Abstract: Greenhouse facilities have been widely used worldwide. In view of the requirements of fully automated control of agricultural greenhouses, the author developed a greenhouse automatic control system with configuration concept. The system uses an industrial control computer or PC as the host computer and an embedded industrial control system as the slave computer. It can comprehensively control the temperature, humidity, light intensity, carbon dioxide concentration and pH and EC values ​​in irrigation in the greenhouse. The design of this system takes into account the concept of configuration and can be applied to most conventional greenhouses. Keywords: greenhouse, configuration, automatic control system 1. Overview 1.1 Introduction to Greenhouse Control Greenhouse control system is a high-efficiency agricultural development technology aimed at saving energy and increasing yield. Greenhouse control system mainly creates a suitable artificial environment for crop growth that is independent of the season through computer control and external actuator operation, so as to achieve high yield and low energy consumption of crops on a large scale. The research and development of greenhouses started earlier abroad. The Netherlands is located in western Europe and now has more than 10,000 square kilometers of glass greenhouses, all of which are operated by computers. Vegetable and flower production each account for almost half, and exports nearly US$2 billion annually [1]. Some leading greenhouse control companies abroad have developed greenhouse configuration control systems with strong functions. The domestic greenhouse control, especially the control of large and medium-sized greenhouses, still has a certain gap from the advanced level. This gap is mainly reflected in the three aspects of greenhouse control configuration, control accuracy and energy efficiency. 1.2 System characteristics The author incorporated the concept of configuration when designing this greenhouse control system. Users only need to set some system configurations and parameters on the human-machine interface to apply it to various conventional greenhouses. Configuration technology is the key to computer control technology. The application of configuration technology can develop real-time monitoring software for industrial control systems, thereby ensuring the reliability and control quality of the control system [2]. Due to the different conditions in various places, the external actuators are not completely the same and the system control parameters will also change. In specific applications, modifications should be made according to the specific situation. For example, some greenhouses add carbon dioxide monitoring and release devices, and the control system should also add corresponding carbon dioxide control. In this case, the greenhouse control system needs to be modified, especially the control part in the system software. This greenhouse control system applies the concept of configuration, considers various equipment and actuators in the greenhouse, and can be applied to various specific greenhouses after system settings. This greenhouse control system has relatively accurate temperature control of the greenhouse. Since crop growth factors (such as temperature, humidity, carbon dioxide concentration, etc.) are mostly nonlinear control variables with multiple inputs and multiple outputs, we use dynamic, feedback-based real-time monitoring to control the greenhouse. Among them, temperature is one of the most important factors affecting crop growth, so we conduct specific research on the specific temperature control algorithm and control strategy. After actual experiments, it was found that the temperature control error of this system is within 1 degree. 1.3 System Functions The greenhouse control system in this paper mainly has the following functions: (1) Display various parameters of each area in the greenhouse and the outdoor environment in the form of data and curves. Users can query the outdoor meteorological parameters, as well as the indoor parameters of the greenhouse and the operating status of the actuators, and the daily and historical data curves. The system collects analog quantities such as temperature, humidity, light intensity, wind direction, wind force intensity, rain sense, carbon dioxide concentration, pH value, and EC value 24 hours a day. It also collects the status switch quantities of skylights, side windows, inner curtains, outer curtains, circulating fans, and supplementary lights, and stores them in the database of the host computer. (2) The host computer human-machine interface centrally displays the current values ​​of various parameters inside and outside the greenhouse. (3) Users can choose automatic or manual control. Automatic control is generally used. In this case, the system automatically adjusts the greenhouse environment according to the parameters set by the user in advance. This system has a relatively complete irrigation control subsystem, which can automatically control irrigation through various control methods, such as: control by daily clock, control by weekly clock, control by light intensity, etc. In manual control, users can control the equipment in the greenhouse through the manual operation interface of the host computer. (4) The human-machine interface of the system includes the engineer debugging and configuration interface. The system engineer can easily configure and debug the entire greenhouse control system through this interface. 2. System Hardware The structural block diagram of the entire control system is shown in Figure 1. [align=center] Figure 1 System structural block diagram[/align] The host computer mainly provides the human-machine interface of the entire system and the storage of historical data. Through the host computer, users can set control parameters, observe the greenhouse environment, configure the system, print reports and other auxiliary functions. The slave computer is the main part of this control system. The motherboard of the lower-level machine adopts a PC104 embedded single-board computer system, which comes with an MMX 300MHz CPU. The board mainly includes a Flash interface, a memory interface, an RS232/422/485 serial port, an RJ45 Ethernet port, a VGA monitor interface, a PS/2 mouse and keyboard interface, and a PC104 bus. The lower-level machine communicates with the outdoor weather station through the RS232 serial port. The weather station adopts a bus structure [3] to transmit outdoor temperature, humidity, irradiance, wind direction, wind speed, rainfall and other data to the lower-level machine in real time. The upper and lower-level machines communicate through the Ethernet port, mainly to transmit control parameters and display parameters. The digital input/output module and the analog input module are both mounted on the PC104 bus of the motherboard. The lower-level machine system and the external actuator are photoelectrically isolated to prevent some external interference and to prevent external circuit short circuits from damaging the lower-level machine. 3. Overall Structure of System Software The software of this control system is divided into upper-level software and lower-level software, programmed using Delphi and Access database. The system control flowchart is shown in Figure 2: [align=center] Figure 2 System Control Flowchart[/align] As can be seen from the figure, the execution of the automatic control part is controlled by software timer interrupts. Once the specified time period arrives, the software will execute the automatic control program. After execution, it waits for the next execution. The system can continuously execute automatic control at certain intervals. Since the climate in the greenhouse is a type of climate with pure time delay, large inertia, and strong disturbance, and the greenhouse control must meet the plant protection requirements, the controlled variables such as temperature and humidity in the greenhouse cannot change drastically. Therefore, users can modify the execution cycle of automatic control in the system settings according to actual needs, but the value of this cycle is generally between 15 seconds and 120 seconds. This range ensures a certain degree of real-time performance of the greenhouse control and avoids situations where the system cannot execute in time due to an excessively short execution cycle. The main functions of the software system are divided into automatic control, manual control, system configuration, and auxiliary functions. The automatic control section is responsible for automatically controlling the greenhouse environment according to the parameters or requirements set by the user. The automatic control section mainly includes temperature control subsystem, humidity control subsystem, irrigation control subsystem and some other smaller control subsystems (carbon dioxide control, supplemental lighting control, etc.). The main function of the irrigation subsystem is to mix water and nutrients thoroughly to prepare the nutrient solution required for crop growth, and then supply the crop with the irrigation facility in a timely and appropriate amount according to the irrigation and fertilization program set by the user to ensure the needs of crop growth [4]. The manual control section can manually control various actuators in the greenhouse in real time. The auxiliary function section manages the user's account and system parameter setting permissions, etc. The system configuration section will be introduced in detail in the system configuration section below. 4. System configuration design 4.1 The core of the configuration - the tag name mechanism The configuration of the greenhouse control system developed this time is based on the tag name mechanism, and establishes a tag name chain that corresponds to the actual system. The tag name is the logical abstraction of each control quantity in the actual system and is a name directly related to the hardware. In the system, all control quantities, including analog and digital quantities, must have this one-to-one mapping relationship. System control is achieved by changing the values ​​of these tag names. Each control variable is distinguished by a tag name, which we call the system tag name. Correspondingly, each intermediate link in the control loop is distinguished by an intermediate tag name. The system tag name reflects the output of the control system, including the operating status of the equipment and the magnitude of analog and digital inputs and outputs. The intermediate tag names are the medium for calculation and processing, facilitating the handling of various links in the control process. The tag name mechanism facilitates both database maintenance and the construction and implementation of the control system. With this mechanism, we can collaboratively develop relatively independent basic functional modules in actual configuration software development, ultimately forming practical configuration software. We choose a DBMS (Database Management System) to manage the tag names. This system uses Access to design the tag name database structure and to complete operations such as inputting, inserting, deleting, and modifying real-time values ​​and alarm information of tag names. Both the screen configuration module and the control configuration module read tag name information through the ODBC interface. Because database software like Access has strong database management capabilities, this approach is highly suitable for situations with many monitoring points. Its biggest advantage is that it can fully utilize database functions, making it easy to query and organize and manage stored data. At the greenhouse site, the real-time data collected by the industrial control computer from monitoring points and the data from the control mechanism are divided into four categories: analog input, analog output, digital input, and digital output. For example, outdoor temperature and humidity belong to the analog input category. Therefore, we correspondingly divide the tag names into four categories: analog input tag names, analog output tag names, digital input tag names, and digital output tag names. Each category of tag names has its own unique data structure. For example, the data structure of analog input tag names mainly includes fields such as identifier name, status word, current workload value, upper limit of range, lower limit of range, filtering method, conversion type, conversion coefficient, upper limit alarm value, lower limit alarm value, sampling period, channel number, and refresh period. 4.2 Configuration Structure The configuration of the greenhouse control software of this system consists of three parts: control strategy configuration, monitoring screen configuration, and system structure configuration. The creation of control strategy configuration is implemented in the strategy editor, which is a design environment for functional modules [5]. The control strategy configuration is responsible for the specific control strategy inside the control subsystem of the system, as well as the setting of some control parameters. For example, the temperature control module of the system is a control subsystem with configuration ideas. It already includes all common equipment or actuators related to temperature in the greenhouse. The software system will automatically judge the equipment status of each area of ​​the greenhouse according to the user's settings of the area equipment in the system configuration configuration, and automatically select the appropriate control process. The control strategy configuration also includes the adjustment of the structural parameters, setting parameters and adjustable parameters of the control module. The structural parameters include functional parameters and connection parameters. Taking the PID functional module as an example, this module can change the actual form of PID control by determining the greenhouse functional parameters. The connection parameters are used to represent the connection relationship between the control module and the outside, and are an indispensable part of the implementation of the tag name mechanism. By establishing this connection relationship, various means such as monitoring and control of the system can be realized. The monitoring screen configuration is the result after calling the configuration, displaying the control site, processing the site data in real time, and alarming in real time. Considering factors such as screen refresh and display effect, each window is an independent space that can be freely configured. The basic sub-modules of the monitoring screen configuration include: (1) Graphical interface generation module: This module provides a variety of greenhouse equipment graphics, can edit various dynamic display points and flowcharts, and can easily connect dynamic points, real-time points and historical points. (2) Alarm module: Users can set alarm points such as lower limit alarm and upper limit alarm in the alarm module, which can easily realize dynamic alarm and sound alarm on the interface, and provide alarm records. (3) Report generation module: This module allows users to edit reports and generate the values ​​of each record point in the database. System structure configuration is the core part of configuration. Both the control strategy part and the monitoring screen part need to control the hardware through it. On the one hand, this module needs to collect field data and perform preprocessing. While writing to the database, it also needs to upload relevant data to the monitoring screen configuration for processing and display according to the configuration requirements. On the other hand, it needs to send back the control commands of the control strategy configuration part to realize the control of the field. It is the bridge between software and hardware devices in the whole system. The basic sub-modules of the system structure configuration include: (1) I/O module: according to the communication protocol, control the analog input and digital input and output of the lower computer. (2) Database generation module: this module includes real-time database and historical database, can edit database records, and convert, connect and archive database records. (3) Network communication module: manages the configuration of the communication protocol between the upper and lower computers, so that the system can run on the network. The system structure configuration is mainly reflected in the system configuration part of the software. Through the system configuration, the user can set the number of greenhouse cells and the status of different devices in each cell according to the actual situation. Different users will use different sensors, and the differences in their performance indicators and parameter conversion will affect the correctness of the system data acquisition. Therefore, in the system configuration, the user can also select different types of sensors, or define the parameters of the sensors used, such as: whether it is a voltage sensor or a current sensor, the range of the sensor, etc. This part is also the key to realizing the configuration idea control in the system. Through the user's settings, the system knows the specific number of cells and the specific devices and their quantities in the cells. In this way, the system can perform targeted control on each cell. 4.3 Configuration Steps The specific configuration steps for this greenhouse control system are as follows: First, analyze the controlled system, formulate a reasonable hardware scheme, and select relevant components. System engineers perform system design, including equipment, structure, and control modes. Second, determine the tag names, i.e., perform data configuration. From a system perspective, this should be considered part of the control configuration. We abstract the entire controlled object using a tag name mechanism, so users don't need to worry about the hardware structure, only understanding their control logic relationships. Third, this is crucial for determining system control. Establishing tag name links means that the inputs and outputs of each module can be linked to another module. This allows the standard control modules provided by the system software to be combined into a fairly complex control structure, fulfilling various system control requirements. Fourth, users can flexibly divide and implement various control screens of the system according to the system monitoring requirements. Following the principles of convenience and practicality, it can intuitively reflect the situation on-site in the greenhouse and the control effect. 5. Conclusion This system has been put into actual operation in the greenhouse. The configurable design has greatly reduced the workload of greenhouse system engineers, and the control effect is ideal. References: [1] Yu Haiye, Ma Chenglin, Chen Xiaoguang; Current status of research on automation of greenhouse facilities in developed countries [J]; Transactions of the Chinese Society of Agricultural Engineering, 1997.9, Vol.13, Sl [2] Lu Lingjie, Tian Yanyan, Chen Dongfang et al.; Design and implementation method of configuration software [J]; Journal of Daqing Petroleum Institute, 2001.3, Vol.25, No.1 [3] Li Shuzhong, Wang Chunfang, Zhang Zhen; Design of configuration software for automatic meteorological data acquisition station [J]; Journal of Qingdao University, 1998.6, Vol.2, No.2 [4] Eldar-Shany Agricultural Computer Automation Technology Co., Ltd., Israel; Automatic control system for irrigation and greenhouse climate [M]; 1999.7: Chapter 9 [5] Yang Chen, Zhong Jingliang, Chang Tao; Development of visualization configuration simulation software for distributed control system [J]; Journal of System Simulation, 1999.8, Vol.11, No.4