Share this

Application Research of Advanced Control Algorithms Based on FF Fieldbus

2026-04-06 06:23:26 · · #1
Abstract: Fieldbus control systems (FCS) have significant advantages over closed DCS systems and have become the mainstream system for process control. Based on an experimental platform designed by the Shenyang Institute of Automation, Chinese Academy of Sciences, using the FF fieldbus, a feasible scheme for applying advanced control algorithms to a single-loop liquid level control system within the FF fieldbus control system is proposed, thus better leveraging the technical advantages of both the FF fieldbus and advanced control algorithms. Keywords: Foundation Fieldbus; OPC; PID control; Advanced control algorithm Study of Advanced Control Algorithm Based on Foundation Fieldbus Abstract: The fieldbus control system (FCS) has more obvious advantages than the closed distributed control system (DCS), which has become the major in process control. Based on the experimental platform of Foundation Fieldbus designed by Shenyang Automation Institute of Chinese Academy of Science, the feasible project of applying advanced control algorithm to the single-loop liquid control of Foundation Fieldbus control system is proposed, which exerts the technical advantages of Foundation Fieldbus and advanced control algorithm. Key words: Foundation Fieldbus; OPC for process control; PID control; Advanced control algorithm Introduction The development of fieldbus technology has enabled control systems to take a big step forward in the process of moving from closed to open. Control systems based on fieldbus will replace closed DCS systems as the mainstream system for process control. FF fieldbus[1] is a fully digital, serial, bidirectional communication network, and also a fieldbus specifically designed for applications in the field of process automation. Therefore, it fully considers some characteristics of the field of process automation at the beginning of its design, such as bus, power supply, intrinsic safety, and high real-time requirements. my country has made great progress in FF bus technology research and the development of field device products that conform to the FF protocol. For example, the Shenyang Institute of Automation of the Chinese Academy of Sciences has developed various Foundation fieldbus products, such as pressure transmitters, temperature transmitters, host interface cards and communication stack software. In the DCS era, advanced control has been proven to bring huge economic benefits to enterprises. The combination of advanced control and fieldbus[2] is undoubtedly what the industry expects and is also the topic of this paper. In the FF fieldbus process control experimental system, the controlled object often has time-varying and time-delay problems, which pose a great challenge to the control system. It is a misconception that the use of fieldbus can solve these problems. The use of fieldbus only provides a better means of implementing control strategies. Dynamic matrix control (DMC) is an advanced control algorithm that can be directly used for time-delay objects without the need for additional control structures. This paper designs a liquid level feedback control loop for a dual-tank system on the networked measurement and control experimental platform based on the FF fieldbus designed by the Shenyang Institute of Automation, Chinese Academy of Sciences [3, 4]. It also proposes a feasible solution to apply advanced control algorithms to the FF fieldbus system based on the specific controlled object. The conventional PID control is implemented on the basis of the OPC server MicroCyber ​​.FFServer.1, and then advanced control algorithms based on OPC technology are implemented on the basis of PID control. Experiments show that the control effect of this solution better leverages the technical advantages of fieldbus and advanced control, and achieves the expected control effect. 1 Networked experimental platform based on FF fieldbus This system consists of two parts [5]: the FF fieldbus part and the field control model part, as shown in Figure 1. The FF fieldbus [6] includes the low-speed fieldbus H1 and the high-speed fieldbus HSE. The low-speed fieldbus H1 has a rate of 31.25Kbps and can be used for temperature, liquid level and flow control, etc. The signal type is voltage signal; the high-speed fieldbus HSE has a rate of 100Mbps and is generally used for advanced control, remote input/output and high-speed factory automation. The field control model can utilize existing laboratory equipment, thus saving on investment. Existing analog instruments can be connected to the fieldbus via current signals to fieldbus signal transmitters. [align=center]Figure 1 System Software Operation Relationship[/align] Taking single-loop liquid level control as an example, the operation of the software in the host computer is as follows: ①HSE Init interface software: Select the H1 network segment. The HSE interface program can interact with HSE devices in the Ethernet segment and H1 network segment devices under the LD device, providing data access interfaces to upper-level software such as configuration; ②Run the configuration program FF-Configurator configuration software, refresh the network segment to obtain the system's field device list and function block list. After refreshing the network segment, establish the application to complete the function block configuration. The connection between function blocks represents the signal connection through fieldbus communication, as shown in Figure 2; ③FF H1 and FF HSE OPC server: refresh once per second to realize the sharing of real-time data and historical data of the device, as well as alarm functions; ④Design SiaView monitoring software: create a new project, select PID in the object and drag it into the view. After connecting with OPC and editing, a PID function block operation panel can be obtained. [align=center]Figure 2 PID Configuration Strategy Configured in Application Window[/align] From the engineering window of the configuration software, select the IF-AI1, IF-PID, and FI-AO1 function blocks and drag them to the application view to connect and configure them into a PID control loop, as shown in Figure 2. Connect the function blocks to establish communication between them, enabling parameter value transfer and downloading configuration information to the field devices. To ensure the correct operation of the level control loop, the parameter values ​​of the function blocks need to be modified. Double-click the IF-PID function block to open its parameter window, and change the TARGET parameter under the MODE_BLK item in IF-PID to "AUTO" mode. Read the function block parameters to achieve automatic and normal operation of the single-loop level control. If you want to implement an advanced control algorithm in the FF fieldbus system, simply change the TARGET parameter under the MODE_BLK item in IF-PID to "MAN" mode, as shown in Figure 3. Then, use OPC technology to achieve communication between process variables and control variables. [align=center]Figure 3 IF-PID Function Block Parameter Table for Implementing Advanced Control Algorithms[/align] 2. OPC-Based PID Control Scheme To implement an advanced control algorithm based on the FF fieldbus, the primary focus should be on implementing conventional PID control. The software containing the advanced control algorithm reads and writes information from hardware devices via the OPC interface (acting as an OPC client). Accessing process data through the OPC server overcomes the differences between heterogeneous network structures and network protocols. 2.1 OPC Server Selection: The Shenyang Institute of Automation, Chinese Academy of Sciences, provided the OPC server MicroCyber.FFServer.1. The server's address space consists of all data items that the server can read and write. The full name of each data item can be obtained and related operations performed as needed. Figure 4 shows the address space diagram of the OPC server. OPC technology is used to implement communication between a client program written in VB6.0 and an OPC server. The server stores the real-time liquid level value (IF-PID-PV.VALUE) and custom variables such as the liquid level setpoint (IF-PID-SP.VALUE). Algorithm control is then performed to obtain the control quantity, which is written to the OPC server's entry (FI-PID-OUT.VALUE) to control the controlled system. [align=center] Figure 4: Address Space Diagram of the OPC Server [/align] 2.2 OPC Automation Interface Standard 2.2.1 Automation Interface To facilitate software development in various environments, the OPC Foundation provides two types of interfaces for data access specifications: automation interfaces and custom interfaces. Development based on custom interfaces requires in-depth COM/DCOM knowledge, which is relatively obscure and difficult to understand. Using automation interfaces has the following advantages: client programs can easily apply the interface without understanding its detailed internal mechanisms; event-triggered mechanisms can be used; and a universal dynamic link library (DLL) or control can be generated for use by all client applications. 2.2.2 Communication Mechanism of Automation Interface The OPC client program accesses the OPC server through the encapsulated OPC automation interface dynamic link library [7], as shown in Figure 5. The dynamic link library translates the customized interface of the OPC server into the automation interface desired by the OPC client program for the client program to call. The communication between the OPC client program and the dynamic link library is in-process communication, while the communication between the dynamic link library and the OPC server is based on COM/DCOM, which can be in-process or local connection, or remote connection. The encapsulated dynamic link library solves the interpretation of the customized interface and the communication between the two, thus greatly simplifying the development of the OPC client program. [align=center] Figure 5 Communication between OPC client program and server Figure 6 Automation object model[/align] 2.2.3 Automation object model The key to the development of OPC client program is to understand the encapsulation structure of the dynamic link library, that is, the automation interface standard. This standard can be represented by the automation object model shown in Figure 6. It mainly consists of 6 types of objects: ① OPCServer: An instance of the OPC server. Information about the OPC server can only be obtained after an OPCServer object is created. This includes searching for OPCGroups and creating OPCBrowser objects; ② OPCGroups: Automatically collects all OPCGroup objects created by clients within the OPCServer scope; ③ OPCGroup: An instance of an OPCGroup object. It contains all status information and provides necessary services for the OPCItems involved in the OPCGroup; ④ OPCItems: Automatically collects all OPCItem objects in the corresponding OPCGroup created by clients within the OPCServer scope; ⑤ OPCItem: An automated object that contains the definition of the "bit number", current value, status information, and last update time; ⑥ OPCBrowser: An object that browses all "bit number" names in the server. Only one OPCBrowser object can be created for each OPCServer. 2.3 Implementing an OPC Client Program for PID Control 2.3.1 Developing OPC Applications Using VB When developing OPC applications using VB, an OPC wrapper DLL that implements the OPC automation interface is essential. This OPC wrapper DLL is generally provided by the OPC server vendor, providing a tool with a good interface for OPC client program software development. This article uses the OPC Foundation's OPC wrapper DLL for illustration. Because the OPC wrapper DLL is not yet registered in the newly created VB project, it must be registered first using the following method: ① From the VB menu, select [Project (P)] - [References (N)]; ② In the list of [Available References (A)], select the corresponding OPC wrapper DLL file name, here we select [OPC Automation 2.0]. 2.3.2 Implementation Steps The OPC client program written in VB includes the following steps [8]: ① Add a reference to the server, create an OPC server object, and connect the client program to the server; ② Create a group collection and add group objects; ③ Add OPC items, use the OPCBrowse method to browse all items in the entire server, select the required items, add them to the specified group, and display their values ​​and status; ④ Display the added groups and items in the main screen. 2.3.3 Program Design The following is an OPC synchronous client program with PID function implemented in VB. [align=center] Figure 7 OPC client program implementing PID control [/align] This single-loop control process is a self-balancing non-oscillating process with mutually influential double-capacity processes. Its mathematical model can be described by the following transfer function: G(s) = Where, K[sub]p[/sub] , T[sub]p[/sub] , τ are the process gain, time constant and time delay. The parameters in the formula can be obtained by graphical method according to the step response curve. The following is a method for determining the parameters of the transfer function: Let the control quantity in Figure 7 be q, the measured value be y, the new steady-state value be the set value, the gain K can be directly calculated from the steady-state values ​​of the input and output, and T[sub]p[/sub], τ can be determined by graphical method. In order to be able to call the measured value y collected on the OPC server at any time and draw the corresponding step response curve, the selected data is stored in the specified database (using SQL2000 database). At the same time, in order to be able to retrieve data from the historical database when performing algorithm simulation, and to continuously refresh the data as the data is collected, and to optimize the algorithm, as shown in Figure 8. [align=center] Figure 8 Storing OPC server data in SQL server database[/align] 3 Advanced control based on OPC control scheme Dynamic Matrix Control (DMC) algorithm [9] includes prediction model, optimization control, and feedback correction. The main function of dynamic matrix control is to determine the control action according to the DMC algorithm. In the design, the DMC control type is abstracted into a DMC control class, and a corresponding object is generated. The attributes describing the entity are encapsulated within the object. That is, the parameters required for DMC control, such as model time domain length N, optimization time domain length P, and control time domain length M, are private members of the DMC control class. External programs cannot access this data, while DMC control initialization and online calculations are public interfaces that external programs can access. The simulation object uses a single-loop control model of the FF fieldbus process experimental system. The liquid level object mainly consists of two water tanks connected in series, each 20cm long, 10cm wide, and 40cm high. The liquid level height h2 in tank 2 is the controlled variable, with a maximum value of 40cm and an operating point typically set around 20cm. Because the system has volumetric hysteresis and pipe hysteresis, the liquid level object is a multi-capacity object with pure time delay. Experimental testing of the system showed a pure time delay of approximately 3 seconds. The model time domain length is 10, the optimization time domain length is 6, the control time domain length is 5, the control cycle is 1000ms, the number of control cycles is 1000, the set value is 20, the error weight is 1, and the control weight is 0.01. [align=center] Figure 9 OPC client interface developed with VB interface Figure 10 DMC algorithm implemented in VB6.0[/align] 4 Conclusions This paper studies the design process of the process control experimental system of FF fieldbus in depth, and then studies OPC technology. The purpose is to develop OPC automation interface client application through VB, realize the application programming of dynamic matrix advanced control algorithm to OPC client application, and facilitate user use and system maintenance and upgrade. At the same time, the database system of SQL Server 2000 is studied for historical data storage and algorithm simulation. How to realize more complex advanced control algorithms can be realized by using OLE automation technology to realize VB and Matlab mixed programming [12]. That is, use VB to write OPC client program to realize data communication, and use Matlab to write advanced control algorithm, so as to analyze field data. This is the focus of the next research. References: [1] Long Ziping, Wan Manying. Overview of FF fieldbus[J]. Automation and Instrumentation, 1999, 14(4): 5-7. [2] Yuan Mingzhe, Liu Yuzhong, Zhou Yue et al. Advanced control and FF fieldbus[J]. Industrial Detection and Automation Device, 2003, (2): 7-9. [3] Liu Dan, Yu Haibin, Wang Hong et al. Typical applications of self-developed FF fieldbus control system[J]. Automation Instrumentation, 2005, 26(3): 46-48. [4] Chen Na, Qi Shufen. Experimental measurement and control system based on FF fieldbus[J]. Journal of Qingdao University of Science and Technology, 2004, 25(1): 80-82. [5] Xu Yonghong. Networked Control Experimental Platform Based on FF [D]. Harbin: Harbin Institute of Technology, 2003. [6] Liu Dan, Yu Haibin, Wang Hong, et al. Basic Principles and Implementation of FF HSE and FF H1 Protocol Gateways [J]. Information and Control, 2004, 33(6): 719-723. [7] Wang Ling, Yang Xinzhao, Zhang Fengxing. Application of OPC server in FF bus system [J]. Journal of Instrumentation, 2004, 25(4): 227-228. [8] Zhao Mingsheng. Predictive control based on OPC technology [J]. Computer Engineering and Applications, 2005(9): 224-228. [9] Li Jinxia, ​​Qiu Gongwei. Simulation study of dynamic matrix control and its improvement method [J]. Journal of Fuzhou University (Natural Science Edition), 2004, 32(5): 560-562. [10] Zhang Wen. Research on monitoring system based on fieldbus process control system [D]. Shanghai: Shanghai University of Science and Technology, 2001. [11] Shao Qike. Research on distributed software technology and its application [D]. Zhejiang: Zhejiang University of Technology, 2002. [12] Mao Yanke. Research on intelligent monitoring system construction technology and application based on fieldbus [D]. Shanghai: Shanghai University, 2002.
Read next

CATDOLL 108CM Coco

Height: 108cm Weight: 14.5kg Shoulder Width: 26cm Bust/Waist/Hip: 51/47/59cm Oral Depth: 3-5cm Vaginal Depth: 3-13cm An...

Articles 2026-02-22