Model predictive control as the cloud control strategy for a battery thermal management system

The fast charges are critical in supporting long-distance travels with battery based electric vehicles and alleviating range anxiety. Nonetheless, the battery heat generation under fast charge damages the battery if the battery thermal management system (BTMS) is not managed correctly. This paper proposes an innovative cloud control strategy for the BTMS that optimizes the electric consumption of the BTMS while managing the battery temperature under extreme events such as fast charges. The optimized control strategy is based on the concepts of model predictive control, which provides the optimized BTMS operation in the next 60 minutes. To do so, firstly, the different components that influence the optimization problem have been modeled. The models are the electro-thermal battery system model, the battery management system model, the energy control unit model and the BTMS model. Secondly, a model predictive control algorithm has been designed to avoid hard constraint violation along with an optimization approach that allows black-box model execution. Finally, the approach has been structured to be integrated as a functional mock-up into a digital twin application to solve current micro-controller computational limitations. The developed cloud control strategy shows a decrease of 1ºC on the maximum achieved temperature at fast charge events. • A MPC optimization algorithm that allows black-box model integration. • Hard constrain violation issue tackled with exponential soft constrains. • Modelling of all elements interacting in the battery thermal performance. • Cloud oriented optimized BTMS control under fast charge events.