Iterative Dynamic Programming Strategy for Electric Vehicle Battery Thermal Management Optimization

Abstract

Lithium‐ion batteries are extensively used in electric vehicles because of their superior performance in the general environment of global energy shortage. However, lithium‐ion batteries generate large amounts of heat as the battery packs are charged and discharged. If this accumulated heat is not dissipated timely, then the power supply system will fail, resulting in safety accidents. Therefore, a high‐efficiency and energy‐saving optimization strategy is required for both ensuring operating temperature of the battery and saving energy. Herein, a lumped thermal model is established on the basis of the heat generation characteristics of batteries and Newton's law of cooling, and its exactitude is proven by comparison with a liquid cooling system established using AMESim. In addition, to reduce computation time and enhance the real‐time capability, an iterative dynamic programming (IDP) method is proposed to determine the optimal values iteratively in a multidimensional search space, in view of the high nonlinearity and time‐variability of a battery thermal management system as well as the complexity of dynamic programming (DP). Thus, the results from co‐simulations performed in Matlab and AMESim verify that the proposed strategy has high cooling efficiency and can save considerable energy for cooling system compared with PID and DP.