基于Internet温室环境远程智能控制系统研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
温室设施依靠检测和控制技术为作物创造良好的生长环境,但是温室内高温、潮湿的恶劣作业环境,使人不能在其中长期工作。世界各国由于温室面积的扩大,如何进行温室的群管理,降低运行成本,提高效率,实现环境的精确控制已经成为目前研究的焦点问题。因此远程控制技术和自动控制技术的提高成为设施农业发展的迫切需求。
    鉴于此,本文进行了温室环境远程智能控制系统的开发研究,该系统在结构上主要由7大部分组成:远程控制器、Internet网络、现场监视器、智能控制软件、模数转换卡、传感器模块、输出及驱动装置。在软件设计上主要包括两部分:远程控制子系统的设计、智能控制子系统的设计。
    温室环境远程智能控制系统基于Internet网络实现远程控制信息的传输。Internet是一个跨越全球的计算机网络,通过Internet可以在全球范围内将信息快速、有效和方便地传递。所进行的基于Internet远程控制程序的编写主要是使用Winsock。Winsock是一套开放的、支持多种协议的Windows下网络编程接口,是Windows网络编程事实上的标准。应用程序通过调用Winsock的API实现相互之间的通信,而Winsock利用下层的网络通信协议功能和操作系统调用实现实际的通信工作。
    该远程智能控制系统的主要功能是在网络上由一台计算机(远程控制器)通过对另一台计算机(现场监控器)远距离地控制以此来实现对温室环境进行控制。当操作者使用远程控制器控制温室内环境参数时,就如同在现场监控器的屏幕前一样,可以进行很多方面的远程控制,包括获取目标计算机屏幕图像、窗口及进
    
    
    程列表;记录并提取远程键盘事件(击键序列,即监视远端键盘输入的内容);打开、关闭目标计算机的任意目录并实现资源共享;提取拨号网络及普通程序的密码;激活、中止远端进程;打开、关闭、移动远端窗口;控制目标计算机鼠标的移动与动作(操作);浏览目标计算机文件目录,任意删除目标计算机的磁盘文件;上传、下载文件,就如操作自己的计算机的文件一样的简单;远程执行目标计算机的程序;强制关闭Windows、关闭系统(包括电源)、重新启动系统;提取、创建、修改、删除目标计算机系统注册表关键字;在远端屏幕上显示消息;启动目标计算机外设进行捕获、播放多媒体/音频文件;控制远端录音、放音设备音量以及进行远程版本升级更新等。
    温室环境智能控制部分主要涉及对温室内温度、湿度、光照、CO2浓度等的多变量控制,各控制变量之间相互制约、相互影响、相互耦合,并且存在滞后性。因此,温室控制系统是一个典型的多变量非线性时变分布参数系统,要建立该系统的精确数学模型是非常困难的。采用传统的控制方法难以实现精确控制的目的,模糊控制不需要建立精确的数学模型,便可根据经验规则对非线性复杂系统进行理想控制。但是,模糊控制规则的提取和隶属度函数的产生,需要工程人员有丰富的经验。神经网络的学习能力恰恰能够解决模糊控制的这一弊端。将模糊系统和神经网络结合在一起,可以发挥两者的长处,提高整个系统的学习能力和表达能力。针对温室控制这一特点,采用模糊神经网络控制技术,实现温室环境的智能控制部分的开发。
    本文在充分总结和吸收国内外关于远程控制技术及智能控制技术的理论和实践经验的基础上,结合本课题的特点和要求,对
    
    
    远程智能控制技术在温室中的应用进行了研究,以LabVIEW 6i为主要开发工具结合MATLAB 6.5及 Visual C++6.0开发了基于Internet的温室环境远程智能控制系统,并用该系统对温室环境进行了检测和控制,取得了良好效果,具有一定的实用价值。主要研究工作和结论如下:
    1.在LabVIEW 6i操作平台上结合MATLAB 6.5编程语言开发温室环境智能控制子系统。该系统运用LabVIEW 6i编写软件的操作界面,运用MATLAB 6.5进行模糊神经网络的建立,利用神经网络的参数和结构的训练方法,自组织地获得隶属函数和规则库,并采用遗传算法对其进行训练。
    2.在LabVIEW 6i操作平台上结合Visual C++6.0编程语言开发基于Internet网络的温室环境远程控制系统。该系统由远程控制器、信息传递媒介、现场监控器、数据采集卡、各类传感器及执行机构六大部分组成。操作者在现场监控器端计算机上直接对温室环境参数进行调整控制,同时也可以实现在远程控制器电脑上通过身份验证后直接登录现场监控器,在两台电脑间传输文件、传递信息及远程关机等控制。温室管理人员还可以远程观测温室内状况。
    3.将远程控制子系统及智能控制子系统结合并将所有的系统硬件连接起来,构建一个完整的温室环境远程智能控制系统。并在吉林大学生物与农业工程学院的玻璃温室内进行试验运行,重点测试了智能控制效果、网络传输速度、网络延迟、传输正确率以及传输不稳定性,得出如下结论:由于考虑到减小模糊神经网络控制的复杂性,本系统主要对温室环境参数中的温度和湿度进行研究,但是其他因素,如光照、CO2等因素对温室环境的影响也较
    
    
    重要,因此在智能控制上有一定的误差;远程控制方面,在非直接控屏状态下,如远程执行命令、传输文件以及传送信息等情况下,能够较好的执行命令,无明显延迟。而在直接控屏状态下,网络传
The greenhouse system supplies fit environment to crops depend on detection and control technique, but it is unwholesome for people to work in the atrocious weather of greenhouse. And how to manage teeming greenhouses, how to reduce the run cost, how to improve the efficiency, how to control greenhouse’s environment accurately, all these problems have been the pivotal matter. So the technique of remote control and auto control in greenhouse environment has become important to development of facility agriculture.
    For these reasons and purposes, study the system of remote intelligent control in greenhouse’s environment. The system includes seven parts in structure: remote-director、internet、monitor on the spot、software、DAQ card、sensors、equipments. And the design of software include two parts: remote control system、intelligent control system.
    The remote-intelligent control system in greenhouse transmits the information via internet. Internet is a computer web spread all over the world, it can transmit the information quickly、efficiently、conveniently. The remote control program is compiled used Winsock. Winsock is an interface of program in web. The application transmit information through call the API of Winsock.
    The main function of the remote-intelligent control system in greenhouse via internet is to control greenhouse environment through one computer (Host/Server) controlled by another remote computer (Remote/Client) via internet. While operator uses the client computer to control the server computer, just like he sitting before the monitor of
    
    
    the server computer. He can startup any application in the server, edit documents, and even working or surfing Internet with the peripheral equipment(printer、scanner…) and communication equipment(modem or other net equipment) of the server, as much as remote control the TV volume、change channel or on-off TV set with telecontroler. But then we must define a notion: the client computer just transmit the instruction of keyboard and mouse to the server, and return the screen information on the server. That is to say, it seems that we operate the client, and the operation come to realization on the server essentially, whatever open a file、browse web page or download. All information and IE cookies are stored in the server.
    The control of greenhouse’s environment include the indoor environment parameters such as temperature、humidity、CO2 concentration and light intensity. And these parameters are interactional、inter-restraining、inter-coupling. So the system is a nonlinear, large lagging, multiple input and output object. It is hard to found the precise model and difficult to achieve a satisfactory result in conventional control methods. It is not necessary to found precise module of the greenhouse used fuzzy control technology, and it is ideal to control the multiple nonlinear system based on experience. Although fuzzy control technology can ratiocinate, it needs veteran to make the rule and membership function. The learning ability of neural network can solve the problem for fuzzy control. The combine of fuzzy control technology and neural network is fuzzy-neural network control, it infers fuzzy logic using neural network configuration, and
    
    
    gains the rule and membership function independent.
    On the basis of summarizing and absorbing the result and the experiences in the study of the theory and the programming about the technique of remote control and intelligent control, combined with the characteristic and demand of the project, the author studied on the application of remote-intelligent control technique in greenhouse. And the writer designed the remote control system via Internet in the greenhouse programming with LabVIEW 6i、 MATLAB 6.5 and Visual C++6.0, and simulated the work process of the check and control of the greenhouse environment. The outcome is impressive and practical. The focus points of the main work in this paper are as follows:
    Designed the intelligent control system in greenhouse based on LabVIEW 6i and MATLAB 6.5. C
引文
[1]汪建,张世禄.Internet远程控制系统.计算机应用,2000.12:62-65.
    [2] H. Chang, C. Tait, N. Cohen, R. Floyd, B. Housel& D. Lindquist 1997. Web Browsing in a wireless Environment: Disconnected and Asynchronous Operation in ARTour Web Express. In MOBICOM 97,1997,260-269.
    [3]谢建玲.Internet---世界最大的互联网.导弹与航天运载技术,1995.5:77-79.
    [4] 于海业,马成林,陈晓光.发达国家温室设施自动化研究的现状.农业工程学报,1997.13(增):253-257.
    [5] http://www.s-soren.or.jp/99_11_09denpob-op.htm.
    [6] http://www.radiosystems.co.uk/.
    [7] http://www.loukfarm.gr/greenhouse_en.asp.
    [8] http://www.x-10europe.com/greenhouse.htm.
    [9] http://www.greenair.com/ghc2man.htm.
    [10]于海业,马成林等.远程控制技术在温室环境控制中的应用现状分析.农业机械学报,2003.11:160-163.
    [11] 乔晓军. 温室生态系统健康研究Ⅰ—环境监控系统开发和瓜蚜种群数量动态分析. 北京:中国农业大学博士学位论文,2000.4.
    [12]佟成安,高英.Internet简介.现代情报,1995.12:27-28.
    [13]付影平,刘进海.Internet通信协议与接入方法的探讨.西安邮电学院学报,2000.3:29-32.
    
    [14] Tim Parker,Mark Sportack. TCP/IP技术大全.北京:机械工业出版社,2000.7.
    [15] Uyless Black. TCP/IP and Related Protocols[M]. McGraw-Hill,1992.
    [16] 陈坚,陈伟.Visual C++网络高级编程.北京:人民邮电出版社,2001.8
    [17] Richard C. Leinecker. Visual C++5.0开发技术内幕.北京:机械工业出版社,1999.
    [18] 于海业,马成林,孙瑞东.温室环境自动检测系统.农业工程学报,1997.13(增):262-263.
    [19] 吴静怡,王如竹.现代农业技术——太阳能温室设施农业,太阳能,1998.4
    [20]刘庆玉.温室环境自动控制系统的研究.农村能源,1998.2:14-17.
    [21]郑秀莲.现代温室气候的专家控制系统.机电工程,2003.20(3):42-45.
    [22]陶然.智能化温室环境控制系统的研究.农机化研究,2003.2:53-55.
    [23]周晨松.智能温室控制系统:[硕士学位论文].哈尔滨:东北林业大学,2001.
    [24] 穆阳,朱伟兴.用VC++开发温室群集散控制系统通信软件.微计算机应用,2000.21(2):81-83.
    [25] Alves-Serodio C M J, Moteiro J L, Couto C A C. An integrated network for agriculture management applications. IEEE International Symposium on Industrial
    
    
    Electronics Proceedings. ISIE’98 p.679~683 vol.2, Publisher: IEE.
    [26] 顾寄南,毛罕平.温室环境智能化控制模型的研究.农业机械学报,2001.32(6):63-65.
    [27] 顾寄南,毛罕平.“大系统”理论及其在温室系统建模中的应用.农业系统科学与综合研究,1999.3(15).
    [28] 宫赤坤.温室环境多变量模糊控制及其仿真.农业机械学报,2000.31(6):52-54.
    [29] 李友善,李军.模糊控制理论及其在过程控制中的应用.北京:国防工业出版社,1993.
    [30] 刘君华.基于LabVIEW的虚拟仪器设计.北京:电子工业出版社,2003:2.
    [31]张袆,韩慧莲.虚拟仪器在远程控制中的软件设计与实现.测试技术学报,2003,17(2):142-145.
    [32]Richard C. Leinecker. Visual C++5.0开发技术内幕.北京:机械工业出版社,1999.
    [33]网冠科技.Visual C++6.0时尚编程百例.北京:机械工业出版社,2001.1.
    [34]陈怀琛.MATLAB及其在理工课程中的应用指南.西安:西安电子科技大学出版社,1999.10:3.
    [35] 闻新.MATLAB神经网络应用设计.北京:科学出版社,2000:2.
    [36] 尹洪珠,马孝江. 基于Internet的远程故障诊断技术中的交互性问题[J]. 计算机测量与控制,2002,10(3):145-147
    [37] 熊焕庭.在LabVIEW中数据采集卡的三种驱动方法.电测与仪
    
    
    表,2001,8:35-37.
    [38]龙志强. LabVIEW与通用数据采集卡的接口方法研究.微计算机信息,2001,17(9):18-19.
    [39] 杨乐平,李海涛等.LabVIEW高级程序设计。北京:清华大学出版社,2003.4:437-446.
    [40] Tencate. A thermal model for greenhouse. 1989, ASAE paper 89~4055.
    [41] Wang.L.X, and Mendel.J.M., Fuzzy basis functions, universal approximation and orthogonal least squares learning[J].IEEE Transon Neural networks, Sept. 1992, 3(5): 807-814.
    [42] 陈延其,路界,神经网络理论及其在控制工程中的应用,西北工业大学出版社,1991.
    [43] Wang.L.X,Stable adaptive fuzzy control of nonlinear systems[J].IEEE Trans Fuzzy Syst,1993,1(2):146-156.
    [44]Chen Y C. Teng C C. A Model Reference Control Structure Using a Fuzzy Neural Network. Fuzzy Sets and Systems. 1995,73:291-312.
    [45] Jang JR. Self-learning Fuzzy Controller Based on Temporal Back Propagation. IEEE Trans Neural Network, 1992,NN-3(5):714-723.
    [46] Kong S G. Kosko B. Adaptive Fuzzy Systems for Backing up a Track-and-Trailer. IEEE Trans Neural Networks, 1992, NN-3(2):211-233.
    [47] 王耀南,宋明,童调声,一类神经网络自组织模糊控制器与
    
    
    应用,模式识别与人工智能,1994,7.
    [48] 杨煜普,许晓鸣,张钟俊.基于模糊神经网络的控制规则获取及置信度估计问题. 模式识别与人工智能,1994,7(1):53~593.
    [49] Beyer H G, Deb K. On Self-adaptive Features in Real-parameter Evolutionary Algorithms[J]. IEEE Transactions on Evolutionary Computation, 2001, 6(3): 250-270.
    [50] Kita H.A Comparison Study of Self-Adaptation in Evolution Strategies and Real-coded Genetic Algorithms[J].Evolutionary Computation, 2001, 9(2):223-241. 
    [51] Kita H, Ono I, Kobayashi S. MULTI_Parental Extension of the Unimodal Normal Distribution Crossover for Real-Coded Genetic Algorithms[A]. In: Angeline P Jed. Proceedings of the Congress on Evolutionary Computation[C]. Piscataway, NJ:IEEE Press, 1999,1581-1587.
    [52]周志坚.一种基于遗传算法的模糊神经网络结构和参数优化.华南理工大学学报(自然科学版),1999,27(1):26-32.
    [53] Joseph Baum. PC-Based environmental control system for plant and animal structures. St Joseph,Mich:1987,ASAE Paper NO.87 4551.
    [54] Korthals R L, Christianson L L, Knight S. Greenhouse modeling for environment control. 1989,ASAE paper 89 4014.

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700