy Sparrow search algorithm (OCSSA) was used to optimize the VMD parameters. Then, the optimal IMF component was extracted and calculated to construct the required feature vectors for the model. The feature vectors are used as inputs to train the CNN BiLSTM model. Finally, the model can accurately identify the fault status of rolling bearings. Experimental results have shown that the diagnostic model based on VMD-CNN-BiLSTM not only achieves a significant reduction in training time but also has an accuracy of 99.33%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Study on Fault Diagnosis of Rolling Bearings Based on CNN-BiLSTM

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xin Hu,
  • Yuanlin Luo,
  • Wenhui Liu,
  • Yuechao Wu

摘要

This article proposes a model for rolling bearing fault diagnosis based on variational mode decomposition combined with a convolutional neural network and a bidirectional long short-term memory network fusion (VMD-CNN-BiLSTM). In order to better extract the features of the original vibration signal, the Osprey Cauchy Sparrow search algorithm (OCSSA) was used to optimize the VMD parameters. Then, the optimal IMF component was extracted and calculated to construct the required feature vectors for the model. The feature vectors are used as inputs to train the CNN BiLSTM model. Finally, the model can accurately identify the fault status of rolling bearings. Experimental results have shown that the diagnostic model based on VMD-CNN-BiLSTM not only achieves a significant reduction in training time but also has an accuracy of 99.33%.