Adaptive Equivalent Consumption Minimization Strategy for Hybrid Heavy-Duty Truck Based on Driving Condition Recognition and Parameter Optimization

Abstract

The accurate determination and dynamic adjustment of key control parameters are challenges for equivalent consumption minimization strategy (ECMS) to be implemented in real-time control of hybrid electric vehicles. An adaptive real-time ECMS is proposed for hybrid heavy-duty truck in this paper. Three efforts have been made in this study. First, six kinds of typical driving cycle for hybrid heavy-duty truck are obtained by hierarchical clustering algorithm, and a driving condition recognition (DCR) algorithm based on a neural network is put forward. Second, particle swarm optimization (PSO) is applied to optimize three key parameters of ECMS under a specified driving cycle, including equivalent factor, scale factor of penalty function, and vehicle speed threshold for engine start-up. Finally, combining all the above two efforts, a novel adaptive ECMS based on DCR and key parameter optimization of ECMS by PSO is presented and validated through numerical simulation. The simulation results manifest that proposed adaptive ECMS can further improve the fuel economy of a hybrid heavy-duty truck while keeping the battery charge-sustainability, compared with ECMS and PSO-ECMS under a composite driving cycle.

Keywords

• hybrid heavy-duty vehicle
• particle swarm optimization
• equivalent consumption minimum strategy
• driving condition recognition