基于关联规则联合收获机全论域作业速度自适应控制系统

发布时间：2018-07-07 14:39

本文选题：联合收获机 + 脱粒系统　；参考：《江苏大学》2016年博士论文

【摘要】：联合收获机在田间收获作业时,田间作物密度、作物含水率、甚至地形的变化都会影响其喂入量的变化,而喂入量的变化会造成割台螺旋输送器、输送槽和脱粒滚筒的转速发生变化,其中脱粒滚筒的转速变化又将直接影响脱粒滚筒的工作性能。因此,割台螺旋输送器、输送槽、脱粒滚筒的转速、前进速度与喂入量、谷物收获损失率之间就存在着某种关联性。分析联合收获机多源作业信息之间的关联性,开展基于关联规则联合收获机全论域作业速度自适应控制系统研究,对探索联合收获机作业速度自适应控制规律以及寻找新的智能控制算法,都具有重要的现实意义和科学研究价值。本文结合国家“863”计划和江苏省科技支撑计划等项目,综合运用模型分析、关联规则数据挖掘技术、动力学分析与建模、计算机仿真、嵌入式技术等技术与理论,开展基于关联规则联合收获机全论域作业速度自适应控制系统研究,主要工作包括:1、在联合收获机作业系统模型分析和作业参数关联规则挖掘的基础上,提取作业参数与喂入量、损失率之间的关联规则,获取各作业参数影响喂入量和损失率有价值的关联知识。通过分析联合收获机作业系统各主要工作部件的数学模型可知,联合收获机主要作业参数(割台螺旋输送器转速、输送槽转速、脱粒滚筒转速、前进速度)与喂入量、损失率之间存在着某种关联性;依据关联规则挖掘技术,对联合收获机作业参数数据样本进行关联规则数据挖掘,获取脱粒滚筒转速、割台螺旋输送器转速、输送槽转速等作业参数对喂入量和损失率有影响的关联规则知识,并根据知识的重要性和置信度,采用归一化的方法评估各作业参数与联合收获机的喂入量和损失率之间的关联规则权重因子。考虑到所得数据的不完整性,权衡各作业参数与喂入量、损失率之间的关联程度,设置脱粒滚筒转速、割台螺旋输送器转速、输送槽转速参数的权值区间为[0.4 0.6]、[0.3 0.5]和[0 0.3]。2、建立了联合收获机脱粒系统动力学模型,并以此为基础构建了作业速度普通控制系统仿真模型,再融合作业参数的关联知识构建了基于关联规则作业速度控制模型。针对已有滚筒功耗模型没有考虑其他工作部件的运动对滚筒转速变化造成影响的这一问题,以XG610型联合收获机为研究对象,通过运动机构的动力学分析,建立了脱粒系统动力学理论模型;构建了联合收获机作业速度普通控制模型,并进行仿真分析。从普通控制模型的脱粒滚筒转速、前进速度仿真曲线变化趋势可以看出,联合收获机在喂入量出现较大变化时,控制系统能够对脱粒滚筒转速、前进速度做出有效的调控,滚筒转速变化没有超出允许变化范围,说明建立脱粒系统动力学模型是合理可行的;同时在普通控制模型基础上,融合作业参数的关联知识构建了基于关联规则作业速度控制模型,并与普通控制模型进行仿真对比。对比结果显示在总体收获性能基本相同的情况下,基于关联规则作业速度控制模型的整体控制性能要好于普通控制模型,前者前进速度的最大相对变化幅度要比后者减小了1.50%,稳态相对变化幅度比后者减小了0.70%,系统调整时间也由后者约16s缩短成约11s,系统整体稳定性好于普通控制模型。3、建立了基于关联规则联合收获机全论域作业速度自适应控制模型,并与基于关联规则的控制模型和普通控制模型进行仿真对比。在基于关联规则联合收获机作业速度控制模型的基础上,从全论域角度出发内建立了一种基于关联规则作业速度自适应控制仿真模型;设计了全论域可调因子模糊控制器,建立了可调因子模糊整定规则,并对三种控制模型进行仿真对比。仿真显示在喂入量增加约15%时,基于关联规则的全论域作业速度自适应控制模型能够满足对作业速度的调控要求,脱粒滚筒转速相对额定值最大相对变化幅度约为5.48%,稳态时滚筒转速相对变化幅度约为2.62%;前进速度相对设定值最大相对变化幅度约为9.00%,稳态时相对变化幅度约为7.80%;系统调整时间大约为8s。喂入量和单位损失率稳态时大小分别为3.88kg/s和0.55%/(kg/s)。对比结果显示,基于关联规则联合收获机全论域作业速度自适应控制模型不仅在控制性能方面优于基于关联规则的控制模型和普通控制模型,而且在总体收获性能方面也好于基于关联规则的控制模型和普通控制模型。4、对联合收获机作业速度控制的硬件系统组成和软件系统开发进行研究,并对控制系统进行室内测试。硬件系统主要由arm9系统、转速信号采集模块、液晶触摸显示屏和联合收获机作业速度自动调控装置等部分组成,同时系统预留了视频监测模块和gps信号采集模块的接口;开发外接硬件设备驱动程序和作业速度控制系统应用软件,应用软件共分为五个部分:系统主界面、参数设定界面、作业速度监测与智能控制界面、视频监测界面和gps定位信息监测界面。在联合收获机室内模拟调速装置上对系统进行了测试。测试结果显示,系统对监测的数据实时稳定,当分别采用普通控制算法、基于关联规则控制算法和基于关联规则全论域自适应控制算法对步进电机的控制符合联合收获机前进速度的调节要求。5、进行作业速度控制系统机载调试,并针对三种控制模型的控制算法开展田间收割试验与对比验证。机载调试主要开展各工作部件转速的标定、前进速度的标定以及自动控制作业测试等工作。分别采用普通控制算法、基于关联规则的控制算法和基于关联规则全论域自适应控制算法进行水稻收割试验,并进行试验数据分析和对比。从三种控制算法的脱粒滚筒转速和前进速度试验数据曲线总体变化趋势上可以看出,滚筒转速与前进速度的变化与仿真分析结果相符合,三种控制算法的脱粒滚筒转速最大变化幅度没有超出额定值7%的允许变化范围,这也进一步验证了所建立的脱粒系统动力学模型是合理可行的。同时对比结果显示,在控制性能方面,基于关联规则全论域自适应控制算法获得的脱粒滚筒转速稳态时平均变化幅度为2.97%、前进速度最大变化幅度9.00%、到达基本稳定状态所需时间约7s,均优于普通控制算法和基于关联规则控制算法获得的控制性能数据;在收获性能方面,基于关联规则的全论域自适应控制算法的联合收获机平均喂入量比基于关联规则的控制算法和普通控制算法下的平均喂入量要略小,但该算法下的平均损失率要比后两种控制算法下的平均损失率分别要低0.29%和0.22%,单位平均损失率要比后两种控制算法分别要低0.06%/(kg/s)和0.05%/(kg/s),损失率降低幅度较大。因此,基于关联规则全论域作业速度自适应控制系统不仅在控制性能方面优于后两种控制系统,而且在总体收获性能方面也好于后两种控制系统。
[Abstract]:When the harvester is harvested in the field, the field crop density, the water content of the crop, and even the change of the terrain will affect the change of the feeding amount, and the change of feed quantity will cause the screw conveyor of the cutting platform, the speed of the conveying slot and the threshing roller change, and the change of the speed of the threshing barrel will directly affect the work of the threshing roller. Therefore, there is some correlation between the rotating speed of the screw conveyor, the conveying tank and the threshing roller, the speed of advance and the feed rate, the loss rate of the grain harvest, and the correlation between the multi source work information of the combined harvester and the study of the adaptive control system of the full domain operation speed based on the association rules and the joint harvesting machine. It is of great practical significance and scientific research value to explore the adaptive control law of the working speed of the joint harvester and to find a new intelligent control algorithm. This paper combines the national "863" plan and the Jiangsu science and technology support program, and comprehensively uses the model analysis, the closed rule data mining technology, the dynamic analysis and modeling. On the basis of computer simulation, embedded technology and other technologies and theories, the research on adaptive control system of full domain operation speed based on association rules is carried out. The main work includes: 1. On the basis of the model analysis of the joint harvester operating system and the mining of the association rules of the operating parameters, the extraction of the operation parameters and the feed rate and the loss rate are extracted. Association rules are used to obtain relevant knowledge about the value of feed volume and loss rate. Through the analysis of the mathematical models of the main working parts of the combined harvester, the main operating parameters of the combined harvester (rotating speed of the screw conveyor, the speed of the conveyor, the speed of the threshing roller, the speed of advance) and the loss of the feed are found. According to the association rule mining technology, the association rule data mining is carried out on the data samples of the joint harvester operation parameter data to obtain the knowledge of the association rules which have influence on the feeding quantity and loss rate, and the knowledge is weighed according to the knowledge, and the speed of the threshing roller, the speed of the screw conveyor and the speed of the conveyor, and so on. To evaluate the correlation rule weight factor between the operating parameters and the feed intake and loss rate of the combined harvester. Considering the incompleteness of the data, the degree of association between the operating parameters and the feed volume and the loss rate is weighed, and the rotational speed of the threshing roller, the speed of the screw conveyor of the cutting platform, and the transmission of the screw conveyor are taken into account. The weight range of the rotating speed parameters of the grooves is [0.4 0.6], [0.3 0.5] and [0 0.3].2. The dynamic model of the threshing system of the joint harvester is set up. On the basis of this, the simulation model of the operation speed ordinary control system is built. Then the association knowledge of operation parameters is used to construct the operation speed control model based on the association rules. The consumption model does not take into account the problem that the movement of other working components affects the change of the rotational speed of the drum. Taking the XG610 type combined harvester as the research object, the dynamic theoretical model of the threshing system is established through the dynamic analysis of the motion mechanism, and the general control model of the combined harvester is constructed, and the simulation analysis is carried out. It can be seen that the control system can effectively control the speed and speed of the threshing drum when the feed volume of the combined harvester changes greatly, and the change of the rotational speed of the drum does not exceed the allowable range of change, indicating the establishment of the dynamic model of the threshing system. It is reasonable and feasible. At the same time, based on the common control model, the operation speed control model based on association rules is constructed and compared with the common control model. The comparison results show that the speed control model based on the association rule operation speed control model is the same as the overall harvest performance is basically the same. The body control performance is better than the ordinary control model, the maximum relative change range of the former speed is 1.50% less than the latter, the relative change amplitude of the steady state is 0.70% less than the latter, and the adjustment time of the system is shortened by about 16S by about 11S, and the overall stability of the system is better than the universal control model.3, and the association rules are set up based on the association rules. The adaptive control model of the machine full domain operation speed is simulated and compared with the control model based on the association rules and the common control model. Based on the association rule combined with the harvest speed control model of the harvester, an adaptive control simulation model based on the operation speed of association rules is established from the point of view of the whole domain. The whole domain adjustable factor fuzzy controller is designed, the adjustable factor fuzzy setting rule is established, and the simulation comparison of the three control models is carried out. The simulation shows that when the feeding amount is increased by about 15%, the adaptive control model of the whole domain operation speed adaptive control model based on the association rules can be full of the operation speed regulation and the threshing roller speed. The relative variation amplitude of the relative nominal value is about 5.48%, the relative change amplitude of the roller speed is about 2.62% in the steady state, the maximum relative change amplitude of the forward velocity relative to the set value is about 9%, the relative change amplitude is about 7.80% in the steady state, and the system adjustment time is about 3.88kg/s and 0.5, respectively, when the 8s. feed and unit loss rate are steady. 5%/ (kg/s). The comparison results show that the adaptive control model based on the association rule combined harvester total domain job speed adaptive control is not only better than the control model and ordinary control model based on association rules in control performance, but also better than the control model based on association rules and the common control model.4 in the overall harvest performance. The hardware system and software system development of the harvester operating speed control are studied and the control system is tested indoors. The hardware system is mainly composed of ARM9 system, speed signal acquisition module, liquid crystal touch screen and joint harvester operation speed automatic control device and so on. Meanwhile, the system is reserved for video monitoring. The interface of module and GPS signal acquisition module; development of external hardware device driver and operation speed control system application software. The application software is divided into five parts: system main interface, parameter setting interface, operation speed monitoring and intelligent control interface, video monitoring interface and GPS positioning information monitoring interface. The test results show that the system is stable to the monitoring data in real time. When the common control algorithm is adopted, the control of the stepping motor based on the association rule control algorithm and the full domain adaptive control algorithm based on the association rules conforms to the adjustment requirement of the forward speed of the combined harvester.5. The operation speed control system is debugged, and the field harvest test and contrast verification are carried out against the control algorithms of three control models. The airborne debug mainly carries out the calibration of the rotational speed of the working parts, the calibration of the forward speed and the automatic control operation test. The general control algorithm and the control algorithm based on the association rules are adopted respectively. The rice harvesting experiment based on the full domain adaptive control algorithm based on the association rules is carried out, and the experimental data are analyzed and compared. The change trend of the speed and forward speed of the three control algorithms can be seen that the change of the roller speed and the forward speed conforms to the results of the simulation analysis, and the three kinds of control are controlled. The maximum change range of the rotational speed of the threshing cylinder is not beyond the allowable range of the rated value of 7%, which further proves that the dynamic model of the threshing system is reasonable and feasible. At the same time, the comparison results show that the rotational speed of the threshing drum is obtained on the basis of the correlation rule full domain adaptive control algorithm in the control performance. The average change amplitude of the steady state is 2.97% and the maximum change range of the forward speed is 9%. The time required to reach the basic stable state is about 7S, which is better than the control performance data obtained by the common control algorithm and the association rule control algorithm. The feeding amount is slightly smaller than that of the control algorithm based on the association rules and the average control algorithm under the common control algorithm, but the average loss rate under this algorithm is 0.29% and 0.22% lower than the average loss rate under the two control algorithms. The unit average loss rate is lower 0.06%/ (kg/s) and 0.05%/ (kg/s) than the latter two control algorithms, and the loss rate is lower than the latter. The rate of rate reduction is larger. Therefore, the adaptive control system based on the association rule full domain operation speed adaptive control is superior to the latter two control systems, and the overall harvest performance is better than the latter two control systems.
【学位授予单位】：江苏大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：S225

【相似文献】