3D-transient identification of surface heat sources through infrared thermography measurements on the rear face

A study of estimation of space and time distribution of heat flux at the front face of a thin plate is presented in this paper. This identification is carried out from the temperature data collected on the rear face by an infrared camera. The inversion procedure is based on a sequential estimation method: the state representation of the 3D parabolic model of heat conduction in solids. Of course, the parameterized heat flux distribution is discrete in both space (2D) and time. The high number of unknowns to be estimated simultaneously, as well as the measurement noise, makes the ill-posed character of this multi-dimensional inverse problem quite high. Regularization is implemented by using two techniques: the stabilization by the function specification method and the Tikhonov penalization. •An inverse problem of reconstituting heat fluxes on a plate from temperature measurements by an infrared camera on the opposite face is performed.•The inversion procedure is based on a sequential estimation method using the state representation “pseudo-inversion” of the 3D parabolic model of heat conduction in solids.•Regularization is implemented by using two techniques: the stabilization by the function specification method and the Tikhonov penalization.•It is possible to reconstruct the spatiotemporal distribution of heat flux densities, for which we have no prior information.