Extended-range electric vehicle energy management system and method considering low temperature heat demand

Abstract

This invention discloses an energy management system and method for range-extended electric vehicles that considers low-temperature thermal demand. By establishing a dual-model training unit, a dual-model fusion prediction reference generation unit, and an MPC real-time decision and closed-loop feedback unit, it solves the pain points of insufficient collaborative optimization of power and thermal management at low temperatures and the difficulty in balancing global optimization and real-time performance. It achieves global optimization of the two state variables, SOC and temperature, through a state grid adaptive dynamic programming algorithm. The optimal reference trajectory and predicted vehicle speed are generated online by the LSTM+RBF dual model. Combined with MPC rolling optimization, it achieves collaborative control, effectively improving fuel economy, reducing battery life degradation, and taking into account the thermal comfort of the passenger compartment. It has strong engineering practicality.

Images