Development, validation, and verification of multi-pass thermo-mechanical welding simulations using the open-source MOOSE framework: NeT TG4 benchmark weldment

This study develops and validates a sequentially coupled thermo-mechanical welding simulation for the three-pass 316L stainless steel NeT TG4 benchmark weldment using the open-source Multiphysics Object-Oriented Simulation Environment (MOOSE) and the Nuclear Engineering Material model Library (NEML). A diffused ellipsoidal heat source was calibrated against thermocouple data and weld macrographs to accurately model the fusion zone geometry and transient thermal fields. Material hardening is represented using the Lemaitre-Chaboche mixed isotropic-kinematic model, while four annealing models—no annealing, single-stage at 1050°C and 1300°C, and two-stage at 800°C/1300°C — were implemented to assess their impact on welding-induced plasticity, distortions, and residual stresses. The simulation predictions were validated against experimental measurements and benchmarked against results from commercial software, demonstrating that MOOSE achieves comparable accuracy with enhanced computational efficiency. This work highlights the potential of open-source platforms like MOOSE for advanced welding analysis. All scripts necessary to reproduce the results are provided to support further development by the open-source community. • Developed MOOSE-based simulations for NeT TG4 multi-pass 316L weld benchmark. • Validated predictions of WIP, WID, and WIRS with experimental measurements. • Introduced single-stage annealing model balancing accuracy and efficiency. • Open-source MOOSE framework matches commercial tools in accuracy. • Provided all codes to promote reproducibility and open science in welding simulations.