Research On Modeling And Nonlinear Control Method Of Start-stop System For Mild Hybrid Electric Vehicles

Idle speed conditions,as one of the five basic working conditions for the operation of an engine used by vehicles,consume a lot of fuel every year.The emergence of the idling start-stop technology not only solves the phenomenon of "out-of-work",but also implements the concept of energy-saving and emission reduction.It is mature and widely used due to its economic and environmental advantages.The mild hybrid diesel vehicle with start-stop technology has lower cost due to the less modification to the original vehicle.At the same time,it combines the advantages of high thermal efficiency and low harmful gas emissions.So it has better fuel economy,clean and environmental protection,and has broader market prospects.However,because the mild hybrid diesel vehicle can not achieve pure electric operation at low speed,the engine needs frequent start and stop.If the start-stop action is not effectively controlled,not only will it cause more fuel consumption,but it will also seriously affect the driver's driving experience.This paper studies the modeling and control of the start-stop system of the mild hybrid electric vehicles.Through the tracking control of the engine speed of the start-stop system,the mild hybrid diesel vehicle can realize the smooth and fast start-stop.Firstly,according to the structural characteristics and functional characteristics of the start-stop system based on the belt driven starter generator(BSG)in the mild hybrid electric vehicle with diesel engine,a BSG model,the belt drive system model and the traditional starter motor model,combined with the four-stroke diesel engine model were built in the AMESim Rev 13 simulation software.They together formed a complete diesel engine start-stop system model.Through the analysis of the characteristic curve of the diesel engine model and the analysis and comparison of the changes of the modules in the start-stop system under different conditions,the simulation model was verified functionally rationality.Then,according to the driving state of the vehicle and the driver's operation,the control strategy of the start-stop system was designed.The conditions for the start-stop function to be turned off or turn on were specified.And when the engine start-stop function was turned on,it can correctly judge the driver's intention to start and stop.For the start-stop control of the engine,combined with the characteristics of the engine not spraying during the starting process of the BSG motor,a feedforward term that can estimate the change of the compressive resistance torque in the cylinder was designed.It can attenuate large fluctuations in engine speed due to the high compression ratio of the diesel engine.On this basis,a linear error-oriented model for control was obtained by linearizing the engine start-stop system model.A linear engine start-stop controller based on the linear model predictive control(LMPC)was designed.The controller can optimize the torque demand of the BSG motor and apply it to the motor to eliminate the engine speed deviation for the goal of smooth start and stop of the engine.And the effectiveness of the control algorithm through the joint simulation of AMESim and Simulink was verified.Finally,in order to avoid the adverse effects of the system's nonlinear dynamic charac-teristics neglected,this paper designed an engine start-stop nonlinear controller based on the nonlinear model predictive control(NMPC).The particle swarm optimization(PSO)algorithm was used to optimize the nonlinear start-stop controller,which reduced the adjustment parameters and reduced the computational complexity of the online solution.The effectiveness of the designed nonlinear start-stop controller was verified by the join-t simulation of AMESim and Simulink under different sets of idle conditions.At the same time,it can be seen that the nonlinear controller had better control effects in both transient and steady state than the LMPC controller.