Research On Intelligent Control Strategy Of Agricultural Diesel Engine Based On The Load Torque Model

Agricultural tractors are one of the most important agricultural production operations and play an important role in the field of agricultural machinery and equipment.However,the operating environment of the tractor is complicated and the workload varies drastically.In actual production,there are problems such as insufficient power,high fuel consumption,and poor discharge performance.On the basis of extensive research on relevant literature,this article focuses on three aspects of tractor load forecasting model,mean value modeling of tractor diesel engine,and feedforward-feedback intelligent control strategy.This paper constructs a forecast model of farm tractor load.The data collection experiment for typical operating conditions of agricultural tractors was designed and carried out.The tractor's deep loose plough was used as an example to establish the dynamic model of the tractor's working unit.The main source of resistance of the tractor's working unit was analyzed and based on the CAN bus data.The data sets of the tractor operating conditions were calculated and the effective operating segments were divided;the resistance of the agricultural implements was separated from the total resistance using a dynamic model,the resistance of the agricultural implements was characterized by the soil specific resistance,and the soil resistivity was calculated by the BP neural network model.The law of variation was studied offline,and a prediction model of soil specific resistance was established.Based on the experimental data,the validity of the load prediction model was verified.Based on MATLAB/Simulink,an average simulation model of the target model WeiChai WP6 turbocharged diesel engine was established.This article analyzes and introduces the mathematical model modeling principles of diesel engines,divides the turbocharged diesel engine into intake,exhaust,intercooler,turbocharger,cylinder and crankshaft,etc.Modeling sections;Based on the classical 4th-order Runge-Kutta formula ODE4 algorithm,the selected average operating point is used to verify the established average model.The simulation results show that the average model has a fast calculation speed and the simulation results are more accurate.The controlled object of diesel engine control strategy research.The research of feedforward-feedback compound control strategy was carried out.The feedforward control part is based on the soil resistivity prediction model and compensates for the load disturbance caused by the soil resistivity change.The feedback control part adopts BP neural network PID control algorithm.The algorithm can adjust PID controller proportionally,integrally,differentially online.Coefficients to realize intelligent adaptive control of load changes;Set up a control system simulation platform to simulate sudden increase or abrupt reduction of operating load;set four sudden simulations of sudden increase in speed,sudden decrease,and match the simulation results with corresponding actual operating conditions Data comparison;results show that the feedforward-feedback compound control algorithm reduces the adjustment time by 47%,66%,69%,and 42%,respectively,compared with the traditional PID control,and the speed change amount decreases by 42%and 29%,respectively,and the control effect is significantly improved.