Scholarlink

In rotating machinery, rolling element bearings (REBs), are crucial mechanical components used for supporting load due to their ability to work in a wide range of speed and loads. Nevertheless, they frequently serve as the primary cause of malfunctions in these machines. Timely identification or assessment of these defects can avert malfunctions or breakdowns during operation. The interpretation and analysis of faults have been greatly influenced by recent breakthroughs in signal processing. Additionally, incorporation of deep learning techniques enhances the predicting accuracy of machine failures. In this study, adaptive superlet transform (ASLT) is applied to the response of bearing system and the resulting image from the transform is then used as the input for a proposed 2-D convolutional neural network (2-D CNN). An accuracy of 99.7% is achieved by using the proposed algorithm under the given operating condition.

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

Ball Bearing Fault Detection Using Adaptive Superlet Transform and 2-D CNN

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Samrat Mandal,
  • Pankaj Bhardwaj,
  • Nirmal Baran Hui,
  • Chintamani Mishra

摘要

In rotating machinery, rolling element bearings (REBs), are crucial mechanical components used for supporting load due to their ability to work in a wide range of speed and loads. Nevertheless, they frequently serve as the primary cause of malfunctions in these machines. Timely identification or assessment of these defects can avert malfunctions or breakdowns during operation. The interpretation and analysis of faults have been greatly influenced by recent breakthroughs in signal processing. Additionally, incorporation of deep learning techniques enhances the predicting accuracy of machine failures. In this study, adaptive superlet transform (ASLT) is applied to the response of bearing system and the resulting image from the transform is then used as the input for a proposed 2-D convolutional neural network (2-D CNN). An accuracy of 99.7% is achieved by using the proposed algorithm under the given operating condition.