Research On Transient Engine Model Of Hybrid Electric Vehicle

The majority of the world's nations have been plagued by global energy crisis and environmental pollution,which leads automakers to develop,produce and sell new energy vehicles so as to reduce energy consumption and tailpipe emissions.As one important category of the new energy vehicles,hybrid electric vehicles are the current focus of the automotive industry and academia.Hybrid electric vehicles contain two power sources,engine and motor,which present different transient responses.Individual control of engine or motor can lead to insufficient power output or successive torque fluctuation,thereby jeopardizing the vehicle driving performance and fuel economy.The shortcoming of individual control necessitates the introduction of coordinated control to the engine and motor.The problem of the existing coordinated control algorithm-engine torque open-loop control+ engine transient torque estimation + motor torque compensation-lies in the engine transient torque estimation.The foundation of engine transient torque estimation is the establishment of engine model which accurately reflects the steady and transient process of the engine.Therefore,it is of vital significance to investigate the establishment of engine model.In this dissertation,based on the state of the art of the engine modeling theory,the problem of the existing hybrid electric vehicle powertrain is looked into,and the theory of engine modeling method is also investigated.The main contents of this study are as follows:(1)Engine bench test is carried out.Based on the test data,the influences of the engine operating parameters(such as engine speed,intake manifold pressure,ignition advance angle,excess air coefficient,intake advance angle)on the engine performance are studied.(2)The BP neural network modeling theory,mean value model(modeling?)theory,computational fluid dynamics modeling theory and thermodynamic process theory are studied.The BP neural network engine model based on the engine bench test data is established.Using the mean value model theory,the mean value model is built.Combined with the test data,the classic models of the average intake sub-model,power output sub-model are revised,and the key parameters of the revised models are calibrated.Based on the computational fluid theory,combined with the engine bench test data and engine structure parameters,the quasi-dimensional combustion model is established using the GT-Power software.In accordance with the thermodynamic theory,the discrete thermodynamics process is established in the cylinder model.(3)Based on the engine model established in chapter 3,the engine steady-state simulation is carried out,and the model simulation results are compared with the bench test data.The transient simulation with GT-Power model,BP neural network model and mean value model is carried out.Based on the GT-Power model simulation results,the results of BP neural network model and the average model are compared.Based on the discrete thermodynamics process in the cylinder model,the influences of the engine operating parameters such as engine speed,intake manifold pressure,spark advance angle are studied.(4)The vehicle model of HEV whose engine model is BP neural network model and mean value model,is analyzed and compared with the simulation result.