Physics-driven Digital Twin Based Architecture of Wire Arc Additive Manufacturing Enabled by Fast Surrogate Calculations

This study presents physics-driven digital twin architecture for wire arc additive manufacturing, designed to integrate high-fidelity physical simulation with data-driven surrogate models for real-time-capable process monitoring and decision support. The proposed multilayer framework defines the interaction between physical, digital and control domains through standardized data exchange and scalable computational workflows. Core components, including finite-element thermal modeling, surrogate training and middleware integration, are developed and quantitatively validated to demonstrate their feasibility within the intended control loop. The obtained results confirm the technical readiness and efficiency of the proposed architecture as a foundation for next-generation intelligent manufacturing systems.