As a new type of vehicle power system with zero emission and renewable energy, fuel cell has high research value. In this paper, a fuel cell energy management strategy based on deep reinforcement learning is proposed for high-speed traffic scenarios. A bus working condition with real high-speed traffic scenarios is obtained through NGSIM's real high-speed road segment data set screening. The energy management strategy effectively reduces the hydrogen consumption by 4.21% and the power fluctuation by 17.2%, which effectively improves the durability of fuel cells.
Fuel cell bus energy management based on deep reinforcement learning in NGSIM high-speed traffic scenario
2022-10-28
4827589 byte
Conference paper
Electronic Resource
English
Making NGSIM Data Usable for Studies on Traffic Flow Theory
Transportation Research Record | 2013
|Developing extended trajectory database for heterogeneous traffic like NGSIM database
Taylor & Francis Verlag | 2022
|Transportation Research Record | 2016
|