Application of entropy analysis in the prediction of flow distribution in parallel channels

Abstract: Multiphase flow in parallel channels is often an efficient approach to manage heat and energy distribution in engineering systems. However, two-phase flow with heating in parallel channels is prone to maldistribution, resulting in sub-optimal performance and in some cases, permanent damage. This challenge requires accurate flow modeling in parallel channels to mitigate or design against the adverse effect of two-phase flow maldistribution. The nonlinear nature of multiphase flow results in a multiplicity of predicted solutions for the same condition, thereby creating significant challenges in modeling flow distribution. Therefore, this study focuses on solving this challenge by applying entropy generation analysis and the conservation of mass, momentum balance, and energy balance to predict two-phase flow distribution in a two-parallel-channel assembly with a numerical model. Both model predictions and experimental data show that equally distributed flow becomes severely maldistributed with a decrease in flow rate, resulting in significant change (>30%) in the entropy generation rate. We show that the entropy analysis can be applied in distinguishing between stable and unstable flow distribution, like the linear stability analysis used in previous studies. We also surpass the limit of applying linear stability analysis by using entropy analysis to identify the most feasible end state in a maldistribution process.