基于鲁棒反馈线性化的联合收获机割台高度控制策略

谷物联合收获机割台高度控制非常重要，有效的割台高度控制有助于提高喂入量的稳定性、降低整机各环节的负荷波动。本文提出一种基于鲁棒反馈线性化的割台高度控制策略，该方法可以使割台跟随地面起伏进行俯仰控制调节。首先，基于割台结构和动力学分析建立系统数学模型，选取正弦角度的近似约简条件，将多变量的复杂非线性系统线性转换为典型的非线性系统；通过分析传统的反馈线性化控制研究可控仿射的模型构建方法，在集成鲁棒优化设计控制器基础上，提出鲁棒反馈线性化（Robust feedback linearization，RFL），通过构建灵敏度方程、选取增益来稳定系统；选取液压控制机构，根据控制液压输出的电流参数设计为基于鲁棒反馈线性化控制系统的控制器。将传统的PID控制和本文提出的鲁棒反馈线性化控制进行对比实验，结果表明，在不同行驶速度、地形正弦振幅和地形周期条件下，鲁棒反馈线性化控制下的高度误差均小于传统的PID控制；以3种不同行驶速度在同一起伏地面上行进，鲁棒反馈线性化控制下的高度误差受行驶速度增加的影响小于传统的PID控制。

Header Height Control Strategy of Harvester Based on Robust Feedback Linearization

In the operation of grain combine harvester, the height control of header is particularly important. Effective header height control helps to improve the stability of feeding amount and reduce the working load fluctuation of each link of the whole machine. A kind of header height control strategy based on robust feedback linearization was introduced, which made the header follow the ground fluctuation to adjust the pitch control, so as to achieve the purpose of consistent stubble height. Firstly, the mathematical model of the system was established based on the analysis of the structure and dynamics of the header, and the approximate condition of sinusoidal angle reduction was selected to linearize the multi-variable complex nonlinear system into a typical nonlinear system. The traditional feedback linearization control was analyzed, and the controllable affine model building method was studied. The reason that it lacked of robustness due to its high dependence on the identification accuracy of the system was obtained. Furthermore, based on the integrated robust optimal design (IROD) controller, the robust feedback linearization (RFL) was proposed, which stabilized the system by constructing the sensitivity equation and selecting the gain;furthermore, the robust feedback linearization was proposed according to the current parameters of hydraulic output, the controller was designed for the output of robust feedback linearization control system. In the contrast simulation experiment, the traditional PID control was compared with the proposed robust feedback linearization control. The height error under the robust feedback linearization control was smaller than that of the traditional PID control under the conditions of different driving speeds, sinusoidal amplitude and terrain period;and the robust feedback linearization was realized under the measured conditions of three different driving speeds on rolling ground. Therefore, compared with the traditional PID control, the robust feedback linearization controller had higher steady-state accuracy and improved the consistency of stubble.