CFD and ANN-based optimization of perforated baffle design in an immersion-cooled BTMS for 21700 lithium-ion batteries

Immersion battery cooling systems have attracted considerable attention for application in battery thermal management system (BTMS) for electric vehicles because of their superior cooling performance. In this study, numerical simulations were conducted by considering baffle position, the distance between the Lithium-ion battery and the perforation ( α ), and installation angle ( θ ) as design variables to identify the optimal conditions for minimizing pumping power, maximum temperature ( T max ), and temperature standard deviation ( TSD ). The flow rate through the perforations, which varied with baffle position and θ , had a significant influence on the pressure drop and temperature distribution within the BTMS. The minimum pumping power of 6.547 mW was obtained under the Left position at θ = 3° and α = 1.75 mm. In addition, the minimum T max of 31.466 °C was achieved under the Right position at θ = 5° and α = 1.25 mm, while the minimum TSD of 0.3103 °C was obtained under the Cross position at θ = 3° and α = 1.25 mm. Furthermore, an ANN prediction model was developed using the CFD results, and the mean relative error was less than 1.5%, demonstrating high predictive accuracy. These results indicate that the ANN model can serve as an efficient surrogate model for BTMS design optimization and can significantly reduce the time and computational cost associated with repeated CFD simulations.