Research On Shifting Strategy And Automatic Control Method Of Heavy Tractor Based On BP Neural Network

Heavy tractors generally refer to tractors with high horsepower.As an important agricultural power machinery,heavy tractors play a huge role in agricultural production and can be used for deep tillage,ploughing,combined harvesting and so on.Because of the different working conditions,the resistance changes at all times,which requires constant adjustment of gear to adapt to the changing working environment.With the increase of power,it is difficult for heavy-duty tractors to be in the best working condition all the time,which reduces the working efficiency.Transmission is the core of tractor power assembly system.Shifting strategy and control method are the core of automatic transmission.Their advantages and disadvantages determine whether automatic transmission meets the requirements of vehicle power,economy and safety.Compared with automobiles,the structure and use of tractors are quite different.It is impossible to directly determine the shift law of tractors by using the shift strategy of automobiles.Therefore,it is necessary to study the shift strategy of tractors.The specific research contents and conclusions are as follows:(1)On the basis of studying the development course and current situation of tractor automatic gear shifting,the shift methods and research emphases adapting to the series of models studied in this paper are put forward.The characteristics of tractor diesel engine are studied.The working characteristics of diesel engine are analyzed concretely.The optimal power and fuel economy working range are determined.The main resistance sources in the working process of tractor are studied and the main working characteristics are defined.(2)According to the load characteristics and working characteristics of tractor diesel engine,its working environment is roughly divided into three types,and different shifting modes are divided and formulated.According to the different operation requirements of each mode,the mathematical models of diesel engine speed,throttle opening and engine torque are established.Choose the appropriate input parameters to determine the shift strategy in each mode.(3)According to the example of tractor,the whole vehicle and other parameters are obtained,and the shift rules are formulated under the specific mode.The shift rules under different working conditions are trained by using MATLAB neural network.The error of the neural network is analyzed according to the fixed input parameters and the ideal experimental results,and the error is gradually reduced.Finally,the ideal shift results are obtained.(4)Simulink simulation was carried out on the experimental results,and the simulation results were compared with the actual situation.Four cases were compared,including shift accuracy,fuel consumption,lane change and bumpy road surface.The results of the research and development of shift law were proved and the future research was prospected.