Fast Reconstruction of Transient Heat-Flux Distributions in a Laser Heating Process with Time-Space Adaptive Mesh Refinement

The study of inverse heat transfer problems (IHTPs) have obtained extensive attention over the recent decades. However, IHTPs are often mathematically ill-posed, namely the solution stability suffers from measurement errors. Although there already exist established methods for IHTPs, their applications are mostly limited to simple geometric configurations in one- or multi-dimensions due to the high complexity. It is still a crucial task to develop special solution techniques that can ensure fast and accurate estimation. Recently, we developed a forward modeling approach and a time-space adaptive mesh refinement (TSAMR) strategy. In this work, we consider a realistic simulation study and use these methods to reconstruct laser heating boundary condition on the front surface of a 3D object subjected to laser beam heating and a combined radiation. By conducting groups of parameters tests, the solution quality is expected to be improved. The estimated heat-flux distributions and temperature distributions are presented to validate the proposed methods. Based on these results, a very large number of tests are supposed to be performed in the supercomputing plat-form in future work to obtain optimal parameters for the TSAMR strategy, so as to solve complicated IHTPs in a more efficient and accurate manner.