Application of Hybrid Monte Carlo Algorithm in Heat Transfer

Abstract

This article presents a new method of estimation of thermophysical parameters using the hybrid Monte Carlo (HMC) algorithm that synergistically combines the advantages of a Markov chain Monte Carlo (MCMC) method and molecular dynamics. The advantages of this technique over the conventional MCMC are elucidated by considering the multiparameter estimation in heat transfer. Four situations were analyzed. The first two involve a two- and a three-parameters estimation in a lumped capacitance model, third involves estimation in a distributed system, and the fourth involves estimation in a fin system. The goal is to establish the potency and usefulness of the HMC method for a wide class of engineering problems.