Scholarlink

Root mean square (RMS) of vibration is one of the most widely used health indicators (HIs) for rotating machinery condition monitoring. However, the RMS is affected by the operation condition (e.g., rotating speed, torque). When the operation condition changes, the RMS usually changes accordingly, leading to possible false alarms. This paper develops a simple yet effective method to attenuate the effects of operation condition on RMS with the use of operation condition data. First, an optimal piecewise linear function is fitted between the RMS and operation condition data when the monitored machine is healthy. The fitted function estimates the effects of operation condition on RMS. For upcoming data stream (including RMS and operation condition data), the fitted function is used to provide the predicted RMS. The ratio of the real RMS and the predicted RMS is the indicator which is named normalized RMS (NRMS) for condition monitoring. The effectiveness of proposed NRMS is validated over a field-collected wind turbine dataset.

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

Normalized RMS for Rotating Machinery Condition Monitoring with the Use of Operation Condition Data

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Meng Rao,
  • Xingkai Yang,
  • Wangshan Xu,
  • Meili Zhang,
  • Yaqiang Jin,
  • Fulei Chu,
  • Mingjian Zuo

摘要

Root mean square (RMS) of vibration is one of the most widely used health indicators (HIs) for rotating machinery condition monitoring. However, the RMS is affected by the operation condition (e.g., rotating speed, torque). When the operation condition changes, the RMS usually changes accordingly, leading to possible false alarms. This paper develops a simple yet effective method to attenuate the effects of operation condition on RMS with the use of operation condition data. First, an optimal piecewise linear function is fitted between the RMS and operation condition data when the monitored machine is healthy. The fitted function estimates the effects of operation condition on RMS. For upcoming data stream (including RMS and operation condition data), the fitted function is used to provide the predicted RMS. The ratio of the real RMS and the predicted RMS is the indicator which is named normalized RMS (NRMS) for condition monitoring. The effectiveness of proposed NRMS is validated over a field-collected wind turbine dataset.