Research On Energy Conservation Of Gasoline Hybrid Power System Based On Energy Flow Experiment And Simulation

Under the guidance of the trend of "electrification,intelligence,networking and sharing",the automobile industry is undergoing huge changes unseen in a century since its birth.Hybrid power system is one of the mainstream directions for the development of energy-saving vehicles at present stage.After years of development,hybrid power technology has been relatively mature,but there are still some defects in aspects like energy flow integration testing technologies and energy efficiency optimization methods.For example,the level of detail in energy decomposition is low,the simulation model cannot restore the actual vehicle operation state well,and the optimization strategy research method is incomplete.As a result,the energy-saving mechanism and the optimization direction can not be understood clearly by researchers.Therefore,it is urgently needed to achieve the breakthrough of methods and techniques in the above research fields to provide an effective way to improve the energy consumption of hybrid electric vehicles.In order to solve the above problems,the research on energy conservation of gasoline hybrid power system was carried out,taking the method of combining energy flow experiment and simulation.Firstly,the energy flow testing platform of hybrid electric vehicle was built,and the energy flow integration testing was carried out under New European Driving Cycle(NEDC)and Worldwide Light-duty Test Cycle(WLTC).Through the analysis of test data,the variation of energy flow at the system level and component level of gasoline hybrid electric vehicle and their relationships with the working process of the powertrain system have been explored.On this basis,the transient simulation model of hybrid electric vehicle was established to highly restore the vehicle operation process.Neural networks,dynamic programming,reinforcement learning and other artificial intelligence methods were introduced into vehicle energy flow simulation model whose optimization potential of energy consumption for hybrid electric vehicle was discussed.The main contents and results of this paper are as follows:(1)The topological structure of the target plug-in hybrid electric vehicle(PHEV)was analyzed.The basic theory of vehicle energy flow calculation suitable for new energy vehicles and conventional internal combustion engine vehicles(ICEV)was studied.A complete set of sensor arrangements and test settings for vehicle energy flow tests was developed,so that the mechanical power flow,electrical power flow and heat flow can be synchronously measured in the actual vehicle operation state providing theoretical and methodological supports for the energy flow evaluation.(2)A PHEV which is in charge sustenance phase and an ICEV with the same internal combustion engine were selected to conduct energy flow tests.The energy distributions under NEDC of cold start and warm start were compared.The research results show that,the working range of engine in PHEV was narrowed to a certain extent,but its tank-to-wheel efficiency had increased little.Compared with the tested ICEV,the energy-saving contribution rates of tested PHEV under NEDC are 5% for idling condition,20% for braking condition,and-5% for driving condition,respectively.(3)The reason why turbocharged gasoline vehicles were generally overcharged in naturally aspirated conditions was explored.To solve this problem,an optimization scheme of hybrid turbocharger based on equivalent consumption minimization strategy was proposed.The oil-electricity equivalence factor was introduced to equate the turbocharger motor power to engine fuel consumption.The one-dimensional numerical model of the engine was integrated into the iterative algorithm.The simulation results show that the energy saving rate of hybrid turbocharger in different driving cycles ranges from 1% to 5%.(4)The whole vehicle model of tested PHEV was constructed,which was coupled multiple physical fields such as machinery,electronic control,heat and fluid.The vehicle model was calibrated based on the energy flow test data.The simulation model provided a verification platform for the optimization strategy proposed later.Dynamic programming was used to get the global optimal solution,which provided a theoretical basis and evaluation benchmark for the optimization of energy management strategy.The simulation results show that,the theoretical maximum fuel-saving rate in the travel simulation of NEDC,WLTC and real driving condition are 11.37%,8.47% and 27.77%,respectively.(5)The optimal control strategy of PHEV based on navigation information was proposed.When the travel information was known,the strategy based on model predictive control was adopted.The radial basis function neural network was used to predict the future vehicle speed,and the dynamic programming algorithm was used in the prediction time domain to optimize the control of the powertrain.When the travel information was unknown,the strategy based on deep Q learning was adopted.The agent used driving experience to optimize the power output of the powertrain components.The simulation results show that,in the travel simulation of NEDC,WLTC and real driving condition,the strategy based on model predictive control can achieve fuel-saving rates of 7.51%,3.17% and 18.69%,respectively,and the strategy based on deep Q learning can achieve fuel-saving rates of 3.32%,2.02% and 14.51%,respectively.In this article,a set of vehicle energy flow evaluation method suitable for hybrid electric vehicles and conventional internal combustion engine vehicles had been established.The energy-saving mechanism of PHEV had been systematically analyzed,and a variety of energy-saving methods for gasoline hybrid system had been proposed.The research results can provide theoretical basis and data support for the design of the powertrain topology of hybrid electric vehicles,and propose effective recommendations for the development of the energy management control strategy of hybrid power system.