Predictive Energy Management Strategy Of Parallel Hybrid Electric Vehicles

With the increasing awareness of energy crisis and serious environmental pollution, Hybrid Electric Vehicle (HEV) has caught people’s great attention than before due to its advantages of energy saving and environmental protection. Hybrid Electric Vehicle, which incorporates the advantages of both the conventional fuel vehicle and the electric vehicle, has at least two kinds of energy sources. The complexity of hybrid drive system determines the complicacy of energy management strategy (EMS). With the prerequisite of satisfying the driving requirements, EMS of HEV controls the magnitude and flow direction of energy between each component, EMS is significant for rational use of carrying energy and realization of energy saving and environment protection. EMS has become one of the chief components in an automobile which also is the hot issues all over the world.In this paper, a new type energy management strategy for Hybrid electric vehicles is designed. Firstly, an overview of the status and trends of domestic and international development and the analysis of the key technologies of HEV is introduced.Then, the various existing energy management strategy is classified and analysis of the respective control principle. After that, the structure of various types HEV is introduced, especially analysis of the single axis parallel type hybrid system. The related parameters of main components are given, and the modeling of engine and batteries, etc is established combined with numerical method and the look-up table method.Parallel hybrid electric vehicle energy management strategy,based on the principle of predictive control, is designed and simulation experiments are done in the European car road test driving cycle, the urban road driving cycle, as well as cycle conditions of rural roads. The experiments result shows that the energy management strategy based on predictive control under the setting SOC can fully make use of the battery’s power and reduce fuel consumption more efficiently to achieve energy saving effect compared with the rule-based control strategy. Additionally a new energy management strategy which offered a prediction model for the training data, based on the relevance vector machine (RVM) algorithm is presented. Comparing with the predictive control EMS, the RVM strategy needs simpler calculation and is capable of faster response, which is much more suitable for real-time application.