ANN based estimation of heat generation from multiple protruding heat sources on a vertical plate under conjugate mixed convection

Abstract

Inverse Heat Transfer Problems (IHTP) are characterized by estimation of unknown quantities by utilizing any given information of the system. In this study, the inverse problem of estimation of heat generation in multiple two dimensional protruding heat sources on a vertical plate, a geometry frequently encountered in the cooling of electronic equipment, is carried out from the information available on the temperature distribution on the substrate on which these sources are mounted. A non-iterative method is applied utilizing Artificial Neural Networks (ANN) and covariance analysis to estimate the heat generation in the protruding heat sources on a vertical plate. The forward model involving laminar, two dimensional, steady, incompressible fluid flow and mixed convection heat transfer is numerically solved with FLUENT 6.3 for known values of heat generation in the protruding sources and the temperature distribution thus obtained on the PCB substrate is utilized to train the ANN for the inverse model. Parametric studies are conducted on the forward model to investigate the effect of Richardson number, Reynolds number, the chip and substrate conductivities on the heat dissipation to the fluid flowing over the heat sources. The trained networks are finally used to estimate the heat generation from the sources for a given temperature distribution on the substrate wall generated, for known values of the heat generation rates, which serve as the “measured” temperature distribution. Use is made of covariance analysis in order to identify the important temperature locations sufficient to carry out the inverse analysis. Finally, a systematic investigation on the effect of noise in the temperature “measurements” on the estimates also has been carried out.