Performance analysis and prediction of hybrid battery thermal management system integrating PCM with air cooling based on machine learning algorithm

Targeted at addressing the battery temperature concerns of rapid discharging and extreme ambient temperature, a hybrid battery thermal management system (BTMS) integrating phase change material (PCM) with air cooling is introduced, in which the Cantor fractal fin and segmented PCM block are employed to enhance thermal management capacity. Through the numerical simulation, the effects of fin parameters, filling strategy of segmented PCM block and airflow direction are analyzed. Moreover, the machine learning algorithm is adopted to predict the thermal control of hybrid BTMS. The results demonstrate that compared to the rectangular fin, the Cantor fractal fin reduces the maximum temperature (T-max) of batteries by 0.53 K, and continually increasing the fractal number of fin improves the cooling performance, but the improvement extent is restricted. The strategy that the middle section is filled with RT35 and the bilateral section is filled with RT35HC provides better temperature management capacity, and so does the strategy that the PCM filling proportion of middle section maintains at 0.5. Applying the segmented PCM block, the T-max and maximum temperature difference (Delta T-max) of batteries are kept at 308.43 K and 1.64 K at 4C discharge rate respectively. Varying airflow direction hardly affects the T-max, however the Delta T-max is obviously decreased when the reverse flow is utilized. The Delta T-max with reverse flow is reduced by 17 % and 21.2 % separately in contrast to those with concurrent flow and staggered flow. The proposed hybrid BTMS owns superior cooling and insulating capacities under rapid discharge rate, high ambient temperature and low ambient temperature. Based on the machine learning algorithm, a back propagation neural network is established and can accurately predict T-max and Delta T-max of batteries.