Multi-scale defect detection and adaptive mitigation for defect-free autonomous laser-directed energy deposition​

Existing in-situ monitoring strategies for laser-based additive manufacturing are mostly qualitative (e.g., process regime classification), lack multi-scale integration, and focus primarily on defect detection instead of predictive forecasting and defect mitigation. The objective of this research is to develop multi-scale, quantitative defect detection, spatio-temporal forecasting, and adaptive control capabilities to enhance the process stability and part quality in coaxial wire laser-directed energy deposition (WL-DED). By integrating hierarchical multi-scale, multi-sensor monitoring with predictive analytics and adaptive process adjustments, the proposed approach will significantly reduce WL-DED production costs and material waste, paving the way towards sustainable, defect-free, autonomous additive manufacturing.