Scholarlink

<p>A systematic review is conducted from three perspectives: defect mechanisms, process monitoring, and defect prediction. First, the formation and influence mechanisms of three representative defect types, including geometric deviation, surface defect, and fracture failure, are analyzed emphatically. Then, online monitoring technologies of key physical quantities such as force, temperature, and strain are sorted out. Furthermore, the applications of physics-based models, <i>machine learning</i> models, hybrid models, and transfer learning strategies in defect prediction for incremental sheet forming are explored in detail. The research results demonstrate that a coupled framework integrating hybrid modeling and transfer learning effectively balances physical interpretability and data-driven generalization capability, thereby enhancing both the accuracy and robustness of defect prediction models. Finally, an integrated incremental sheet forming intelligent system framework encompassing monitoring, prediction, and control is proposed, providing a theoretical foundation and technical pathway for achieving high quality, high efficiency, and high stability flexible forming of metal sheets.</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

A review on monitoring and defect prediction of incremental sheet forming process

  • 期刊论文

  • 发表日期：

    2026年7月18日

查看全文
  • Jianing Song,
  • Lin Wang,
  • Fuzhen Han,
  • Shaoqi Song,
  • Tingyu Ge,
  • Heng Chen,
  • Chenglong Yang,
  • Yanle Li

摘要

A systematic review is conducted from three perspectives: defect mechanisms, process monitoring, and defect prediction. First, the formation and influence mechanisms of three representative defect types, including geometric deviation, surface defect, and fracture failure, are analyzed emphatically. Then, online monitoring technologies of key physical quantities such as force, temperature, and strain are sorted out. Furthermore, the applications of physics-based models, machine learning models, hybrid models, and transfer learning strategies in defect prediction for incremental sheet forming are explored in detail. The research results demonstrate that a coupled framework integrating hybrid modeling and transfer learning effectively balances physical interpretability and data-driven generalization capability, thereby enhancing both the accuracy and robustness of defect prediction models. Finally, an integrated incremental sheet forming intelligent system framework encompassing monitoring, prediction, and control is proposed, providing a theoretical foundation and technical pathway for achieving high quality, high efficiency, and high stability flexible forming of metal sheets.