适于丘陵地区的甘蔗收获机入土切割深度自动控制系统研究

At present, the buried cutting depth of the sugarcane harvester suitable for the hilly lands cann't be automatically adjusted along with the furrows and ridge surfaces of the rugged topography so that the cane stump may be long or short, which will affect the broken head rate and harvesting losses rate. The project researches the automatic control system of the buried cutting depth of the sugarcane harvester, and studies the control accuracy and response speed. It aims to solve the detection of complex information characteristic based on the neural network, fuzzy reasoning and genetic algorithm of the multi-sensors fusion signal processing. The knowledge base system is established to describe the relation between the loading pressure and the buried cutting depth of the cutter. It also studies the control theory of the hydraulic system of the automatic adjustment of the cutting depth. The intersectant application of such manners as intelligent computing and simulation analysis is used to develop the intelligent optimal control system of the buried cutting depth. The control system is analyzed on the simulation software to research the automatic adjustment function under different conditions. According to the simulation results, the control system is optimized. Finally, the project develops the experiment platform of the control system to verify the correctness of the theory and simulation model. It also be applied to the prototype of the sugarcane harvester to carry out the performance test. In innovation view, the project presents that the integrated modeling method based on the intelligent computing and mathematical model is used to build the relation forecast model between the loading pressure and the buried cutting depth. According to the forecast model the closed-loop optimal control of the automatic control system is realized.

项目针对丘陵地区甘蔗收获机切刀无法随着田地垄沟和垄面的起伏而自动调整，从而造成蔗蔸有长有短，影响宿根破头率和收割损失率的问题，系统研究切刀入土切割深度自动控制系统，探讨系统的控制精度、响应速度等特征。通过神经网络、模糊推理和遗传算法等技术的多传感器信息融合处理方法, 解决复杂信息特征的检测问题，构建描述切刀负载压力和入土切割深度关系的知识库系统；研究入土切割深度自动调节的液压系统控制原理，交叉应用智能计算、仿真分析等手段，研究开发切刀入土切割深度智能优化控制系统；模拟与再现控制系统在不同工况下的自动调节功能，优化系统设计及控制算法。研制入土切割深度自动控制系统试验平台，验证理论和仿真模型的正确性，并运用到自主研发的样机上进行田间试验。本项目在创新层面上，提出了结合智能计算、数学模型等的集成建模方法建立切刀负载压力与入土切割深度关系的预测模型，依据此模型实现对自动控制系统的闭环优化控制。

由于丘陵地区田地高低起伏变化较大，甘蔗田地垄沟和垄面很不平整，如果甘蔗收获机切割器切刀入土切割深度不能随着垄面和垄沟的起伏而自动调整，势必会造成蔗蔸有长有短，影响宿根破头率和收割损失率。因此本项目通过研究广西丘陵地区不同土壤及甘蔗品种特性的基础上，深入探讨了甘蔗收获机刀盘负载压力与入土切割深度之间的关系，找出了影响切割负载压力的关键影响因素。并通过采用数值仿真分析与物理试验研究相结合的方法，分别在计算机及试验平台上再现了刀盘负载压力与入土切割深度的关系模型。在此基础上，构建刀盘升降自动控制系统，通过仿真分析与试验研究方法验证了其可行性。.（1）切刀负载压力与入土切割深度的相互关系研究。根据甘蔗生长的复杂环境，通过大量的实地土壤取样进行试验，构建了土壤剪切强度、土壤硬度等数学模型，为室内真实复现土壤特性提供了基础数据。通过大量的剪切试验、刀盘负载压力与入土切割深度的试验研究，并结合神经网络、遗传算法、孪生支持向量机等技术构建了不同的工况下切刀负载压力与入土切割深度的相互关系数学模型，从而为自动控制系统提供有力的支持。.（2）采用数值模拟方法对甘蔗收获机入土切割深度自动控制系统理论模型进行建模和仿真分析，设计了以PLC为控制核心，以切割负载压力为反馈信号的小型甘蔗收割机切割深度自动控制系统。运用AMESIM仿真软件建立了小型甘蔗收割机切割系统的液压仿真模型并进行仿真分析，同时在自主研制的甘蔗茎秆切割实验平台对自动控制系统进行试验研究。.（3）考虑到理论与实践的差异和自动控制系统的实用性，研制了甘蔗收获机入土切割深度自动控制系统及其试验平台。以甘蔗根部切割质量为考察目标，测试控制系统对土壤切割深度的自动测量、控制精度、响应速度等特征，从而验证了理论和仿真模型的可行性。