Multi-objective optimization of a microchannel membrane-based absorber with inclined grooves based on CFD and machine learning

A novel microchannel membrane-based absorber with inclined grooves is proposed and studied by a three-dimensional CFD model. Parametric analysis is carried out to analyze the effects of structural parameters on the absorption rate and pressure drop. Results indicate that the groove introduces a swirling effect in the solution channel, interrupting the boundary layer at the solution-membrane interface and increasing the solution residence time inside the microchannel. The absorption rate in the grooved channel is up to 1.55 times higher, while the pressure drop is 0.77–0.96 times lower. To optimize the novel absorber geometries and maximize the integrated performance, the Pareto front is obtained by performing a multi-objective optimization, in which a machine learning method based on ANN and NSGA-ΙΙ is developed. The optimal design parameters from the Pareto front are identified by two well-known decision-making methods, LINMAP and TOPSIS. Compared to the basic smooth channel, these methods generate 1.41 and 1.47 times improvement in volumetric cooling capacities, at a much lower solution pressure drop. Moreover, a high absorption rate equivalent to that of a 200 μm-thick smooth channel is achieved by LINMAP and TOPSIS, with pressure drops lower by 6.29 and 5.63 times, respectively.