Scholarlink

Drilling rig winch plays a vital role in oil and gas production. Due to long-time operation and complex working environment, the winch may encounter various faults, and its fault analysis and intelligent diagnosis are very important to ensure safety in production. Taking the JC70DB winch as the research object, this paper first introduces the function, working principle and fault causes of the JC70DB winch, then analyzes the working conditions and operating characteristics of the winch, develops the winch multivariate signal recognition technology, and innovates to extract time domain and frequency domain features for winch faults excited by multiple sources, and construct fault sensitive feature sets combined with self-attention mechanism. An early warning model based on the combination of dynamic theme model and student t distribution is established, which improves the accuracy of fault early warning.

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

Research on Fault Analysis and Intelligent Diagnosis of Drilling Rig Winch

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Deng Jia,
  • Xiao-guang Yang,
  • Zhi-xiong Zhou,
  • Xue-li Luo,
  • Yi Zhang,
  • Yang Li,
  • Wei-dong Zuo,
  • Chang-liang Wu,
  • Ling Jin,
  • Na Zhang,
  • Yong Su

摘要

Drilling rig winch plays a vital role in oil and gas production. Due to long-time operation and complex working environment, the winch may encounter various faults, and its fault analysis and intelligent diagnosis are very important to ensure safety in production. Taking the JC70DB winch as the research object, this paper first introduces the function, working principle and fault causes of the JC70DB winch, then analyzes the working conditions and operating characteristics of the winch, develops the winch multivariate signal recognition technology, and innovates to extract time domain and frequency domain features for winch faults excited by multiple sources, and construct fault sensitive feature sets combined with self-attention mechanism. An early warning model based on the combination of dynamic theme model and student t distribution is established, which improves the accuracy of fault early warning.