Scholarlink

The rotor-bearing system plays a crucial role in heavy industrial applications like sugar crane crushers and cement mills for power transmission. This study investigates the dynamic characteristics of a misaligned rotor-bearing system through experimental analysis and the utilization of a trained artificial neural network (ANN) algorithm. Experimentation encompasses varying levels of critical factors including misalignment angle, rotor RPM, and static load to comprehensively estimate their influence on vibration patterns. ANN technique has been employed for angular misalignment fault detection within the rotor-bearing system. Notably, the ANN demonstrates exceptional accuracy in predicting rotor angular misalignment, achieving over 99% accuracy in training and 95% accuracy in testing with five nodes in hidden layer of neural model. In the context of Industry 4.0, this research holds significant potential as a key component in machinery condition monitoring, offering precise and reliable fault diagnosis.

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

Fault Diagnosis in Rotor-Bearing Systems Using ANN-Based Vibration Analysis for Industry 4.0 Machinery Condition Monitoring

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ganesh L. Suryawanshi,
  • Sachin K. Patil,
  • Ramchandra. G. Desavale

摘要

The rotor-bearing system plays a crucial role in heavy industrial applications like sugar crane crushers and cement mills for power transmission. This study investigates the dynamic characteristics of a misaligned rotor-bearing system through experimental analysis and the utilization of a trained artificial neural network (ANN) algorithm. Experimentation encompasses varying levels of critical factors including misalignment angle, rotor RPM, and static load to comprehensively estimate their influence on vibration patterns. ANN technique has been employed for angular misalignment fault detection within the rotor-bearing system. Notably, the ANN demonstrates exceptional accuracy in predicting rotor angular misalignment, achieving over 99% accuracy in training and 95% accuracy in testing with five nodes in hidden layer of neural model. In the context of Industry 4.0, this research holds significant potential as a key component in machinery condition monitoring, offering precise and reliable fault diagnosis.