Scholarlink

Quality-related fault detection and root cause diagnosis are the key problems in fault diagnosis framework, which are the effective ways to ensure operational safety and product quality. In this chapter, a modified canonical variable analysis (MCVA) model is developed for quality-related fault detection. Subsequently, the classical generalized reconstruction-based contribution (GRBC) is used for fault identification. Then, a transfer entropy (TE)-based causality analysis method is put forward for root cause diagnosis of quality-related faults. In the end, sufficient simulation experiments are conducted by a typical industrial process, the hot rolling process (HRP), to demonstrate the superiority of the developed framework compared with some classical algorithms.

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

Data-Based Quality-Related Fault Diagnosis Scheme for Fault Detection and Root Cause Diagnosis

  • 章节

  • 出版模式：

    开放获取

  • 发表日期：

    2026年

查看全文
  • Kaixiang Peng,
  • Liang Ma

摘要

Quality-related fault detection and root cause diagnosis are the key problems in fault diagnosis framework, which are the effective ways to ensure operational safety and product quality. In this chapter, a modified canonical variable analysis (MCVA) model is developed for quality-related fault detection. Subsequently, the classical generalized reconstruction-based contribution (GRBC) is used for fault identification. Then, a transfer entropy (TE)-based causality analysis method is put forward for root cause diagnosis of quality-related faults. In the end, sufficient simulation experiments are conducted by a typical industrial process, the hot rolling process (HRP), to demonstrate the superiority of the developed framework compared with some classical algorithms.