A novel parametric spectral technique to solve inverse heat problems for the estimation of wall thermophysical properties
Résumé
This study proposes reducing the computational cost of gradient-based method for solving inverse heat conduction problem of the thermal diffusivity estimation. It relies on using a spectral parametric reduced order model to build a solution dependent on spatial and temporal coordinates, as well as on an extra-parameter of the problem, such as the thermal diffusivity. This parametric solution of the direct problem and its sensitivity function is computed off-line significantly lowering the computational cost of the iterative inverse problem algorithm. Synthetic data are used to verify the reliability of the method. Results show its potential for building diagnostics and energy efficiency, offering a reliable tool for characterizing building wall thermal properties.
Domaines
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |