Research On Gear Position Decision Of Hybrid Electric Vehicle Based On Neural Network

Hybrid electric vehicle(HEV)combines the advantages of traditional vehicle and electric vehicle,which can achieve the purpose of energy saving and emission reduction,and becomes a research hotspot in the field of new energy vehicles.Gear position decision is an important research issue in the automotive transmission systems.Research on intelligent and adaptive gear position decision method is of great significance to improve the power performance,fuel economy and ride comfort of HEV.Moving vehicle is a complex driver-vehicle-road closed-loop system.It is difficult to determine the optimal shift point to meet the customers' demands for vehicle performance in different driving intentions and driving conditions.Therefore,in this paper,taking the single shaft parallel hybrid electric vehicle as the research object,the neural network algorithm is adopted to learn the sample data containing driver and environmental information and generalize the vehicle gears.A nonlinear model is established between the vehicle state parameters and the optimal gear position.In view of with the deficiency of neural network algorithm,the following studies are carried out.(1)Analysis of the relationship between driver,vehicle and road,and the gear position decision scheme of this paper is worked out,which has considered the influence of driving intention,driving environment and vehicle status.(2)The driving intention and driving environment recognition method is studied.The vehicle running state parameters are collected,and the driver's intention is identified by fuzzy reasoning based on the vehicle parameters such as the acceleration pedal signal and the brake pedal signal.At the same time,Lagrange interpolation method and moving average method are used to identify the driving environment.(3)The neural network model is established which take the throttle,speed,acceleration,transmission input shaft speed as the control parameters,and the gear position as output.The neural network structure is designed as well,including network layers,the number of nodes in each layer,and so on.In order to avoid the local minimum of neural network,which leads to the local convergence of the network,genetic algorithm is adopted to optimize the weights and thresholds of the neural network.(4)In MATLAB,based on the recognition results,the gear position decision models are built respectively on the condition of rapid acceleration,uphill and bumpy,and the simulation and analysis are carried out.The simulation results show that the trained neural network can accurately predict the vehicle's gear position under the special driving intention and driving environment;and the precision of neural network model optimized by genetic algorithm is improved.