Scholarlink

<p>The adoption of smart, data-driven technologies is reshaping manufacturing, yet Wire Arc Additive Manufacturing (WAAM) lacks an integrated cyber-physical architecture capable of supporting qualification and certification workflows. Existing AI-based solutions typically operate as isolated analytics, limiting their ability to transform heterogeneous sensor data into traceable information that can guarantee product compliance. This work introduces a unified, qualification-oriented cyber-physical framework for WAAM, in which sensing, modelling, optimisation, monitoring and control are treated as interdependent components of a single product-centric workflow. The framework exploits Artificial Intelligence (AI) to link offline qualification, process planning and sustainability-driven optimisation with online, multimodal monitoring and adaptive control. A case study on Invar 36 material demonstrates how the proposed architecture enables parameter optimisation, layer-level quality estimation and closed-loop corrective actions. Although validated on WAAM, the modular and process-agnostic design provides a generalisable pathway toward intelligent and certifiable additive manufacturing.</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

Data-driven smart wire arc additive manufacturing: a qualification-oriented cyber-physical system framework

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月27日

查看全文
  • Giulio Mattera,
  • Alessandra Caggiano,
  • Andrew Kusiak

摘要

The adoption of smart, data-driven technologies is reshaping manufacturing, yet Wire Arc Additive Manufacturing (WAAM) lacks an integrated cyber-physical architecture capable of supporting qualification and certification workflows. Existing AI-based solutions typically operate as isolated analytics, limiting their ability to transform heterogeneous sensor data into traceable information that can guarantee product compliance. This work introduces a unified, qualification-oriented cyber-physical framework for WAAM, in which sensing, modelling, optimisation, monitoring and control are treated as interdependent components of a single product-centric workflow. The framework exploits Artificial Intelligence (AI) to link offline qualification, process planning and sustainability-driven optimisation with online, multimodal monitoring and adaptive control. A case study on Invar 36 material demonstrates how the proposed architecture enables parameter optimisation, layer-level quality estimation and closed-loop corrective actions. Although validated on WAAM, the modular and process-agnostic design provides a generalisable pathway toward intelligent and certifiable additive manufacturing.