Estimação de parâmetros e seleção de classes de modelos em problemas de convecção-difusão em microcanais

Abstract

The analysis of heat transfer at the microscale has been widely employed in recent decades due to technological advancements in the industry of miniaturized fluid transport systems. In this work, inverse problems of parameter estimation and model class selection in convection-diffusion problems in microchannels are formulated and solved using a Bayesian approach. The first physical system of interest involves cooling a heat source simulating a processor through a thermal dissipation system with microchannels. In the second problem to be analyzed, a microchannel is modeled, taking into account slip flow and temperature jump at the walls. The results, using simulated temperature measurements, show that the method employed is capable of qualitatively and quantitatively identifying the most plausible model for different physical problems in the context of heat transfer in microchannels.