Scholarlink

<p>Machine learning (ML) is reshaping additive manufacturing (AM) with its potent capability of data analysis, antonomous learning and intelligent decision-making. ML empowers designers by enhancing the capabilities of structural design, improving process optimization, and elevating product performance, thus further help to propose designs that can adapt to diversified manufacturing conditions and meet multi-functional requirements. We herein comprehensively review the cutting-edge advances of ML-based AM in various domains. ML technologies and methods used in multiple AM domains are summarized and the technical features are introduced. ML-based materials preparation, structure design, performance prediction and optimization of AM are comprehesively compared and discussed. Lastly the encountered challenges and the future developments are demonstrated. With an in-depth analysis, we hope this review can propel the applications of ML in the intelligence-led design of AM.</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

Machine learning in additive manufacturing: enhancing design, manufacturing and performance prediction intelligence

  • 期刊论文

  • 发表日期：

    2025年1月27日

查看全文
  • Teng Wang,
  • Yanfeng Li,
  • Taoyong Li,
  • Bei Liu,
  • Xiaowei Li,
  • Xiangyu Zhang

摘要

Machine learning (ML) is reshaping additive manufacturing (AM) with its potent capability of data analysis, antonomous learning and intelligent decision-making. ML empowers designers by enhancing the capabilities of structural design, improving process optimization, and elevating product performance, thus further help to propose designs that can adapt to diversified manufacturing conditions and meet multi-functional requirements. We herein comprehensively review the cutting-edge advances of ML-based AM in various domains. ML technologies and methods used in multiple AM domains are summarized and the technical features are introduced. ML-based materials preparation, structure design, performance prediction and optimization of AM are comprehesively compared and discussed. Lastly the encountered challenges and the future developments are demonstrated. With an in-depth analysis, we hope this review can propel the applications of ML in the intelligence-led design of AM.