Machine Learning-Based Pressure Drop Prediction for Enhanced EV Battery Thermal Management

It is important to develop active thermal management systems for electric vehicle (EV) batteries to ensure that they work safe and durable. The major challenge is to find the right pressure loss in the coolant circuits as it directly affects the energy conversion efficiency and the cooling system. A new approach to the development of a machine learning based approach to predict pressure drop across a wide range of flow methods The study presents a machine learning-based approach to predict pressure drop across various circuit topologies and flow rates. A dataset containing information on computational fluid dynamics (CFD), coolant properties, experimental data, and circuit geometry was curated to address limitations in existing methods. This dataset was utilized to model complex nonlinear relationships using supervised learning algorithms, including gradient boosting and neural networks. The proposed models were found to be accurate and generalizable to other datasets and were validated using experimental data and CFD simulations. The results show that the application of machine learning can bring about a significant change in the development of cooling systems of EV batteries and can be useful in improving the current industry standards. This work presents a data driven method to enhance the thermal management of next generation EVs.