Towards Inverse Estimation of Properties, Process Parameters and Residual Effects for Friction Stir Welding

Forward and inverse modeling of the friction stir welding (FSW) process is an important endeavor that can be used to optimize process parameters that play a significant role on achieving manufactured part specification requirements. One such requirement is the minimization of permanent deformation induced by the plastically deformed zone close and inside the stirred zone. However, at the same time modeling FSW requires the estimation of process parameters that are not directly measurable. In this work we present a computational framework for applying an inverse approach for the heat conduction equation with experimentally guided volume heat source term that can be used to identify such parameters and subsequently use them for solving the associated forward coupled thermo-plastic problem in a Lagrangian frame of reference. We demonstrate the versatility of our framework by determining the feasibility of determining plastic (and hence residual) deformation.