基于NSGA-Ⅱ的锂离子电池液冷板的优化设计

Optimal Design of Liquid Cooling Plate for Lithium-ion Battery Based on NSGA-Ⅱ

为了研究纯电动汽车锂离子电池的液体冷却方式,提出了一种仿生叶状通道的热管理系统(BTMS)来研究锂离子电池的冷却性能。以通道角度、通道宽度、质量流量为设计变量,综合考虑液冷板的目标函数(平均温度、温度标准差、压力损失)之间的比重,得到液冷板结构参数及质量流量的最优参数;通过搭建单体电池实验平台,得到单体电池在不同放电倍率下的单位体积产热量;采用拉丁超立方抽样(LHS)在设计空间中选取40个样本点;通过克里金(Kriging)代替模型建立设计变量与目标函数之间的关系;利用NSGA-Ⅱ算法对Kriging模型进行优化寻优;利用CFD验证优化结果的准确度。研究结果表明:优化设计结果能将液冷板平均温度下降1.42 K(3.88%),温度标准差下降0.35 K(15.29%),压力损失下降7.49 Pa(8.78%)。

In order to study a liquid cooling method for lithium-ion batteries in pure electric vehicles,a bionic leaf-shaped channel thermal management system(BTMS)was proposed to study the cooling performance of lithium-ion batteries.Taking the channel angle,channel width,and mass flow as design variables,the proportion among the objective functions of liquid cooling plate(average temperature,temperature standard deviation,pressure loss)was comprehensively considered,and the structural parameters of the liquid cooling plate and the optimal parameters of the mass flow were obtained.Firstly,by setting up a single battery experimental platform,the heat production per unit volume of the single battery under different discharge rates was obtained.Then Latin hypercube sampling(LHS)was used to select 40 sample points in the design space.The relationship between design variables and objective functions was established by replacing the model with Kriging.Finally,the NSGA-Ⅱalgorithm was used to optimize the Kriging model,and CFD was used to verify the accuracy of optimization results.The research results show that the optimized design results can reduce the average temperature of the liquid cooling plate by 1.42 K(3.88%),the temperature standard deviation by 0.35 K(15.29%),and the pressure loss by 7.49 Pa(8.78%).