关键词: 拖拉机, 换挡规律, 神经网络, 速度修正, 自适应换挡

Abstract: The tractor pulls  different agricultural implements  to complete different operations, and its operation requirements change with the different traction machinery. Aiming at the problem that highpower tractors have different operating requirements under different working conditions, a multiworking condition shifting adaptive control method is proposed. Firstly, the requirements of different conditions on the power and economy of the tractor are analyzed. Slip rate, the throttle opening and vehicle speed are chosen as the control parameters, according to the requirements of the workings conditions, calculating the theoretical shift schedule that takes into account the power and the economy, and adopt neural network offline training shift schedule to realize intelligent control of shift control. Secondly, in view of the problem of cyclic shifting caused by random load fluctuations under heavy loads, the acceleration and the throttle opening changes are introduced as parameters. Fuzzy logic is used to judge the tractor load and the drivers operation intention to obtain the speed correction coefficient, and the shifting speed is corrected to expand the range of shift. A longitudinal dynamic simulation model of a highpower tractor is built by Simulink to verify the effectiveness of the shifting strategy. The simulation results show that in terms of fuel economy, the fuel economy of road transportation and light load operating conditions decreases by 5.78% and 3.28%, respectively. In terms of power performance, while ensuring the overcoming of traction resistance, the acceleration time under light load and heavy load conditions is faster, the speed fluctuation is reduced, and the problem of cyclic shifting under heavy load conditions is effectively avoided.

Key words: tractor, shift schedule, neural network, speed correction, adaptive control