Scholarlink

This paper investigates the application of multi-target machine learning to enhance non-destructive testing (NDT) of high-pressure natural gas storage tanks using pressure sensor data. The paper addresses the need for improved safety and reliability by simultaneously predicting pressure monitoring status, leak detection, and fire suppression needs. Various machine learning models were evaluated, and results demonstrate the potential of multi-target machine learning to provide a comprehensive assessment of tank integrity for more proactive and efficient NDT strategies in the oil and gas industry. The XGBoost model achieved the highest overall accuracy of 95% in predicting pressure status, leak detection, and fire suppression needs.

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

Non-Destructive Oil and Gas Tanks Testing by Using Pressure Sensors Based on Multi-Target Machine Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mohamed S. Elshebani,
  • Salwa M. Ibrahim

摘要

This paper investigates the application of multi-target machine learning to enhance non-destructive testing (NDT) of high-pressure natural gas storage tanks using pressure sensor data. The paper addresses the need for improved safety and reliability by simultaneously predicting pressure monitoring status, leak detection, and fire suppression needs. Various machine learning models were evaluated, and results demonstrate the potential of multi-target machine learning to provide a comprehensive assessment of tank integrity for more proactive and efficient NDT strategies in the oil and gas industry. The XGBoost model achieved the highest overall accuracy of 95% in predicting pressure status, leak detection, and fire suppression needs.