Abstract:

Understanding the thermal physical properties and heat generation distribution patterns of lithium batteries is crucial for designing, modeling, and evaluating the performance of battery thermal management systems. The complex internal structure of lithium batteries leads to non-uniform heat generation. During charging and discharging processes, significant temperature gradients occur between the core and surface of cylindrical batteries, as well as along the axial direction. Although the current majority of simulations of battery thermal behavior assume a uniform volumetric heat source, this simplification inevitably introduces errors and compromises the accuracy of analytical and simulation models. Moreover, in thermoelectric cooling-based battery thermal management systems, the temperature differential across the batteries often approaches the thermal design limit more readily than the maximum temperature itself—further underscoring the importance of developing and applying non-uniform heat generation models for batteries. To address this issue, this study divided a single 18650 battery into five regions. Pulse discharge tests were conducted to obtain surface temperature data for each region and ambient temperature data during discharge at each 10% state of charge (SOC) interval. The heat generation rate of each battery region was calculated, and a non-uniform heat generation model of the single battery in both time and space was established using Fluent software. The results showed that the maximum error between the simulated data and experimental data at each measurement point was 1.28℃. Furthermore, a thermoelectric battery thermal management system model considering the non-uniform heat generation of lithium batteries was proposed. Simulation results indicated that, compared with model using uniform heat generation, the average errors of the maximum temperature and the maximum temperature difference of the proposed model were reduced by 69% and 43%, respectively. The model established in this work offers advantages such as short computation time, low complexity, and high accuracy, aiding in effectively predicting thermal-runaway risks and designing battery thermal management systems.

Key words: non-uniform heat generation, thermoelectric cooling, battery thermal management system, simulation

摘要：

厘清锂离子电池热物理特性及生热分布规律，对设计、建模和评估电池热管理系统的性能至关重要。锂离子电池复杂的内部结构导致其生热不均。在充放电过程中，圆柱形电池的芯体和表面之间以及沿轴向方向均会产生显著的温度梯度，而当前大多数电池热行为仿真中使用均匀体积热源，这一假设不可避免地会导致误差，其解析和仿真模型面临着精度挑战。此外，在基于热电制冷的电池热管理方案中，电池组温差往往更容易达到热设计上限，而非最高温度，更凸显出建立并运用电池非均匀生热模型的重要性。针对这一问题，本研究将单体18650电池划分为5个区域，通过脉冲放电实验获取每10% 荷电状态(State of Charge, SOC)下降期间各区域表面温度和环境温度数据，计算电池各区域生热率，利用Fluent软件建立单体电池在时间及空间上的非均匀生热模型，结果表明各测点与实验数据的最大误差为1.28℃。进一步提出了一种考虑锂离子电池非均匀生热的热电电池热管理系统模型，仿真结果显示，与采用均匀生热的模型相比，该模型最高温度和最大温差的平均误差分别降低了69%和43%。本研究所建立的模型具有计算时间短、复杂度低、精度高等优势，有助于有效预测热失控风险和设计热管理系统。

关键词: 非均匀生热, 热电制冷, 电池热管理系统, 仿真