Machine learning prediction and multiobjective optimization for cooling enhancement of a plate battery using the chaotic water-microencapsulated PCM fluid flows

Abstract

BackgroundThe cooling of electric vehicle batteries is a crucial issue to increase the performance of the next-generation cars and develop market acceptance. There are various ideas for heat transfer enhancement of the batteries of an electric vehicle. However, it still requires more attention and research to raise the thermal management systems for the batteries. MethodsThe current study seeks the effects of simultaneous use of the chaotic flow and microencapsulate phase change materials for cooling a plate battery. Three values for Reynolds numbers from 100 to 300 and volume fractions from 0 to 10% in various geometries are evaluated. The results are presented through the variations of thermo-hydrodynamic parameters. A multiobjective optimizer combined with an artificial neural network finds optimum working conditions aiding some decision makers. Significant findingsThe Nusselt number of the suspension in the proposed geometries can increase up to 23% compared to that for the pure water in the straight channel. However, it may follow the friction factor strengthening by 160%. It is shown that the proposed geometries are mostly affected by the latter parameter. The efficiency of the proposed geometry accelerates from that of the conventional channel by increasing the Reynolds number and volume fraction.