Scholarlink

<p>The paper deals with fault detection in industrial processes using a novel combination of adaptive principal component analysis with artificial intelligence (AI)-based optimized wavelet denoising parameters. To solve the problem, meta-heuristic optimizer known as particle swarm optimization (PSO) is used to further formulate the selection of parameters for the proposed technique. Two important factors that contribute to the efficacy of fault detection, i.e., the false alarm rate and the missed detection rate, have been addressed in the design. The fault detection system is designed so that robustness against noise and sensitivity to faults are enhanced. The efficacy of the fault detection system has been demonstrated by applying it to a rotating machinery test rig system which was widely used in industrial pump system. The results have shown that the proposed technique can reduce the false alarm rate and the missed detection rate to less than 1%, lower compared to the techniques mentioned in the literature, such as classical principal component analysis and adaptive principal component analysis.</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

AI-Based Resilient Fault Detection Approach for Industrial Process Control Using Meta-Heuristic Parameter Optimization of Wavelet Denoising

  • 期刊论文

  • 发表日期：

    2025年8月17日

查看全文
  • Faizan E. Mustafa,
  • Abdul Qayyum Khan,
  • Abdus Samee,
  • Ghulam Mustafa,
  • Muhammad Abid,
  • Zahid Riaz

摘要

The paper deals with fault detection in industrial processes using a novel combination of adaptive principal component analysis with artificial intelligence (AI)-based optimized wavelet denoising parameters. To solve the problem, meta-heuristic optimizer known as particle swarm optimization (PSO) is used to further formulate the selection of parameters for the proposed technique. Two important factors that contribute to the efficacy of fault detection, i.e., the false alarm rate and the missed detection rate, have been addressed in the design. The fault detection system is designed so that robustness against noise and sensitivity to faults are enhanced. The efficacy of the fault detection system has been demonstrated by applying it to a rotating machinery test rig system which was widely used in industrial pump system. The results have shown that the proposed technique can reduce the false alarm rate and the missed detection rate to less than 1%, lower compared to the techniques mentioned in the literature, such as classical principal component analysis and adaptive principal component analysis.