Driving Pattern Recognition and Energy Management for Extended Range Electric Bus

Driving pattern recognition is an efficient way to ensure that the energy management strategy is suitable to the current driving pattern, so as to obtain a better performance of fuel economy. In this work, we propose an approach to recognize the driving pattern online and apply the results of recognition to minimize the fuel consumption of the bus with state-of-the-art equivalent consumption minimization strategy (ECMS). Firstly, we characterize the driving pattern with a feature vector composed of five statistical parameters; secondly, we evaluate the similarity between the current driving pattern and four typical driving patterns by calculating the Euclidean distance between their feature vectors respectively; thirdly, we choose the one typical driving pattern with the minimum Euclidean distance to the current driving pattern as the recognized driving pattern; finally, based on the results of recognition, we adjust the equivalent factor of ECMS to implement an optimal energy management strategy online. Simulation results show that ECMS with the approach of driving pattern recognition can realize an optimization of fuel economy.