An efficient computational method for anisotropic thermal conductivity estimation

This article presents an attractive and straightforward computational strategy for estimating the anisotropic thermal conductivity in a wide range of materials. It results in a reliable and efficient approach with many potential applications. The proposed method is based on the mathematical model solution of a thermal process to generate some synthetic measurements simulating sensors located at the center of each face of a body under study. This work implements three optimization techniques for solving the formulated inverse thermal problem: Levenberg-Marquardt Algorithm, Particle Swarm Optimization, and Symbiotic Organism Search. Plus, we use an anisotropic cubic piece of solid material as a demonstrative case. Results show an excellent agreement between the estimated anisotropic thermal conductivities and the proposed solutions for the model. Furthermore, we notice a strong impact of the noise level on the measurement system, which affects the precision of the estimated conductivities.