An all-in-one design method for plug-in hybrid electric buses considering uncertain factor of driving cycles. (1st November 2019)

• Record Type: Journal Article
• Authors: Hou, Daizheng; Sun, Qun; Bao, Chunjiang; Cheng, Xingqun; Guo, Hongqiang; Zhao, Ying
• Is Part Of: Applied energy. Volume 253(2019)
• Publication Date: 2019-11-01
• Subjects: Plug-in hybrid electric bus -- All-in-one method -- Typical driving cycle -- Energy management -- Receding horizon control; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2019.113499
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: A typical driving cycle method is proposed for fixed city bus route. A robust all-in-one method is proposed for the fuel economy improvement of PHEB. A receding horizon control-based energy management strategy is proposed. A predictive model of terminal SOC is designed for energy management control. Abstract: The interaction between components and energy management is an obstacle for the fuel economy improvement of plug-in hybrid electric buses (PHEBs). Although the systematic method of design, optimization and energy management is recognized as a promising solution, two problems including the robust design of component matching considering the influence of uncertain driving cycles and the integrated design of energy management are still need to be solved. This paper proposes a novelty all-in-one method to address the issues. Firstly, a typical driving cycle construction method for city bus route is proposed to solve the first problem, based on a series of historical driving cycles. Secondly, a co-optimization framework including an outer layer constituted by multi-island genetic algorithm, and an inner layer constituted by dynamic programming (DP) is proposed, to find the best component matching. Specially, in the outer layer, two engines and six motors are taken as discrete design variables, meanwhile, the speed ratios of AMT and final driver are taken as continuous design variables; in the inner layer, the fuel consumption is taken as the objective of the outer layer. Finally, a receding horizon control (RHC)-based energy management strategy together with a predictive model of terminal state of charge (SOC) are proposed to solve the second problem, based on the same DP algorithm in the co-optimization framework. Simulation results demonstrate that the proposed all-in-one method can find the robust component matching; the RHC strategy can realize the real-time control, and its fuel economy is better than the rule-based strategy.