Research On Energy Management Strategy Of Hybrid Electric Vehicle Based On Driving Cycle Identification

The introduction of the new version of the national emission standards has accelerated the transition from traditional fuel vehicles to new energy vehicles.As an excessive model of the development history of new energy vehicles,hybrid electric vehicles have the outstanding advantages of low emissions and long mileage,and have become the research hotspot of many companies and scientific research institutes.The formulation of reasonable energy management strategies is to reduce the energy consumption and emissions of the entire vehicle.The essential.However,the traditional energy management strategy based on rules and global dynamic optimization cannot make corresponding adjustments and poor applicability following the changes in vehicle driving conditions,resulting in low energy utilization.In order to improve fuel economy,there is an urgent need for energy management strategies that can be adjusted in real time with driving conditions and have better applicability.The thesis is based on the "Xi’an Electric Vehicle Typical Driving Conditions Construction Project" and takes a plug-in single-axle parallel hybrid vehicle as the research object to carry out multi-mode identification energy management strategy research.The thesis takes the urban roads of Xi’an as the background,obtains the operation data of electric vehicles through the actual vehicle test method and divides it into short trips,adopts the principal component analysis method and the K-means clustering method to obtain 4 types of representative working conditions,and establishes based on the representative working conditions Typical urban operating conditions with multiple operating conditions;based on the established operating conditions,the power system parameters of the selected vehicle model are matched and the power system model is established.After the rule-based engine switch control strategy simulation verification,the power system performance meets the requirements of driving conditions;The thesis constructs an energy management strategy through the dynamic programming algorithm with global optimization capabilities and the principle of minimum equivalent fuel consumption(ECMS)with local optimal capabilities,and performs simulation verification under typical driving conditions.The results show that the paper adopts energy management The strategy can achieve good fuel economy,and can adapt to the characteristics of vehicle driving conditions and random changes in traffic operating environment.In order to adapt to the driving characteristics of multiple operating conditions,based on the learning vector quantization(LVQ)neural network and ECMS principle,the operating condition recognition model and the optimal control strategy database under different operating condition categories were established,and the simulation was performed under multiple operating conditions.The results show that the energy management strategy of multi-condition recognition used in this paper can improve fuel economy,improve the efficiency of condition recognition,and achieve the expected research goals.The research results of the thesis have important practical significance for improving the fuel economy of hybrid vehicles and achieving energy saving and emission reduction.