Intelligent energy management of a fuel cell vehicle based on traffic condition recognition

This paper presents a methodological approach for intelligent control of fuel cell vehicles based on traffic condition recognition. For this purpose, employing an extensive real driving pattern database, a six-mode representative traffic condition is developed for the city of Tehran by means of fuzzy subtractive clustering approach. Subsequently, an adaptive fuzzy logic controller is designed, with the assistance of particle swarm optimization algorithm. Finally, a traffic condition recognition algorithm is proposed to establish the most probable driving mode. The fuzzy logic controller is employed as a real-time controller and its modes are singled out with respect to the traffic condition recognition algorithm results. Moreover, effectiveness of the proposed controller has been examined during several real driving periods containing various traffic conditions. Simulation results prove successful performance of the proposed intelligent controller under different traffic conditions according which a nine-to-seventeen percent fuel consumption improvement has been achieved.