Authors

Abstract

As the core component of the tractor operation system, the electro-hydraulic hitch system is the key equipment to realize the implement hitching, lifting adjustment, tillage depth control and adaptive posture control. The hydraulic valve-controlled cylinder system serves as the core control unit of agricultural tractor electro-hydraulic hitch systems. Given that the hydraulic system of tractor hitch devices exhibits strong nonlinearity, uncertainty, and time varying parameters, this study focuses on the valve-controlled cylinder electro-hydraulic servo system and proposes a composite control strategy that integrates linear active disturbance rejection control (LADRC) with parameter optimization via genetic algorithm (GA). This strategy utilizes the linear extended state observer (LESO) of LADRC to estimate and compensate for the total disturbance of the system in real time, while adaptively adjusting the parameters of LADRC using the multi objective optimization characteristics of GA, overcoming the problems of observation lag and gain conflict in traditional trial and error parameter tuning. To verify the control performance, a co-simulation model of the servo valve-controlled cylinder system based on MATLAB-AMESim platform was constructed. The simulation results demonstrate that GA-LADRC achieve significantly superior control performance compared to the PID controller, and in step signal tracking, GA-LADRC reduces both the overshoot and tracking error by more than 50% compared to general LADRC. For sinusoidal signal tracking, GA-LADRC exhibits an 18% reduction in phase lag, an 18.3% improvement in accuracy, and a 34.8% decrease in the integral square error (ISE) compared to general LADRC. Furthermore, under disturbance conditions, GA-LADRC also demonstrates superior anti-interference ability. These results confirm the stability and effectiveness of the proposed GA-LADRC strategy, and the developed method is expected to provide technical support for the fine plowing operation of tractors.

Abstract in Chinese

电液悬挂装置作为拖拉机作业系统的核心组成部分，是实现农具挂接、升降调节、耕深控制及作业姿态自适应的关键装备，液压阀控缸系统是农用拖拉机电液悬挂装置的核心控制单元。针对拖拉机悬挂装置液压系统具有强非线性、不确定性和时变参数的特点，本研究以阀控缸电液伺服系统为研究对象，提出了一种融合线性自抗扰控制（LADRC）与遗传算法（GA）参数优化的复合控制策略。该策略通过LADRC的扩张状态观测器（LESO）实时估计并补偿系统总扰动，同时利用GA的多目标优化特性自适应调节LADRC的参数，克服传统试凑法参数整定中存在的观测滞后与增益冲突。为验证控制性能，构建了基于MATLAB-AMESim平台的联合仿真模型。仿真结果表明，在阶跃响应和正弦跟踪仿真试验中，GA-LADRC的控制效果明显优于PID；且GA-LADRC在阶跃信号仿真分析中，相较于传统LADRC在超调量和跟踪误差上均降低了50%以上；在正弦信号跟踪中，GA-LADRC相较于传统LADRC相位滞后降低18%，精度提高18.3%，ISE降低34.8%。此外，在干扰条件下，GA-LADRC还表现出优越的抗干扰能力。研究结果表明，所提出的GA-LADRC方法稳定有效，可为拖拉机精耕作业提供技术支持。