基于模糊PID控制的再生稻自适应仿形割台性能试验与分析

为促进水稻再生季穗头萌发，头季稻机械收获时需保证300～450 mm的留茬高度，但水田泥脚深浅不一，收获时割台高度上下浮动，留茬高度难以保证，严重影响再生季产量。为此，该研究设计了一种自适应仿形割台，可实现割台高度及水平自适应调整。首先，对自适应仿形割台结构进行设计。然后，在Matlab/Simulink中搭建控制系统仿真模型，设计模糊规则，采用模糊PID控制方法对自适应仿形割台进行性能仿真，以超调量、响应时间和稳定性为指标，验证控制方法的可行性；以阶跃信号作为激励，对比分析了传统PID和模糊PID的控制效果，结果表明，模糊PID控制比传统PID的上升时间和到达稳态所需时间分别减少78.9%和81.6%，超调量由46 mm下降到8.2 mm。最后，搭建割台试验平台进行性能试验。结果表明，采用模糊PID控制时割台提升过程的平均响应速度约为0.216 m/s，下降过程的平均响应速度约为0.244 m/s，割台高程调节的平均误差6.75 mm，水平调节的平均误差为0.64°，割台调整迅速，定位准确度高，可满足再生稻头季收获使用需求。

Performance test and analysis of the self-adaptive profiling header for ratooning rice based on fuzzy PID control

Abstract: Ratooning rice has been one of the rice planting modes for a long history in China, dating back to 1700 years ago. The rice bud can grow again, after the rice harvest in the first season. As such, the ratooning rice can be harvested twice at a time. Planting areas of the ratooning rice can be required for the amiable temperature, light and water conditions for the stable and high yield. The rice panicles are collected in the harvesting process, where 1/3 of the plant and root system are left for the tiller growth again. A high stubble height remains during mechanical harvest in the first season of ratooning rice, in order to promote the panicle germination in the regeneration season. There are different depths of soil feet in the field during the operation of ratooning rice harvester, where the header height of the harvester fluctuates up and down. The driver needs to adjust the header height in real time to ensure the stubble height. The uneven stubble height has posed a great threat to the operation efficiency and tiller growth in the regeneration season. The height and level adaptive adjustment function can be a high demand for the header of the ratooning rice harvester for the consistent stubble height of the field. It is necessary to automatically adjust the height and level in the header, according to the ground fluctuation. In this study, an adaptive profiling header was developed to realize the adaptive adjustment of height and level. Firstly, the specific structure was designed to include the header support frame, conveying groove, rotating disc, elevation oil cylinder, leveling oil cylinder, header, and profiling ground wheel. Among them, the header was rotated around the conveying groove at the given angle. Secondly, a simulation model of the control system was established to optimize the fuzzy PID controller using the Matlab/Simulink platform. The performance of the adaptive profiling header was then evaluated to verify the feasibility of the fuzzy PID control, in terms of the overshoot, response time, and stability index. Taking the step signal as the excitation, the better control of fuzzy PID was achieved, where the rise time and the time required to reach the steady state were reduced by 78.9%, and 81.6% than before, indicating almost no overshoot in the system. The simulation results show that the fuzzy PID control system presented a better performance in the response speed and overshoot. Finally, the test platform of the header was built to conduct the header adjustment accuracy test, header response speed test, and overall system simulation. The bench test results show that the average response speed of header lifting, falling, leveling to the left, and leveling to the right were about 0.216 m/s, 0.244 m/s, 2.38°/s and 3.22°/s, respectively, using the fuzzy PID control system. The header height was measured to obtain the data output using the angle sensor system, and then to accurately represent the height between the header and the ground in real time, when the traveling speed of the soil tank trolley was 1 m/s. Consequently, the header height always maintained the set value during the movement, with the average and maximum errors of 6.75 and 11.63 mm, respectively, indicating the better control effect of header height. The output angle of the inclination sensor was measured for the inclination angle of the undulating road in the field. The header remained horizontal during the movement, with the average and maximum errors of 0.64° and 2.35°, respectively. Anyway, the control parameters of PID control were better than that of the traditional. The header was also adjusted quickly under the stable state. There was a relatively small error between the experimental and the simulation, indicating the better fitting degree. Therefore, the fuzzy PID control system can fully meet the adjustment requirements of the header.