Scholarlink

In recent years, Prediction and Health Management (PHM) technologies have become one of the important indicators various while monitoring the conditions of health bearings and predicting fault of different equipments. In this study, Multi-objective Moth Swarm Algorithm (MOMSA) utilizes the FEMTO dataset for diagnosing the different faults for bearing fault and introduced an new methodology named as the Improved Grey Wolf Optimization (IGWO) where the Remaining Useful Life (RUL) of machinery components have been identified. In this proposed methodology, a diagnostic accuracy of 97.2%, precision of 96.5%, and recall of 95.8% has been achieved while detecting the faults with an 15% improvement in computational efficiency when compared to traditional methods. While predicting RUL, this proposed IGWO model gives the value of MAPE of 4.2% which outperforms the existing models by an average of 10%. For these results, the efficacy of MOMSA has been evaluated by considering the values of specific entropy values and IGWO's precision in RUL prediction.

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

Detection of Faults by Optimization Driven Methodology: A Comprehensive Study on the Heath of Bearings

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Subhranil Das,
  • Rashmi Kumari,
  • Raghwendra Kishore Singh

摘要

In recent years, Prediction and Health Management (PHM) technologies have become one of the important indicators various while monitoring the conditions of health bearings and predicting fault of different equipments. In this study, Multi-objective Moth Swarm Algorithm (MOMSA) utilizes the FEMTO dataset for diagnosing the different faults for bearing fault and introduced an new methodology named as the Improved Grey Wolf Optimization (IGWO) where the Remaining Useful Life (RUL) of machinery components have been identified. In this proposed methodology, a diagnostic accuracy of 97.2%, precision of 96.5%, and recall of 95.8% has been achieved while detecting the faults with an 15% improvement in computational efficiency when compared to traditional methods. While predicting RUL, this proposed IGWO model gives the value of MAPE of 4.2% which outperforms the existing models by an average of 10%. For these results, the efficacy of MOMSA has been evaluated by considering the values of specific entropy values and IGWO's precision in RUL prediction.