شبيه‌سازي و ارزيابي عملكرد يك سيستم مديريت حرارتي باتري‌هاي ليتيومي به كمك الگوريتم يادگيري ماشين

Lithium-ion batteries generate a significant amount of heat during continuous operation with rapid charging and discharging, which can lead to a considerable increase in temperature, affect battery performance, reduce cycle life, and potentially cause a safety hazard. To address this issue, a phase change material (PCM)/mini-channel coupled battery thermal management system (BTMS) has been designed in this study to control battery temperature and improve performance. Additionally, a three-dimensional thermal model of the battery has been developed. Three-dimensional thermal simulations of an 18650 lithium-ion battery and a 75V lithium-ion battery pack consisting of 21 18650 battery cells are conducted based on a multi-domain, multi-scale battery modeling approach. The effects of different cooling methods on the battery and battery pack thermal management under rapid discharge conditions are investigated and compared. It has been found that for natural convection methods, the battery with PCM and the battery with PCM and fins easily exceeds 40°C under a C3 discharge rate. Under external short-circuit conditions, the cell temperature rises sharply and reaches 80°C in a short period, which can lead to thermal runaway and potentially result in catastrophic battery fires. On the other hand, the battery with the coupled BTMS effectively limits the battery temperature to a tolerable level under high-rate discharge conditions and increases the time to reach 80°C in short-circuit scenarios. Moreover, according to the 75V battery pack simulation, the thermal management system in this study results in better temperature uniformity. The effects of various mini-channel factors on battery temperature, especially the effect of different mini-channel geometries, type of mini-channel fluid, fluid velocity, and coolant fluid temperature, were also investigated. Additionally, with the increasing use of artificial intelligence (AI) in various fields, exploring AI approaches to eva‎luate different BTMS types seems valuable. Therefore, in this study, an artificial neural network (ANN) model was developed to predict the temperature of a lithium-ion battery equipped with a BTMS. The model inputs include the discharge rate (C1, C2, C3, and C4), mini-channel coolant fluid inlet temperature (15°C and 20°C), mini-channel coolant fluid inlet velocity (0.1 m/s, 0.5 m/s, and 1 m/s), ambient temperature (25°C and 40°C), and time (s). The model output is the battery temperature (°C). A total of 79,909 data points were used for training, validation, and testing of the model. The results of this study demonstrated the ANNʹs ability to predict battery temperature under various BTMS operating conditions. The mean squared error and mean absolute deviation of the model were 0.12 and 0.09, respectively. The results of this study confirmed the suitability of the ANN for predicting the performance of the coupled BTMS.

5

76156