An Online Energy Management Strategy for a Fuel Cell/Battery Vehicle Considering the Driving Pattern and Performance Drift Impacts

Energy management strategy (EMS) has a profound influence over the performance of a fuel cell hybrid electric vehicle since it can maintain the energy sources in their high efficacy zones leading to efficiency and lifetime enhancement of the system. This paper puts forward an online multi-mode EMS to efficiently split the power among the components while embracing the effects of the driving conditions and performance degradation of the fuel cell system. In this regard, firstly, a self-organizing map (SOM) is trained to cluster the driving patterns. The SOM competitive layer in this work is composed of ten driving features as inputs and it classifies the driving patterns into three classes in the output. Subsequently, a three-mode fuzzy logic controller (FLC) is designed and optimized offline by the genetic algorithm for each driving pattern. Unlike the other similar works, the output membership function of the FLC is designed based on the online identification of the maximum power and efficiency of the fuel cell system which change over time. Finally, the SOM is utilized to recognize the driving mode at each sequence and accordingly activate the most sui