ABSTRACT
The standard heat conduction problem refers to the process of solving the internal temperature distribution and changing process of a given object's geometric shape, thermophysical parameters, initial conditions, and boundary conditions. The inversion of surface heat flux is one of the most typical inverse heat conduction problems, which has a broad application in aerospace, nuclear physics, metallurgy, and other industrial research fields. The load inversion method developed in recent years has become an effective means to obtain the heat flow load of the reentry vehicle by arranging temperature sensors inside the reentry vehicle and measuring its inner wall temperature and then solving the inverse heat transfer problem to obtain the outer wall heat load. The vigorous development of high-performance scientific computing has promoted the application of deep learning technology in computational physics. The temperature at the sampling time is affected by the heat flux at different times and locations during the whole heat transfer process of the experiment.
