Supervised-Learning-Based Optimal Thermal Management in an Electric Vehicle

Abstract

Due to the increasing market share of electric vehicles (EVs), the optimal thermal management (TM) of batteries has recently received significant attention. Optimal battery temperature control is challenging, requiring a detailed model and numerous parameters of the TM system, which includes fans, pumps, compressors, and heat exchangers. This paper proposes a supervised learning strategy for the optimal operation of the TM system in an EV. Specifically, for TM subsystems, individual artificial neural networks (ANNs) are implemented and trained with data obtained under normal EV driving conditions. The ANNs are then interconnected based on the physical configuration of the TM system. The trained ANNs are replicated using piecewise linear equations, which can be explicitly integrated into an optimization problem for optimal TM scheduling. This approach enables the application of a mixed-integer linear programming solver to the problem, ensuring the optimality of the solution. Simulation case studies are performed for the two operating modes of the TM system: i.e., integrated and separate modes. The case study results demonstrate that the ANN-based model successfully reflects the operating characteristics of the TM system, enabling accurate battery temperature estimation. The proposed optimal TM strategy using the ANN-based model is verified as effective in reducing the total energy consumption, while maintaining the battery temperature within an acceptable range.