Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Recursive Heat Flux Estimation in Nonlinear Heat Conduction using Kalman Filter and Kirchhoff Transform
Submission Author:
César Cunnha Pacheco , RJ , Brazil
Co-Authors:
Arthur Anastácio, César Cunnha Pacheco
Presenter: Arthur Anastácio
doi://10.26678/ABCM.COBEM2023.COB2023-1071
Abstract
This paper deals with the problem of estimating high-magnitude heat fluxes applied at the surface of a flat plate. The inverse problem is solved by assuming that transient temperature measurements are available at the opposite side of the plate. Previous works have shown that solution methodologies based on Kalman filtering and reduced-order modelling are capable of yielding estimates in good agreement with reference values. However, since this reduced-order model assumes constant thermal properties, when high temperatures are achieved, the onset of physical nonlinearities lead to inaccurate results. To address this issue, the Kalman Filter is rewritten, now considering the heat conduction model in terms of the Kirchhoff transform, which addresses partially these nonlinearities, alleviating this modelling errors up to a certain degree. The inverse problem was proposed in this approach as a state variable estimation problem, typical of dynamical systems such as the mathematical model of the direct problem. The solution of the direct problem was written in the form of an evolution-observation model (EOM), in its linear form, according to the equations resulting from the Finite Volume (FVM) analysis. The proposed approach requires only the assumption of constant thermal diffusivity, allowing for some variation in the thermal conductivity and thermal capacity.
Keywords
Inverse problems, State estimation, heat conduction
