Modeling and Model Predictive Control of a Battery Thermal Management System Based on Thermoelectric Cooling for Electric Vehicles

Abstract

The active battery thermal management system is critical for the security of electric vehicles. In this article, a novel battery thermal management system and the control strategy based on thermoelectric cooling are proposed. A coupling model between the thermoelectric cooler and the battery pack is built by MATLAB/Simscape software. The model precision is verified through the experimental bench test, with a maximal deviation of 0.56 °C (the accuracy of the temperature sensor is ±0.1 °C). Further, a battery thermal management strategy with model predictive control (MPC) is proposed. In the results, it is elucidated that the MPC strategy has a superiority over the proportional‐integral‐derivation (PID) strategy in both the response time and energy consumption. Notably, the MPC strategy achieves a 35.17% reduction in response time and a 28.65% decline in energy consumption under a constant current of 2 C. Furthermore, during the variable H_N_U_F cycle conditions, the control effects are also evident, with an overshoot diminution of 12.2%, a maximum temperature error decrease of 23.85%, a response time reduction of 31.15%, and an energy utilization decline of 31.85%. In this study, a novel perspective in advancing battery thermal management systems through the application of thermoelectric cooler is provided.