Study Of Energy Management Strategy For Plug-in Hybrid Electric Bus Based On Driving Cycle Recognition

With the increasingly serious environmental and energy problems,the development of pure electric vehicles has become the consensus of the global automotive industry.Hybrid electric vehicles play a transitional role from traditional fuel vehicles to pure electric vehicles.Compared with hybrid electric vehicles,hybrid electric buses can not only alleviate traffic congestion,but also reduce per capita pollutant emissions,and have better energy saving and emission reduction effect.According to the characteristics of multi-energy sources in hybrid electric system,reasonable energy management strategy can ensure the coordinated operation of components of hybrid electric system and improve fuel economy,which is one of the key technologies in hybrid electric bus research.At present,the energy management strategy of hybrid power system mainly includes rule-based control strategy and optimization-based control strategy.Combining the advantages of the two control strategies,this paper designs an energy management strategy based on driving cycle recognition.Firstly,according to the characteristics of Hefei's mixed traffic,the representative urban working conditions of Hefei are constructed by using principal component analysis and fuzzy C-means clustering method.Based on the working conditions and performance indicators,the key components and parameters of the sample car are selected and matched.Secondly,a dynamic programming algorithm suitable for plug-in hybrid electric bus is designed to calculate the global optimal energy management strategy and extract the optimal energy management rules.Then the LVQ neural network is used to identify the driving conditions,and the working modes are determined according to the recognition results.In each working mode,the global optimal energy management rules are used to control the hybrid power system.Finally,the simulation results based on AMESim and Simulink joint simulation platform show that the proposed energy management strategy has a good control performance.