Study on Thermal Control Strategy of Integrated Thermal Management System for Electric Vehicle

Abstract

To address the challenges of poor temperature control precision in electric vehicle battery and cabin, as well as high overall energy consumption, an innovative control strategy for an integrated thermal management system (ITMS) is proposed. Through systematic analysis of the thermal management system’s (TMS) dynamic characteristics and intricate coupling relationships, critical control requirements for battery and cabin thermal management were identified. A novel deep learning model, integrating a convolutional neural network (CNN) and a long short-term memory (LSTM) network, is developed. This unique architecture effectively extracts spatiotemporal features, enabling precise temperature trend prediction and adaptive control. Simulation results demonstrate significant improvements over conventional PID control. Thermal stabilization was reduced by approximately 40–43%. Temperature control precision was enhanced to within ±0.3°C across diverse operating conditions, while compressor energy consumption was significantly decreased. These advancements resulted in an average 2.1% increase in energy efficiency coefficient and a 1.8% improvement in exergy efficiency, accompanied by a substantial reduction in performance fluctuations. Simulation evaluations across the new European driving cycle (NEDC), worldwide harmonized light vehicles test cycle (WLTC), and China light-duty vehicle test cycle (CLTC) confirm the superiority of the proposed CNN-LSTM strategy in temperature regulation, energy utilization, and system stability. This approach provides an effective solution for optimizing next-generation TMS for electric vehicle.

Keywords: Electric vehicle, Integrated thermal management systems, CNN-LSTM