Driving Cycle Recognition Based Adaptive Equivalent Consumption Minimization Strategy for Hybrid Electric Vehicles

Abstract

Hybrid electric vehicles (HEVs) are considered the most practical option for reducing emissions and fuel consumption of conventionally powered vehicles. Energy management strategies (EMSs) are the core technology of HEVs because of decreasing the cost of the system and limiting its negative effects. Equivalent consumption minimization strategy (ECMS) can achieve instantaneous optimal control and has attracted attention in recent years. In this study, an adaptive equivalent consumption minimization strategy (A-ECMS) based on driving cycle recognition is constructed for a parallel HEV. First, select the standard driving cycle and analyze its characteristic parameters. And then training learning vector quantization (LVQ) neural network-based driving cycle recognizer to achieve an average of 98% accuracy. At last, the optimal equivalent factor (EF) is selected for ECMS by recognizing the current driving cycle. It is jointly simulated and analyzed by AVL CRUISE and MATLAB/Simulink software under NEDC and CHTC-LT driving cycle. The results show that compared with the logic-based EMS, in the NEDC driving cycle the 100 km fuel consumption of A-ECMS decreases by 3.8%, and the battery state of charge (SOC) increases by 1.1%. In the CHTC-LT driving cycle, fuel economy improves by 3.6%, proving the superiority of the A-ECMS.

Keywords

• Hybrid electric vehicle (HEV)
• energy management strategy (EMS)
• equivalent consumption minimization strategy (ECMS)
• driving cycle recognition
• learning vector quantization (LVQ)