基于模糊控制策略的温室远程智能控制系统的研究

发布时间：2018-03-18 08:16

本文选题：远程控制　切入点：遗传算法　出处：《吉林大学》2015年博士论文　论文类型：学位论文

【摘要】：近些年来，随着计算机技术、自动化技术和网络技术的飞速发展，智能化和网络化也成为温室环境控制的发展方向，对温室作物的栽培和管理也开始注重节约能源及其可持续发展性。吉林地区冬季漫长而寒冷，夏季短暂而温热，昼夜温差大，其基础温度及温差变化同其他地区相比有很大的不同。现有的控制系统是已经做好的固定的控制系统，不能随用户的要求和季节的变化而随意更改。本研究针对吉林省地处北方寒冷地区的特点，通过对温室作物生长规律及控制算法、软硬件设计等方面的研究，设计了一套带有远程控制功能的温室智能控制系统，通过调节温室的温度、湿度、光照度和二氧化碳浓度等，为温室作物的生长提供适宜的生长环境，从而提高作物的品质、产量并节约能源。本研究还设计了温室控制系统的模糊PID控制器，并用改进了的遗传算法对两输入、三输出的模糊PID控制器进行了三角形隶属函数底宽和模糊控制规则的优化。对优化后的模糊PID控制器进行了仿真实验，并与常规PID控制器和模糊PID控制器进行了比较。常规PID控制器对被控对象的数学模型依赖很大，存在响应速度慢，超调量大等问题。运用模糊推理对PID控制器的三个参数Kp、 Ki、 Kd进行调整，其控制效果明显优于传统的PID控制器，提高了系统的动静态性能，系统的超调量减少，响应速度变快。但是由于模糊PID控制器受到论域的划分均匀和专家经验合理性的影响，其控制效果也没达到最佳。而基于改进的遗传算法优化的模糊PID控制器，通过遗传算法所具有的超强的全局搜索能力，可以得到理想的控制规则和合理的模糊划分。系统的超调量很小，响应速度也很快。将遗传算法应用于模糊PID控制器，对其进行参数的优化可以明显提高控制器的控制品质。仿真实验结果表明经过改进的遗传算法优化的模糊PID控制器具有较好的控制效果。本研究基于模糊控制设计了一套温室远程智能控制系统，本控制系统由四部分组成：一台作为客户端的远程控制PC机、一台作为服务器端的上位机与作为温室控制器的单片机stc15f2k60s2及数据采集系统和执行机构组成。温室智能控制系统软件大致可分成六个功能模块：人机界面模块主要是完成系统登录和密码管理等交互功能；数据接收显示模块主要完成实时数值和图形显示温室内的温度、湿度、光照度和CO2浓度等环境参数，存储从温室控制器接收到的数据，进行超限报警。参数设置模块是根据作物不同生长阶段的要求对温室环境各个参数进行设置，为智能控制提供参考数据；控制策略模块用于分析和处理从温室控制器接收的数据信息，为温室提供智能自动控制或者人工手动控制；状态显示模块用于实时显示现场各执行机构的运行状态；通信模块完成通讯协议的设定和数据的接收和发送。温室控制器可以实时的显示温室的温度、湿度、光照度和CO2浓度等各个环境参数及执行机构的运行状态。并能够根据上位机发送的控制指令对温室内的执行机构进行控制。现场层主要包括温度传感器、湿度传感器和光照度传感器等各种传感器构成的数据采集系统和各执行机构。控制策略方面，将控制模式分为春夏秋季模式和冬季模式两种。春夏秋季模式中，将湿度分为高、中、低三个等级，将一天分为午前（6：00-14：00）、午后（14：00-19：00）、傍晚（19：00-24：00）和夜晚（0：00-6：00）四个时间段进行模糊变温控制。冬季模式中将一天24小时分为白天和夜晚两个时段进行变温控制。远程控制PC机和上位机之间的通信由Visual C++6.0编程实现，只要在温室远程控制软件中输入上位机的网络地址，当网络连接成功后，就可以实时显示温室的环境参数和执行机构的运行状态，并且可以手动控制各个执行机构的启闭。经过试验表明：系统界面友好易于操作，能够实时显示温室环境参数和执行机构的运行状态，并能够根据控制策略智能的控制温室环境，环境参数能够控制在设定范围内，，为黄瓜创造适宜的生长环境。此外，系统的温湿度等控制参数易于修改以便适应不同作物的需求。技术人员或者管理人员在不同的省份城市甚至不同的国家都可以通过远程控制机方便的监控温室环境和执行机构的运行状态，达到减员增效的目的。
[Abstract]:In recent years, with the rapid development of computer technology, automation technology and network technology, intelligent and network has become the development direction of greenhouse environment control, cultivation and management of greenhouse crops has also begun to focus on energy conservation and sustainable development in Jilin area. The winter diffuse long and cold, summer is short and hot, the temperature difference between day and night and the base temperature and the temperature change are quite different compared with other regions. The existing control system is the control system of fixed ready, can change with the user's requirements and the seasons change.
Based on the characteristics of Jilin province is located in the northern cold region, the growth of greenhouse crops and control algorithm research, software and hardware design, design a set of intelligent greenhouse control system with remote control function, by adjusting the greenhouse temperature, humidity, light intensity and carbon dioxide concentration, provide a suitable environment for the growth of for greenhouse crop growth, improve crop quality, yield and save energy.
The study also designed a fuzzy PID controller of the greenhouse control system, and uses improved genetic algorithm to the two input, three output fuzzy PID controller optimized triangle membership function width and fuzzy control rule. A simulation of fuzzy PID controller after optimization, and compared with the conventional PID controller and the fuzzy PID controller.
The mathematical model of the controlled object of the conventional PID controller depends has slow response speed and overshoot problem. Using fuzzy inference three parameters of the PID controller Kp, Ki, Kd adjusted PID controller, its control effect is better than the traditional, improve the system static and dynamic performance, the system overshoot the reduced response speed. But because of the fuzzy PID controller is affected by the division of the domain of uniform and expert experience and rationality, the control effect is not optimal. And the fuzzy PID controller based on improved genetic algorithm, which is through strong global search capability of genetic algorithm, the control rules can get ideal and the reasonable fuzzy partition. The overshoot of the system is very small, the response speed is very fast. The application of genetic algorithm to optimize the parameters of fuzzy PID controller, the controller can significantly improve The simulation results show that the fuzzy PID controller optimized by the improved genetic algorithm has good control effect.
A set of remote intelligent control system design of fuzzy control based on this research, the control system consists of four parts: a remote control as the PC client, as a PC server as greenhouse controller chip stc15f2k60s2 and data acquisition system and an actuating mechanism.
The intelligent greenhouse control system software can be roughly divided into six functional modules: man-machine interface module is mainly to complete the interactive function of system login and password management; data receiving and displaying module mainly completes the real-time numerical and graphical display of the greenhouse temperature, humidity, light intensity and CO2 concentration and other environmental parameters from the data storage controller receives the greenhouse for, overrun alarm parameter setting module is set up. The greenhouse environment parameters according to the different growth stage of the crop requirements, to provide reference data for intelligent control; control strategy module for data processing and information received from the greenhouse controller, providing intelligent automatic control or manual control for the greenhouse; the state display module for real-time display operation the state of each actuator site; sending and receiving communication module to complete communication protocol and data set.
Greenhouse controller can display the greenhouse temperature, humidity real-time operation state, light intensity and CO2 concentration of various environmental parameters and actuator. And can be controlled according to the control instruction sent by the host computer to the greenhouse actuator. The field layer includes a temperature sensor, data acquisition system composed of humidity sensor and light sensor all kinds of sensors and actuators.
The control strategy, the control mode is divided into spring and summer autumn and winter mode two modes. In spring and autumn, the humidity is high, low, three levels, will be a day before noon, afternoon (6:00-14:00) (14:00-19:00) (19:00-24:00), evening and night (0: 00-6:00) four time the fuzzy temperature control mode in winter. 24 hours a day is divided into two periods of day and night temperature control.
The communication between the remote control PC and PC is realized by Visual C++6.0 programming, as long as the input of the PC network address in the greenhouse remote control software, when the network connection is successful, it can display the running state of greenhouse environment parameters and actuators, and can manually control every actuator to open and close.
The experiment shows that the system has friendly interface and easy to operate, can display the running status of greenhouse environment parameters and actuators, and be able to control greenhouse environment intelligent control strategy, environmental parameters can be controlled in the range of set value, create a suitable environment for the growth of cucumber. In addition, the temperature and humidity control parameters can be easily modified to adapt different crop needs. Technical personnel or management personnel in different provinces, city and even different countries are available through the running state of the remote control of greenhouse environment monitoring machine convenient and actuator, to achieve the purpose of shedding workers.

【学位授予单位】：吉林大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TP273.4;TP18

【参考文献】