Optimized design of novel serpentine channel liquid cooling plate structure for lithium-ion battery based on discrete continuous variables

The thermal management system of power batteries plays a crucial role in regulating the temperature of the battery pack and ensuring efficient operation. By optimizing the design of the thermal management system structure, cooling performance can be significantly enhanced. In this study, a novel structure called novel serpentine cooling plate (NSCP) is designed to improve the cooling performance of the liquid-cooled plate. This study utilizes numerical simulation to investigate the impact of coolant mass flow rate, coolant inlet temperature, coolant flow direction, and the number of flow channels on the performance of the liquid cooling plate under a 2C discharge rate, using a 50 % volume concentration of ethylene glycol solution as the coolant. The results indicate that the cooling performance of the NSCP surpasses that of the traditional SCP. Furthermore, to enhance the cooling performance of the NSCP, we optimize the channel width (A), channel depth (B), and wall thickness (C) utilizing both discrete and continuous variables. Initially, structural parameters A, B, and C are considered as discrete random variables, and significant factors along with optimal combinations are identified through orthogonal test polar analysis and analysis of variance. Subsequently, these parameters are treated as continuous random variables, and sampling is conducted using the optimal Latin hypercube sampling method. The response surface method is employed to establish the functional relationship between design variables and performance indices, while NSGA-II is utilized for optimization, leading to the derivation of Pareto-optimal solutions using the CRITIC weighting method. A comparison of the two optimization methods reveals that the continuous variable optimization is more effective; thus, this method is selected for implementation. The results demonstrate that the BTMS optimized with continuous variables can reduce the maximum temperature, maximum temperature difference, and average temperature of the battery pack by 2.1 degrees C, 1.2 degrees C, and 2.8 degrees C, respectively, compared to the initial structure. Finally, validation under different discharge rates and environmental temperatures demonstrates the robustness of the optimized structure under various operating conditions. These research findings provide effective theoretical basis and practical references for the design and optimization of liquid-cooled lithium-ion battery cooling structures.