Abstract
This study focuses on the development of a Digital Twin system for predicting thermal-elastic-plastic fields during gas heating processes. In large-scale structures, assembly welding often induces out-of-plane deformations, requiring skilled workers to apply corrective heating. Considering the decreasing number of skilled workers, it is crucial to automate these processes. A Digital Twin system that integrates FEM analysis with temperature and displacement measurements is proposed to accurately predict and optimize gas heating conditions. This system estimates unknown parameters such as heat distribution and temperature-dependent yield stress, achieving high accuracy in replicating actual phenomena. Using SS400 material, we conducted gas heating experiments to validate our system's effectiveness. Results showed that our system accurately reproduced experimental temperature and displacement fields. Moreover, the system identified optimal heating conditions that minimized deformation to within ±1 mm, demonstrating its potential for practical applications in the fabrication of complex structures. This Digital Twin approach represents a significant advancement in automating the gas heating process, reducing dependency on skilled workers, and contributing to precise deformation control.
