Scholarlink

<p>To address the challenge of real-time sensing and accurate prediction of the wear status of marine water-lubricated stern bearings under complex operating conditions, a wear monitoring system based on multi-source sensor fusion was developed. The system utilizes eddy current displacement sensors and vibration acceleration sensors to construct a distributed hardware acquisition architecture. Based on LabVIEW and Python hybrid programming technology, a monitoring software system integrating multi-channel data acquisition, real-time display and control, anomaly warning, and trend analysis was developed. Furthermore, two machine learning algorithms, Long Short-Term Memory and Gated Recurrent Unit, were integrated to address the difficulty in characterizing nonlinear wear trends. Accelerated wear experiments were conducted using a large-scale simulation test bench, constructing a large-scale sample database with high-density data. Experimental results demonstrate that the system possesses good adaptability under complex operating conditions. The trend prediction accuracy for key parameters reached 85 percent, effectively realizing the perception of wear status and the analysis of wear evolution trends for water-lubricated stern bearings. These research results hold significant reference and application value for the operation and maintenance of water-lubricated stern bearings.</p>

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

Development of marine water-lubricated stern bearing wear monitoring device and system

  • 期刊论文

  • 发表日期：

    2026年7月2日

查看全文
  • Xiaxia Xiang,
  • Zhenyang Guo,
  • Yanmu Chen,
  • Yujie Li,
  • Yeming Lu,
  • Xiaofang Wang

摘要

To address the challenge of real-time sensing and accurate prediction of the wear status of marine water-lubricated stern bearings under complex operating conditions, a wear monitoring system based on multi-source sensor fusion was developed. The system utilizes eddy current displacement sensors and vibration acceleration sensors to construct a distributed hardware acquisition architecture. Based on LabVIEW and Python hybrid programming technology, a monitoring software system integrating multi-channel data acquisition, real-time display and control, anomaly warning, and trend analysis was developed. Furthermore, two machine learning algorithms, Long Short-Term Memory and Gated Recurrent Unit, were integrated to address the difficulty in characterizing nonlinear wear trends. Accelerated wear experiments were conducted using a large-scale simulation test bench, constructing a large-scale sample database with high-density data. Experimental results demonstrate that the system possesses good adaptability under complex operating conditions. The trend prediction accuracy for key parameters reached 85 percent, effectively realizing the perception of wear status and the analysis of wear evolution trends for water-lubricated stern bearings. These research results hold significant reference and application value for the operation and maintenance of water-lubricated stern bearings.