Genetic algorithm-based fuzzy optimization of energy management strategy for fuel cell vehicles considering driving cycles recognition. (15th January 2023)

• Record Type: Journal Article
• Authors: Wang, Yichun; Zhang, Yuanzhi; Zhang, Caizhi; Zhou, Jiaming; Hu, Donghai; Yi, Fengyan; Fan, Zhixian; Zeng, Tao
• Is Part Of: Energy. Volume 263:Part F(2023)
• Publication Date: 2023-01-15
• Subjects: Energy management strategy -- Fuel cell vehicle -- Fuzzy logic control -- Driving cycles recognition -- K-means clustering; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2022.126112
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The energy management in fuel cell vehicles (FCVs) is crucial to maintain the economical operation of FCVs and the fuzzy logic control (FLC) is mainly used to manage the energy split between the fuel cell and other energy sources. To overcome the limitation of traditional FLC, the dependence on expert knowledge leading to the insufficient energy split, this paper proposes strategy optimization based on FLC with driving cycles recognition achieve near-optimal fuel economy and stable battery charge sustenance. Initially, the whole FCV model is established, which includes electrical system, vehicle dynamic system, energy management system. Additionally, with the objective function which is the minimum equivalent hydrogen consumption of four typical driving cycles, the centers and widths of FLC membership function are optimized by genetic algorithm (GA), respectively. Finally, the driving cycles recognition is achieved based on K-means clustering method, and characteristic parameters are extracted and classified. Compared with GA-optimized fuzzy EMS and the traditional fuzzy EMS, the simulation results demonstrate that the equivalent hydrogen consumption based on proposed strategy is reduced by 16.55% and 40.50%, respectively. Therefore, the proposed strategy can effectively smooth the output of proton exchange membrane fuel cell (PEMFC) and enhance the total fuel economy. Highlights: The driving cycles recognition based on unsupervised method is developed. An optimized fuzzy logic control optimized is investigated. An entire FCV model is established consisting of multiple subsystems. Smoother output of fuel cell is achieved while enhancing the total fuel economy.