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

Hybrid electric vehicle(HEV),as one of the new energy vehicles,can effectively reduce pollutant emissions.At the same time,the demand for new energy infrastructure is low and the technology is relatively mature,which is the most suitable new energy vehicle model for popularization at the current stage.As the soul of a hybrid electric vehicle,energy management strategy is an important direction of hybrid electric vehicle research.Superb energy management strategy can reasonably distribute the energy between engine and motor,so that the vehicle can reduce pollutant emissions as much as possible while meeting the dynamic requirements.At present,research on energy management strategies is mostly directed to a specific driving cycle,which is quite different from the actual driving condition,so that the vehicle cannot achieve the optimal performance in actual driving.In this paper,the energy management strategy based on driving cycle recognition is studied for a new type of three-gear hybrid electric vehicle,in order to reduce fuel consumption while meeting the vehicle dynamic performance indexes.1)Parameters matching and modeling of power system.According to the structure and parameter requirements of the new hybrid electric vehicle,the whole vehicle,engine,motor,transmission ratio and battery are matched and selected.The whole vehicle and power sources are modeled mathematically according to the matching results.2)Establish two energy management strategies based on known driving cycle.Establishing logic threshold energy management strategy based on vehicle demand power and battery SOC value as threshold.Taking the battery SOC and transmission ratio as the state variables,engine torque,the motor 2 torque and gear as decision variables to establish a global optimization energy management strategy.To verify that the global optimization strategy has better fuel economy,the vehicle performance under the same driving cycle using the above two control strategies is simulated.3)Selection of typical driving cycle.Thirty-three typical driving cycle are classified into five categories by using 24 characteristic parameters,so that the typical driving cycles can be selected to represent most of the driving conditions.The optimal energy allocation strategy for each typical driving cycle is calculated by using global optimization algorithm,and then the energy allocation law of the optimal strategy for each driving cycle is learned and extracted by using neural network.By extracting the characteristic parameters of real-time driving cycle and comparing them with the typical driving cycle,then the real-time driving cycle is identified and classified,so that the strategy of driving cycle identification is established.4)Establish energy management strategy based on cycle recognition.Combining the strategy of cycle recognition with the global optimization energy management strategy based on neural network learning,energy management strategy of hybrid electric vehicle based on cycle recognition is finally established.In order to verify the feasibility of the energy management strategy,the simulation and comparison of the hybrid electric vehicle energy management strategy based on cycle recognition,the logic threshold and the global optimization are carried out in Matlab software.This dissertation designs and optimizes the energy management strategy of the new three-speed hybrid vehicle,and the simulation results is shown that the energy management strategy can effectively reduce the fuel consumption of the vehicle.