热电偶
计算流体力学
瞬态(计算机编程)
热的
机械
对流
材料科学
传热
稳态(化学)
自然对流
代表(政治)
计算机模拟
产量(工程)
温度测量
热力学
强迫对流
近似误差
均方误差
强度(物理)
对流换热
计算机科学
校准
模拟
观测误差
核工程
机械工程
作者
Duccio Fedeli,Marco Lagnoni,Claudio Scarpelli,Francesco Giuseppe Quilici,Antonio Bertei,Giovanni Lutzemberger,Filippo Fruzza,Maria Vittoria Salvetti,Alessandro Mariotti
出处
期刊:Fluids
[Multidisciplinary Digital Publishing Institute]
日期:2026-08-18
卷期号:11 (8): 204-204
标识
DOI:10.3390/fluids11080204
摘要
A computational fluid dynamics and conjugate heat transfer (CFD+CHT) methodology is developed for the thermal simulation of a commercial prismatic LiFePO4 cell under charging and discharging operating conditions. The approach couples a three-dimensional representation of the battery, including a simplified description of its internal layered structure, with an electrochemical–thermal heat-generation model implemented as a temperature- and time-dependent volumetric source term. The heat source is applied within the active layers of the cell and updated during the transient simulation according to the local thermal state and to the evolution of the state of charge. The methodology is applied to 1C and 2C cycles under natural convection and forced-air cooling at free-stream velocities of 10ms−1 and 20ms−1. A dedicated wind-tunnel campaign is carried out on the same cell, instrumented with type-K thermocouples distributed over its external surfaces, to provide experimental data for model validation. The results show that the proposed framework accurately reproduces the main wall-temperature trends observed experimentally. Under natural convection, the temperature distribution remains nearly uniform, whereas forced convection produces more pronounced vertical and in-plane gradients. For the charge cycles, the comparison between CFD predictions and end-of-cycle measurements yields a mean absolute error (MAE) of 0.66∘C and a root-mean-square error (RMSE) of 0.82∘C over 168 measurement locations. The discharge cycles yield a comparable level of agreement (MAE 0.65∘C, RMSE 0.81∘C over 168 probe points), confirming the predictive capability of the methodology for both operating modes.
科研通智能强力驱动
Strongly Powered by AbleSci AI