7.DIFFERENCE BETWEEN DCS BASIC STRUCTURE AND PLC
DCS is the abbreviation of Total Distributed Control System.It refers to the decentralized control of hazards,centralized management and display.In the late 1960s,someone developed a programmable logic controller for logic operation.It is called PLC for short.Mainly used in automobile manufacturing industry.In the mid-1970s,DCS,which completed analog quantity control,was introduced to the market to replace analog instrument control mainly based on PID operation.The idea of DCS was first proposed by the original instrument manufacturers,which were mainly used in the chemical industry at that time.Later,the computer industry was engaged in the development of DCS.
In the 1970s,microcomputer technology was not mature and computer technology was not developed enough.The operation station,controller,I/O board and network interface board are all self-developed by the DCS manufacturer,that is,all components are dedicated.
In the early 1970s,someone used a minicomputer such as PDP/1124 to replace the original centralized installation of analog instrument control.There are many cables connected to the central control room.For example,a minicomputer is used as both a controller and a CRT connected to the minicomputer as a display device(i.e.,a human-computer interface).A minicomputer needs to receive signals from thousands of transmitters or other sensors to complete hundreds of loop calculations.Obviously,the danger is a little concentrated.There are as many cables as the analog instrument,and once the minicomputer breaks down,the control and display are lost.The digital control did not achieve the desired purpose.
Later,someone proposed to separate control and display.One computer completes the control calculation task,and the other computer completes the display task.In addition,a process as a controlled object may have many points that need to be displayed and controlled,some of which also need closed-loop control or logic operation.Each part of the process as a controlled object will be relatively independent,and can be divided into several independent processes.Then,the input and output points that need to be displayed and controlled on the independent processes in the computer control system will be allocated to several computers,The computing task originally completed by a minicomputer is completed by several or dozens of computers(controllers).The failure of one of the machines does not affect the overall situation.The so-called strategy of"wolves replacing tigers"means danger dispersion.The display,operation,printing and other management functions are integrated together,and the above two parts for control and display are connected into a system by network.At that time,someone called this kind of system as the distributed system.
How much is the appropriate dispersion of the danger?This is related to the development level of computer technology at that time.In the mid-1970s,complete decentralization was the operation of a loop completed by one controller.At that time,because people were not familiar with digital technology,they were used to analog instruments.In the late 1970s and 1980s,loop controllers were popular.Digital controllers were made almost identical to the original analog instruments in appearance,without changing their operating habits,and PID operations were digitized internally.One instrument(one computer)completes the control task of a loop.It is expensive,but the danger is dispersed.Then communication network is used to connect each controller and CRT based human-computer interface into a system.At this time,the network structure is usually star structure.The manufacturing cost of the loop controller is too high,and the price/performance ratio is not good.Later,in order to reduce the cost,there were two loop and four loop controllers,whose price/performance ratio was slightly better.For a large and medium-sized system,the price/performance of DCS is better than that of the system composed of loop controllers.Some loop controllers are still used in some special places.
If there are too many loops to complete,such as a controller collecting thousands of points and completing hundreds of loops,the danger is too concentrated.In this case,the danger must be dispersed.With the development of computer technology,the computing power,storage capacity and reliability of a computer are constantly improved,and the tasks completed by a computer can also be increased.The completed tasks can also be concentrated.In addition,redundant technologies such as controllers and networks have also been developed,and control operations can also be centralized.
From the perspective of current DCS,a controller can complete dozens of loop operations,hundreds of points of acquisition,plus an appropriate amount of logic operations.After field use,the effect is relatively good.This leads to the problem of controller upgrade.Sometimes the distance between the controller and the detection element is still far,which promotes the development of field bus.Such as CAN,LOONWORKS,FF and other field buses,as well as HART protocol receiving boards,are used in the DCS system.
DCS is divided into three parts,controller with I/O board,communication network and human-machine interface(HMI).The I/O board is directly connected to the production process through the terminal board to read the signal from the sensor.There are several different types of I/O boards,and each I/O board has a corresponding terminal board.
L Analog input,4-20 mA standard signal board and millivolt signal board used to read thermocouples;4-16 channels are different;
L Analog output,usually 4-20mA standard signal,usually has few channels,4-8 channels;
L Switching value input;16-32 channels:
L Switching value output,switching value input and output are also divided into boards of different voltage levels,such as DC 24V and 125 V;AC 220 V or 115 V,etc;8-16 channels are different;
L Pulse input,used to collect the signal of speed;4-8 channels are unequal;
L Fast interrupt input;
L HART protocol input board;
L Fieldbus I/O board;
Each I/O board is connected to the I/O bus.For the safety and integrity of the signal,the signal shall be renovated before entering the I/O board.The inspection,temperature compensation and filtering of the upper and lower limits can be completed on the terminal board or separately.The boards that complete the signal renovation are now called signal conditioning boards.
The I/O bus is connected to the controller.In the 1980s,due to the weak computing ability of the controller,DCS separated the tasks of the controller in order to increase the I/O points.In fact,there were three types of controllers.Namely,the controller,analog data collector and logic solver that complete the closed loop operation.They have their own I/O buses,and the I/O buses of various DCS are different.If fast speed is required,parallel bus is preferred.Generally,serial bus is used more.In particular,there are many RS485 buses,and I/O points of analog data collector and logic solver can be more.
The closed loop controller,analog data collector and logic solver can be directly connected to the communication network with the human-computer interface.Each different controller on the network acts as an independent node on the network.Each node performs different functions.They shall have network interfaces.In order to save network interfaces,some DCS pre connect all the equipment used for process control,namely,closed-loop controller,analog data collector and logic solver,to the control bus,which is called process control station.This can increase the I/O points that can be received by the process control station and save interfaces.Then it is connected to the network through the interface and connected to the human-computer interface.With the development of computer technology,the calculation ability of the controller is constantly enhanced.For example,a controller made by a PC has a strong ability to receive both analog and switching logic operations.A controller becomes a node on the network.It is connected with human-computer interface through network.
The controller is the core component of DCS,which is equivalent to a PC.Some DCS controllers are PCs themselves.It mainly includes CPU,RAM,E2PROM,ROM and other chips,as well as two interfaces.One receives the signal from the I/O bus downward,and the other interface sends the signal to the network upward to connect with the human-computer interface.ROM is used to store control algorithms for completing various computing functions(some DCS are called function block library).Store function blocks in the library,such as control algorithm PID,PID with dead zone,integral separation PID,arithmetic operation plus,minus,multiply,divide,square,square root,function operation primary filter,sine,cosine,X-Y function generator,lead lag;More advanced algorithms include Smith estimation,C language interface,matrix addition and matrix multiplication;Logical operations include logical AND,logical OR,logical NOT,logical AND,etc.Usually,the station function block not only combines analog quantity with switching value,but also connects with people.The more function blocks,the more convenient it is for users to write application programs(i.e.configuration).The configuration connects the function blocks to form a control scheme according to the process requirements.Store the control scheme in E2PROM.Because E2PROM can be rewritten,the configuration will change with the process change,so the configuration is stored in E2PROM.Different users have different configurations.During configuration,the user selects the desired function block from the function block library,fills in the parameters,and connects the function blocks.The control scheme is formed and stored in E2PROM.At this time,the controller is in the configuration mode and becomes the operation mode after it is put into operation.
Operating system,function block configuration software and communication software are installed in the controller.
For the safe operation of the system,the closed-loop controller must be redundant,one for use and one for standby,and hot standby.In order to make redundancy successful,the following points should be noted:the hardware and software versions of the two controllers must be consistent;Check whether the sending receiving chip is intact;Whether the redundant chips are intact.Whether the settings of the two modules are the same,and whether there is a manual operation station.
The communication network connects the process station and the human-computer interface into a system.There are several different types of communication networks.Such as bus type,ring type and star type(see figure).The bus shape is also logically circular.Star type is only applicable to small systems.Whether it is ring or bus type,broadcast type is generally adopted.Other protocols have been used less.The speed of the communication network is about 10M and 100M.
The human-computer interface has four different forms of nodes,which are operation station,engineer workstation,historical trend station and dynamic data server.
U The operating station is installed with the operating system,monitoring software and driver software of the controller.Display labels,dynamic flow chart and alarm information of the system.
U The engineer workstation can configure the controller(CAD)and the operation station(as a dynamic flow chart).If the monitoring software has strong drawing ability,the drawing can be completed independently by the monitoring software.Another function of the engineer station is to read the configuration of the controller,which is used to upgrade the controller and find faults.We call it the Reverse Engineer Station.
U Historical trend station is used to store historical data,generally using disk array(called RAID technology).
U Dynamic data server is the interface between DCS and MIS system,and also the isolation equipment between DCS and Web.
The design principles of DCS and PLC are quite different.PLC was developed by imitating the original relay control principle.In the 1970s,PLC only had switch logic control,and the first application was in the automobile manufacturing industry.It stores the instructions for performing logic operation,sequence control,timing,counting and operation;And through digital input and output operations,to control various types of machinery or production processes.The control program compiled by the user expresses the technological requirements of the production process and is stored in the user program memory of PLC in advance.During operation,execute the stored program item by item to complete the operation required by the process flow.There is a program counter in the CPU of the PLC that indicates the storage address of the program step.During the program operation,the counter automatically adds 1 for each step of execution.The program executes from the starting step(step number is zero)to the final step(usually the END instruction),and then returns to the starting step for cyclic operation.The time required for PLC to complete a cycle operation is called a scanning cycle.For different models of PLC,the cycle scanning cycle is between 1 microsecond and tens of microseconds.The cycle operation of program counter is not available in DCS.This is also the reason why the redundancy of PLC is not as good as that of DCS.DCS is developed on the basis of operational amplifier.All functions and the relationships between process variables are made into functional blocks(some DCS systems are called expansion blocks).In the mid 1970s,DCS only had analog control.For example,in the TDC2000 system,one controller can complete the operation of eight PID loops in one second.The first application is the chemical industry.The main difference between DCS and PLC is in the logic calculation of switching value and the calculation of analog quantity.Even though they infiltrated each other later,they still had differences.After the 1980s,except for logic operation,PLC also had some algorithms for control loops,but it was still difficult to complete some complex operations.PLC was programmed with ladder diagram,and the analog quantity operation was not very intuitive in programming,which was more troublesome.However,in terms of solving logic,it has the advantage of being fast.At the microsecond level,it is less than 1 millisecond to solve 1K logic program.It treats all inputs as switching values,and 16 bits(also 32 bits)are an analog quantity.DCS regards all inputs as analog quantities,and bit 1 is the switching value.Solve a logic in the order of several hundred microseconds to several milliseconds.For PLC to solve a PID operation,it takes tens of milliseconds,which is comparable to the operation time of DCS.Large PLC uses another CPU to complete the analog calculation.Send the calculation results to the PLC controller.The time required to solve PID is different for different DCS models,but it is on the order of tens of milliseconds.For example,in the early TDC2000 system,the control operation of eight loops is completed within one second.With the development of chip technology,the time of solving an algorithm is shortened.The time required to solve an algorithm is related to the arrangement and configuration of function blocks.
In terms of grounding resistance,the requirements for PLC may not be high,but DCS must be less than a few ohms(usually less than 4 ohms).Analog isolation is also very important.Intrinsic safety barriers shall be provided in places with explosion hazards.
For systems with the same I/O points,the cost of using PLC is lower than that of using DCS(about 40%can be saved).PLC has no special operation station,and its software and hardware are universal,so the maintenance cost is much lower than DCS.A PLC controller can receive thousands of I/O points(up to 8000 I/Os).DCS controller can only have hundreds of I/O points(no more than 500 I/Os).If the controlled object is mainly equipment interlocking and there are few loops,PLC is more appropriate.If it is mainly analog control and there are many function operations,it is better to use DCS.DCS is much better than PLC in terms of redundancy of controller,I/O board,communication network,etc.,advanced computing,and special requirements of the industry.Because PLC adopts general monitoring software,it is easier to design the management information system of enterprises.
In particular,the special operation station of DCS is not justified.It is formed by historical reasons.If the DCS manufacturer does not open the operation station and connect with the management information system of the factory,individual DCS will be in danger of disappearing from the market.
With the birth of new technologies,negative impacts also follow.The opening of new operation stations makes it easy for viruses and hackers to invade the system.During design,passwords are set on the operation station,and more isolation and firewalls are added to the system.Minimize negative impacts.
8.MULTI SCREEN OPERATION STATION,BACKDOOR NETWORK OF OPERATION STATION AND DUAL COMPUTER DUAL NETWORK
A DCS system can have several operation stations,each of which can display the same content,which is called redundant operation.It can also be different content.For a power plant,a 300MV generator set has a relatively centralized process,and it is generally considered appropriate to adopt redundant operation of two operation stations.However,for large units,due to the large number of input and output points,one display is not enough,and it is better to have multiple displays.At this time,the monitoring software should support multiple screens.Early DCS used to add more display cards on the host of the operation station,such as adding 2 display cards and 3 display cards.Two CRTs or three CRTs can be connected respectively.This method is used to increase the man-machine interface.For production processes other than power plants,such as cement plants and iron and steel plants,a production process is long,geographically dispersed,and controlled by a set of DCS systems.It has many input and output points.In order to save investment,an operation station often corresponds to some production processes.For example,the rotary kiln and the mill in the cement plant are far away,and the operation station can be installed in two places.At this time,one operation station is configured with one printer.If the host of one of the operation stations is broken,the content displayed by this operation station will disappear.For the safe operation of the system,the overlapping configuration of tags is used to achieve safe operation.If the system has two operation stations,station A and station B,the system has 2000 labels,and operation stations A and B need to configure 1300 labels respectively,of which 300 labels are overlapped.Even if operation station A fails,only operation station B is used for monitoring,and the system works normally.Because the important labels are on the B operation station.It is worth noting that overlapping configurations can also have negative effects.In case of operation failure,the responsibilities of the management personnel of operation stations A and B will be unclear.
Due to the rapid development of network technology,the number of human-computer interfaces can be unlimited.The operation station is divided into master station and slave station.The master station is connected to the DCS network.The data in each controller database can be read directly from the DCS network through the DCS network interface.The number of master stations is determined by the number of interfaces.The slave station is connected to the master station.It has no database.When the flow chart screen is displayed,the master station displays the complete process screen,while the slave station displays the process screen of the master station.The number of slave stations is unlimited.Both the master station and slave station are connected by backdoor network.With the backdoor network,when maintaining the system,each operation station can make a copy of the text.When the system is running,even if one of the master operating stations fails,its slave can be used as the slave of the other master.In the network,when connected to the print server,the server has many channels,and several operation stations share one or more printers.The network connecting master station and slave station has no relationship with DCS network.So it is called backdoor network.ARCNET was used in the early backdoor network.Later,because some DCS operating station hosts used DEC minicomputers,DECNET or Ethernet was used in the backdoor network.During tag configuration in the operation station,whether important I/O point overlapping configuration should be determined by the user.
With the development of network and database technology,there is a new way to connect human-computer interface with DCS network.Connect the server to the DCS network.The server and human-computer interface form the server/client format.The server and human-computer interface are connected by Ethernet.Ethernet is cheap and usually adopts dual networks.In order to improve the security performance of the server,the redundancy mode is adopted,which is dual computers and dual networks.Remote operation is possible.There are operation station,engineer station,historical trend station and MIS system interface in the client.The database in the server can be configured.
It will be very convenient to transfer information through the network in local areas.
The DCS system of Beijing Hollysys and the PROCONTROL P of ABB are of this structure.If the process control is connected to the INTERNET network,it will be invaded by viruses or hackers.The security level and firewall must be added.From the perspective of use,DCS made in China has reached a good level in structure and hardware manufacturing.The development of functional blocks needs to be further strengthened.
The protocols between Ethernet networks follow the seven layer protocol of the OSI model(Open System Interconnection).The seven layer protocols are physical layer,data link layer,network layer,transport layer,session layer,presentation layer and application layer.
The physical layer describes the transmission medium,connector and signal pulse standard.The repeater or hub is also a device in the physical layer,which is independent of the specific content of the information frame transmitted.A frame is a group of digital pulses sent to a medium.Used to transmit information.The frame size is 64-1518 bytes.The frame includes pre synchronization signal,message header,data information and frame check sequence.A repeater or hub merely amplifies the electrical signal in the wire and continues to transmit it forward.
The data link layer describes the topology and communication standards between local systems.Ethernet can work with multiple physical layer standards(twisted pair cables,optical cables)and multiple network layer standards.Connect the physical aspects of the network(cables and digital pulses)to the abstract world of software and data flows.The information transmitted in the network is transmitted in frame format.There are message headers and footers in the frame,and datagrams are packaged into the frame.Detect errors and adjust data flow.The frame message consists of source and target media access control MAC addresses,and uses this information and the content of the data area to establish the CRC tail.The data link layer uses topology rules according to the network to send frames to the physical layer(network cable).Bridges and switches are devices at the data link layer.Because they support frames.Both use the information in the frame header to regulate traffic.The frame header is responsible for identifying who sent the information and where the information was sent.The frame message header contains two areas,which are used to identify the source and target of the transmission.They are the node addresses of the source and target systems.The size of the frame header is always 14 bytes.
The network layer describes how systems on different network segments look for each other,and also defines the web address.A web address is a system name or number assigned to a group of physical connections.It is the cornerstone of all the upper layers.The unit is datagram.For example,the IP datagram delivery protocol of the network is an example of network layer functions.The term web address is called differently depending on the protocol used.IP is adopted,which is called subnet.
The transport layer handles the actual operation of data and is ready to transmit it over the network.If the data is too large to be used as a frame,the transmission layer will decompose it into smaller parts and prepare a serial number.The serial number allows the transmission layers on other systems to re combine the data into the original content.The CRC check of frames is carried out at the data link layer,and the transmission layer can play a backup role.The functions of the transport layer,such as the IP transmission protocol(TCP),and the unit is segment.
The session layer handles the establishment and maintenance of connections between two or more systems.It ensures that specific service requests are completed correctly.
If the system is running multiple network applications,the session layer is responsible for maintaining the correct communication order and ensuring that the incoming data is introduced to the correct application.
The presentation layer ensures that data can be received in the form of applications.Its task is encoding and decoding.
The task of the application layer is to determine when access to network resources is required.
9.LOOP CONTROLLER AND DCS SYSTEM
In the 1970s,because instrument engineers working on the site were not familiar with digital control,they wanted the instrument panel of digital control to be the same as the original analog instruments,so that they could not change their operating habits,but also for the sake of hazard dispersion,thus loop controllers appeared.For example,HONEYWELL KMM series,FOXBORO SPEC 200 and Japan Beichen(later merged with Yokogawa)HOMAC series tables have process variables(PV),settings and bar diagrams of control output(CO)on the panel,process values displayed by pointer,hand/automatic switching,alarm confirmation,etc.This digital instrument,which can complete PID-based loop control,has only one or two loops(two single loops).Some people called this instrument a programmable regulator in order not to be confused with a programmable logic controller(PLC).By the 1980s,a series of digital instruments,such as double circuit(two cascade circuits),four-circuit controller,32-way data collector and paperless recorder,appeared.The bar diagram and hand/automatic switch are still retained on the panel,and digital display is used instead of pointer display.These digital instruments are collectively called loop controller.
So far,there are many kinds of such instruments.The most characteristic is that it can communicate with the upper position(i.e.man-machine interface)to form a system.Each table can be connected with fieldbus.I/O can be expanded as required.The internal algorithm of this system can be pre-programmed as a successful energy block in the form of ROM.Configuration(not programming)can be performed according to the desired control strategy.For example,the 353 product of MOORE company in USA has more than 80 functional blocks.Control configuration is realized by configuration diagram and connected with human-machine interface by MODBUS.The monitoring software of human-machine interface can be either IFIX or INTOUCH.Because both software have drive software for this programmable regulator.Domestic monitoring software can also be used as monitoring software for this system as long as the driver software of this programmable regulator is available.Otherwise,it can not be connected.353 itself only has 4 AIs and 2 AOs,which means it can control 2 PID loops.If a control circuit is to be added,a fieldbus is to be used because it has an interface with LONWORKS,on which a module of LONWORKS can be connected.Software bundles are required when making up a system.For data security,a photoelectric converter is added between the table and the bus.This system is shown in Fig.1.Like FOXBORO and SMAR circuit controllers,it has multiple loops.And it also has the interface with programmable logic controller(PLC),so it is widely used in China.
There are several analog outputs,called controllers for several loops.Analog inputs can be more than analog outputs.In addition,there can be a small amount of input and output switching.
The other is that it can not be connected to fieldbus.Its I/O points can not be expanded,but it can also form a system with upper communication.Such products are similar to UDC of HONEYWELL in USA,other products of FOXBORO,Fisher's 900 series,Yokogawa in Japan,Fuji,Toshiba and ABB in Europe.The most representative product is the system 6000 in Hong Kong and Europe.It uses circuit controllers from several British manufacturers,programmable controllers in the United States and monitoring software in the United States.It is integrated into S6000 system by Hong Kong people.Because of its good sales network,it is widely used in the mainland.Especially in the pharmaceutical industry.
The overwhelming number of digital tables that can be described as products is a fixed PID algorithm,which outputs control quantities(CO)as long as process variables(PVs)are connected to the terminals of the instrument and set manually.In other words,there is only one function block of an instrument,which is PID.There are a lot of such products both in China and abroad,such as Qianyo Table of Japan,Chaoyui of Hong Kong,Tianjin Instrument Factory of China,Instrument Factory of coastal provinces and so on.They are sometimes called S-series instruments in China.The appearance of the product is shown in Fig.3.It has a very single function,only one function per table.The feature is that it cannot be configured.
In general,there are few input and output points for the loop controller.If there are hundreds of I/O points to complete the control tasks of dozens of PID loops,and each table needs a housing and panel with bar chart display,the price is much higher than that of pluggable modules.In addition,its function block is much less than that of DCS controller because of its small memory,especially when there are many analog acquisition quantities,switching quantities and advanced operations,it is not economical to use loop controller.If there is no upper position to display,it does not feel intuitive.Systems integrated by loop controllers,together with CRT-based human-machine interfaces,are much more expensive than DCS consisting of modules.Another disadvantage is that the system composed of loop controllers is usually connected with human-machine interface by MODBUS and can only use star structure,i.e.only one main station.This is a defect for power plants that require redundant operating stations.So it only applies to smaller systems,i.e.less I/O points and more controlled objects with more control loops.Pharmaceutical plants that produce antibiotics are more suitable.In addition,especially for some chemical enterprises,it is often used for outdoor operation and has its own advantages.Like power plant bypass systems,they are also suitable for use.That is,it also has a certain market share.Their design principle is no different from that of DCS.Programmable regulator is the combination of the controller of DCS and its corresponding hand operating station.So loop controller is a branch of DCS system and belongs to DCS in essence.
DCS,PLC and programmable regulator are all computer control systems.Their basic structures are similar,and some of them are indispensable,but there are some differences with the size of the system and the controlled objects.To read the field signal into the control system,first connect the signal wire of the sensor to the terminal.If the sensor is not a single one,it has more signal lines and needs some signal preprocessing,then it needs terminal board.Analog signals are converted to digital signals and analog input boards are used.After operation,output board is needed to convert digital signal to analog signal,and then to terminal board.Finally,the signal is sent to an actuator(e.g.a valve).Due to the development of communication technology,only the operational part(i.e.controller)of the control system can be discussed.If Foundation Fieldbus is used,this part can be omitted.The functional blocks for operation are stored in the memory of the sensors and actuators respectively.The configuration is called through the network.At present,the bus application is not wide enough because there are not many functional blocks developed and the benefits of manufacturers limit it.FISHER-ROSEMOUNT does a lot of work for fieldbus.
For larger systems,there must be a CRT(or LCD)based human-machine interface to make it easier to use.For users,the system with less investment,more convenient use and full functions is suitable.
10.DATA ACQUISITION SYSTEM AND SOFT DCS,SOFT PLC
The data acquisition system only collects the field signals into the control computer,in which there is no loop control function block.It is only displayed on the HMI.It should be simple.Hardware structure bar can be divided into three levels or two levels.I/O board,controller and human-machine interface.Simple can be divided into I/O board and human-machine interface.However,some industries have some special requirements.For example,the control system of a power plant often separates the boiler control from the data acquisition system.DCS is used for boiler control,while DAS is used for others.There are also many manufacturers producing data acquisition systems in the world.Generally,data acquisition system is technically OK,but there are some special requirements in power plants.There are points in the power plant that need to record the sequence of events.It has a fast interrupt function(usually 2 milliseconds).There are not many such points,about 100
The amount of switch around the point.The resolution of the input board should be in milliseconds.The difficulty is event sequence recording(SOE).There should be a function to store historical data on the input board.When the human-machine interface is called up for viewing,the time sequence of events can be displayed.SOE(SEQUENCE OF EVENT)is manufactured by a dedicated manufacturer.This is also the case with some DCS systems dedicated to power plants.
The most widely used data acquisition system is OPTO22 product in USA.One small module along the way is switching quantity(DI,DO)and analog quantity(AI,AO).There are 42 small modules such as different voltage levels for user selection.When using,install the small module on a large board(i.e.B1 board,B2 board),8 analog blocks/blocks,16 switch blocks/blocks,in addition to these small modules on B1 and B2 board,there are also serial communication ports.Install the two panels on one rack and the rack in the cabinet.CRT-based human-machine interface runs on the WINDOWS platform.Installed monitoring software with MODBUS driver software,PAREGON software was used before,but fewer people use it now.Now more FIX and INTOUCH are used.The man-machine interface can read I/O signals from small modules and display them on CRT.More than 20 plants in the United States produce similar products.It is in the same grade as STD bus products.STD bus products are gradually withdrawing from the market due to the heavy workload of users.
In some pure acquisition systems,OPTO products are widely used.The main advantage is low maintenance.If closed-loop control is required,add a controller similar to PC(LC4)between the human-machine interface and I/O.There are some functional blocks in the controller which can configure various control strategies.Generally,OPTO22 is used as a large control system with thousands of I/O points because the number of functional blocks is relatively small.
The most widely used small module for OPTO switching quantity is the isolation module as the input module for DCS switching quantity.Protect DCS itself from lightning strikes.Due to the simplicity of the manufacturing process,the domestic module user response is also very good,but the price is less than half of the imported price.
At the beginning of 1990s,due to the development of computer technology,it was pointed out that ordinary micro-computer(PC)has been mature enough to meet the control requirements both in operation speed and memory.Operate in front and back office.Both DCS and PLC need no controller.It is quite difficult to transfer the function code and trapezoid diagram of DCS and PLC controller into the PC of NT operating system.The computer is directly connected to the I/O board.Because the control algorithm already exists in PC,PC serves as both human-machine interface and controller.It is called soft DCS and soft PLC.AB company in USA has soft PLC.Ethernet connection between PCs.So far,the application is still not very wide.SOFT DCS and SOFT PLC are used in some small systems.
11.FUNCTION BLOCK IN DCS CONTROLLER
DCS not only completes the functions of original analog instruments,but also greatly exceeds those of analog instruments.This is because it uses 4C technology such as advanced computer technology,communication technology,CRT technology and control technology.With digital control,the algorithms pre-stored in ROM in the controller can be said to be infinite,and each algorithm represents a function.These functions are realized by analog circuits in analog instruments,which are limited by drift,resistance,capacitance and other devices of analog circuits.It is very expensive or almost impossible to make a highly accurate analog instrument.The arithmetic of digital control is realized by program.Replace the functions that analog circuits can achieve with programs.In theory it is infinite,and that is a great advance.The principle is the same for all DCS systems.Various algorithms are commonly referred to as function blocks.The assembly of function blocks is called function block library.
The controller of DCS mainly consists of CPU,ROM,RAM,E2PROM,address setting switch,etc.CPU completes the calculation,ROM is used to store the operating system and function block library,and the arrangement of function blocks in ROM is definite,so users can not change them.RAM is used to store CPU operation results and I/O signals.Control scheme of E2PROM memory function block connection.Control schemes are not lost when power is switched off.Moreover,the control scheme is not appropriate and can be modified.Modified by ultraviolet radiation,wipe out the contents of E2PROM.The controller should have addresses in the network set by these address switches.The arrangement of function blocks in ROM is called function code by DCS.The order of codes is the address of the function block in the library.DCS from various manufacturers handles algorithms differently,so the names are different.If DCS refers to these algorithms as internal instruments,the essence of them is a procedure and the essence is the same.When using these blocks,you should indicate the relationship between the inputs and outputs of the block and the many parameters required for the operation.On the basis of the controlled object,
When determining the control strategy to form the control scheme,select the function blocks needed for control in the function block library,make clear the connection relationship between each function block,and first define its address in E2PROM,and the connection with other function blocks is also expressed by address.Fill in the required parameters at the same time.These tasks are called configurations in order to distinguish them from programming in language.When configuring,it can be used as a diagram,i.e.CAD.CAD has a software that draws the functional blocks required for each control circuit and the connections between them.It is called an engineer's station.Use it to configure the controller.When configuring,draw the SAMA or ISA diagram first.When using small function blocks,draw SAMA diagrams,large function blocks,draw ISA diagrams.
The most important function block in function block library is PID function block,whose relationship between output Y(t)and input X(t)is proportional-integral-differential,i.e.it plays an extremely important role in process control.The PID function block must be used to complete the closed loop control.Block diagram of one control loop and cascade control loop of closed loop control.
12.INTERCONNECTION OF HETEROGENEOUS SYSTEMS,TIME SYNCHRONIZATION AND SDSL TECHNOLOGY
An enterprise,especially large and medium-sized enterprises,has many technological processes,each of which is relatively independent.Different types of DCS systems will be used for construction at different times or for technical renovation at different times.Several different types of PLC systems will be used for interlock and control of equipment.In order to establish an integrated management information system in an enterprise,the interconnection and time synchronization of different systems become very important problems.Early interconnection between DCS and PLC used PLC as an input board of DCS.DCS experts designed an interface board with PLC and connected to the I/O board of DCS itself on the same I/O bus.The controller reads the signal from PLC and sends the signal from PLC to DCS operation station through the network of DCS.One major disadvantage of this interconnection method is that the fast signal of the PLC is displayed on the operating station of the slow reaction DCS.It has been determined that the action of a microswitch takes more than a few seconds to be displayed on the operating station of the DCS.Some DCS even take 8 seconds.Another interconnection method is to connect individual PLC signals with hard wires to the input board of DCS,so that the number of points input is small and the cost is high.Some DCS designs a node on the DCS network,which is specially used to read PLC signals.It is difficult to send PLC signal to the network of DCS.The disadvantage is the high cost of developing interfaces.Interfaces often fail even when development is successful.A typical example is a power plant that combines a steam turbine with a boiler.The steam turbine is imported from the United States or other countries.The control system of the steam turbine(PLC)is combined with the main equipment and the control system of the boiler is selected by the user.The user would like to display both information on one operating station.To develop this operation station,the development cost is over tens of thousands of dollars.Failures often occur even when development is successful.
For PLC,special operation station was developed in the middle of 1970s,but it was not very successful.Later,it did not develop a special operation station.At that time,the PLC only handled the switching quantity,and the digital display on the controller can also meet the requirements as a man-machine interface.A network connection is used between the controller of the PLC and I/O.By the end of 1980s and 1990s,analog control was introduced into PLC.Users feel that it is inconvenient to have no CRT-based human-machine interface.Users choose general monitoring software to run on NT platform and use common micro-computer as hardware platform to program driving software of each PLC.Thus,PLC also has an operation station.
For general monitoring software,such as INTOUCH software,support Microsoft developed communication protocol DDE(DYNAMIC DATA EXCHANGE),fast DDE,network DDE,etc.
U DDE allows machines used in Windows environments to establish client/server relationships,send and receive data,and send instructions to each other.Data is provided by the server and interested data requests are received from other machines used.The requesting machine is the client and the data is sent by the server.
UFast DDE provides many DDE information packaged into a single DDE information.Packing improves efficiency and reduces DEE data transfer between servers and clients.
U-network DDE extends the standard DDE functions,including LAN communication through serial ports.The extension of the network allows different computers connected to the network to operate as DDE connections for server and client applications.For example,network DDE supports DDE between an IBM compatible computer connected to a LAN or modem and a non-PC-based platform such as in VMS,UNIX and other operating environments.
U SuiteLinkK is a TCP/IP based protocol designed to meet industrial needs such as data integrity,high throughput and ease of diagnosis.This protocol standard only applies to Windows NT4.0 or higher.
U OPC(OBJECT LINKING EMBED PROCESS CONTROL)client,data can be retrieved from the server.Comply with TCP/IP protocol.
Many vendors develop monitoring software because of its need.In the 1990s,there were more than 100 kinds such as FIX,INTOUCH,PARAGON,ONSPEC,CIMPLICITY,etc.China has King of Configuration,SYNALL,etc.Many drive software of DCS and PLC have been developed by each monitoring software,among which FIX and INTOUCH are the most popular.On NT platform,if we use FIX and INTOUCH software,we can exchange data with each other as servers or clients.
When DCS was introduced in the 1970s,the computer technology was relatively low and there were few parts available in the market.The operation station developed by DCS manufacturer has special operation system,special monitoring software,special interface hardware and CRT,special printer,special hard disk and soft drive.Even cables and plugs and sockets for connections are dedicated.This does not preclude DCS manufacturers from considering the maximum commercial benefits.They are the only providers of DCS software and hardware.They can only guarantee that their own systems can communicate with each other.Although they have done a lot of work,they can not connect with other systems.Some people use the analogy of dedicated systems that are similar to religious organizations and are almost incompatible with others,and each business tries to maintain market share with its trustees.The term"religious war"is often used when we describe discussions between supporters of two different proprietary products or operating systems.
Due to the rapid development of computer technology,DCS is in a relatively stable stage of development.Later,there were a lot of micro-computer components on the market,but DCS could not be connected,and different systems could not be interconnected.DCS manufacturers did not modify their software according to market changes,so DCS became a true"automation island".Users of DCS feel very inconvenient.In addition,it is easily damaged compared with the control station.For about 7-8 years,many users decide not to use the original DCS manufacturer's operating station when replacing the operating station because it is expensive and hard to buy spare parts.The general monitoring software is installed on PC and NT platform,and the driver software of a specific DCS is developed to make a general operation station instead of the original special operation station.
Some DCS manufacturers simply remove special operation stations and equip them with general operation stations.For example,MOORE's APACS system adopted INTOUCH software(FIX can also be used)in 1996.MOORE only provides the configuration requirements of APACS's drive software and hardware,and the hardware is provided by users themselves.It works well after use.It is a little better than MOORE's own monitoring software.This may be related to fewer software developers from DCS manufacturers than from General Software Development.For example,the operation station of MEASUREX also uses general monitoring software.At present,the most common alternative operation stations are FIX and INTOUCH.
In the early DCS,manufacturers developed their own special monitoring software.When programming,the driver software and display software of the controller are not written separately.If general-purpose computers and NT platforms are used,they will cost less than general-purpose software and require less hardware than general-purpose software.However,due to the small sales volume with the system,the problems of software itself are not fully exposed.If the hardware can meet the general-purpose and special-purpose requirements,special-purpose deactivation is more serious than general-purpose deactivation.It is understood that a set of DCS system,with PC as its hardware and NT as its operating system,is assembled with special monitoring software and can be shut down several times a day.As software versions are upgraded,downtime continues unabated.Repairs were made several times a year and deadlock persists.Compared with special software and general software,special software is more closed now than general software.There is no network suite software available for use on the network.If general monitoring software is used,the man-machine interface between DCS and DCS,between DCS and PLC,between PLC and PLC can be interconnected.As long as it is Microsoft's WINDOWS interface,it can act as a client.Data are read by DDE,fast DDE and Suelink.They can act both as servers and as clients.An integrated management information system can be formed within an enterprise.To ensure the safe operation of the operating station,a single HMI can be used for interconnection.Common monitoring software usually has network version software.After configuration,it can form server/client format through network and transmit real-time information to the remote.In the client,you can view the data in the dynamic data server database made of general software in real time.It can also send information to standard database of MIS system(such as SQL,SYBASE,ORACLE).Universal monitoring software in their package software,also has Web software,which is installed as Web server through Web server.Browse the data in the dynamic data server.It is not real-time data.Microsoft's Glance software also connects to the Web.Browse the data in the MIS database.Using a Web server,browsing data is slower than using a server/client.
The advantages of using universal operating stations can be summarized as follows:
Low price
Functional Extension of DCS
User familiarity
The connection of different machines must also comply with OSI seven-tier model and adopt general monitoring software.All network suites are written according to open rules.Using common monitoring software can also have negative effects and some specific functions may not meet the requirements.
Time synchronization of systems is also an important issue.Network Time Server is a high-quality time synchronization device for all online workstations.Generally,the absolute time of DCS system is formed on the operating station(man-machine interface),and there is only relative time in the controller.For time synchronization,satellite time GPS(GLOBAL POSITION SYSTEM)can be used as a reference to calibrate the time of DCS.Specific implementation methods are as follows:the satellite receives the antenna and the time server receives the atomic clock of each satellite in the GPS system,then sends the time signal to the dynamic data server or the operation station of DCS.The time signal is sent to the network of DCS through the driver software of DCS in dynamic data server or operation station.Ethernet was rarely used in DCS network,but now it is also used because it is cheaper.It is widely used in Management Information Network(MIS).Ethernet also uses a broadcast protocol,but differs from token networks.It uses random access link control.In the Ethernet network,each node can send data randomly.But when two or more nodes send data at the same time,a data sending conflict occurs.Conflicting nodes are returned and sent back using a random delay algorithm.That is,carrier monitoring/collision detection multi-access technology is used.There are many random delay algorithms which are applied to Ethernet at present.There are now gigabytes available for point-to-point communication.
At present,MIS system has been connected with DCS and PLC to form a system.DCS and PLC data are sent to Ethernet via dynamic data server.Most of its transmission carriers are optical fiber or coaxial cable.Coaxial cables are susceptible to lightning strikes.The dispatch room is usually a central area,based on which it is connected into a star network structure.Servers are configured in the central area.Such as ORACLE,SYBASE,SQL,etc.Each sub-module is stored in the server.Such as financial management sub-module,personnel management sub-module,etc.The signals from DCS and PLC are in the module of real-time production subsystem.The production of real-time data is huge,it can be self-contained,set up its own server,and then connect with the database of MIS system.At present,the transmission carrier is also optical fiber or coaxial cable.The cost of laying optical fiber is relatively high(about 20,000 RMB/km).In MIS system using optical fiber as carrier,some places may be far away from the central area,but a few data also want to be sent to the center for display.At this time,optical fiber is not necessarily used,but telephone line can be used to transmit the network.Data in places such as balance,coal conveyance and water treatment are scarce.Just add a router with special Ethernet to the switch of the telephone master and a slight adjustment in the wiring of the data source plus MODEN.Ethernet routers and cats cost less.Fiber optic and telephone lines can be used in Ethernet at the same time.This is the SDSL(Symmetric Digital Subscriber Line)technology.The remarkable advantage is that the telephone is smooth and the network is smooth.Therefore,maintenance is simple.In addition,the technology can be used with a telephone switchboard already in place,eliminating the cost of repeated wiring.
SDSL(Symmetric Digital Subscriber Line)technology is now widely used in plant networks abroad.If the enterprise is small,it can use all telephone lines without optical fiber and the communication rate is 2.3M.The command rate from the information side to the information side is the same.
13.DCS OPERATING STATION COMPONENT REPLACEMENT SKILLS AND DCS PRINTING SYSTEM
According to incomplete statistics of DCS systems imported from abroad before 1995,there are about 3000 sets and more than 20 different types of DCS systems,which come from USA,Japan and Europe respectively.Some systems are special and enter our country together with the main equipment.For example,some control systems of paper mill such as Measurex ODX may enter our country with the main equipment of paper making.Most DCS enter our country directly.Another example is Continental System 6000,which is an integrated DCS system.Loop controller is used for process control and PARAGON software is used as monitoring software in man-machine interface.Using serial port of PC,connect PLC into HMI.There are several different types of DCS from the same manufacturer,which are designed due to system upgrades or different system sizes.Put the same manufacturer together and observe the use of several DCS systems in China.Users have more than dozens of systems as follows:TDC3000(TDC2000,R150,S9000),μXL,(CENTUM),I/A(SPECTRUM),N90(INFI90),PROVAX,RS3(ΔV),APACES,MOD300(MODIII),AC450(AC210,AC500),WDPF,HIACS,TELEPUM and MAX-1(MAX-1000).There are also some systems with less than 10 users,such as PROCONTROL P14,only 4 sets in China,5 sets in D/3 and not many TOSDIC systems in China.No matter how many or how few users are in China,the common feature of these DCS systems is that many components of the human-machine interface are also purchased from third parties,not manufactured by themselves,and most of them have now been discontinued.If DCS operation station host uses DEC VAX orαMachine,which was merged by COMPAQ in 1996,has been out of production for 5 years now.Not only does the machine not produce,but also the chip.If DCS like this are still imported now,spare parts will be a difficult thing to do.
Another example is the hard disk used in the operation station,most of the interfaces with the hard disk are ST412 interfaces(such as ST225,ST4096,ST4097,XT1085,etc.)or IDE interfaces,which are large in size but small in capacity.Production has been suspended for many years now.When the hard disk of the operating station is damaged,it will be difficult to buy the original hard disk.It can only be replaced by a slightly modified BIOS hard disk of the same series as the original hard disk.A small number of SCSI interfaces are easier to buy.There are also low-resolution CRTs,which are generally RGB inputs.At present,CRTs on the market can not connect with most DCS operating stations.Once CRTs fail,the operating stations can not function.The solution is to connect a converter to the current display and DCS operation station to display the information of DCS system.Also,ARCHIVE tape drives,TEAC floppy drives,etc.are not available.Finding old products is difficult.If found,it may not be useful.For example,the hard disk also needs low pass or initialization.
If you want to use the original station host,you need compatible hard disk,CRT,keyboard,etc.DCS imported in the 1980s,even the manufacturers of DCS,sometimes can not supply some spare parts.Sometimes it is available,but the delivery time is too long to meet emergency requirements.The price can be said to be sky-high.Because DCS manufacturers of these components are also purchased from third parties and need to look for them,the supply time is very long.
The value refers to the special printer for each DCS system,such as GENICON printer,which can print graphics.Its printer head,color band,etc.are difficult to buy.Another example is the alarm printer equipped with the original DCS.As long as there is an alarm in the system,the printer will print continuously,and the sequence of alarm generation is not stored.It is impossible to find out which alarm occurred first.Some useful information is submerged in a large amount of useless information.Not only wastes a lot of printed paper,but also useful information is difficult to extract.
Printing system is developed under the above background.Printing software is used in printing management system of various DCS systems.This printing system can be connected with operation stations of various DCS systems to form a printing information management system,including online alarm records,screen copies and event lists.It not only completely replaces the special printer,but also has functions of storage and query,so it is more suitable for field application.Each printing system can be linked to four DCS operating stations.
Print management system,whose core is print management software,is installed on a common PC and runs on the WINDOWS platform.In order to ensure data security,RAID technology is used and double hard disks run redundantly.This software is a set of print information management software developed for improving the printing function of computer automatic control(DCS)operation station and enhancing the ability of information storage,access and copy.It has the features of strong function,friendly interface and convenient operation.It can fundamentally avoid the phenomenon of information loss often caused by printer failure when only using printers in the past,and also can fundamentally avoid useful information being inundated with a lot of useless information.The printout information from the operating station can be stored on the hard disk for a long time as a file(depending on the capacity of the hard disk and the amount of printed information,which can last months or even years)or copied to other storage devices(e.g.compact discs).
The printing system can automatically identify alarm records,control operations,statistical reports,event triggering reports,picture copies and configuration lists in the information of the operating station and process them separately.It can observe the alarm record,control operation,statistical report,event triggering report,picture copy and configuration list of the operation station in real time and process them separately.The alarm record and control operation of the operation station can be observed in real time,and data and status requiring immediate response can also be easily accessed.It is possible to selectively print the received information using Offline Printing or to immediately output the printout from the operator's station to the printer using Online Printing.
The printer system works according to the principle that each printer manufacturer produces printers with drivers connected to the WINDOWS operating system.Most of the operating systems of the original DCS are special,and only a few printer drivers are available.Commonly used printers on the market do not have drivers,so they cannot be connected.In fact,the task of printing system software is to write the anti-drive software of WINDOWS operating system,which is to make the signal from DCS operating station recognized in WINDOWS environment.Then connect the common printer through the driver of WINDOWS.Printing system software needs to identify different data formats because the printers connecting different types of DCS operation stations have their own data formats,such as ANSI format,IBM format,etc.Before connecting the printing system to the operation station,make sure of its data format so that the data format can work in the WINDOWS environment.Printing system software shall be written with`anti-drive software in various data formats'.In addition,software for printing graphics and text differs greatly.Printing text only,its software is simple and printing graphics is complicated.For example,if you use an HP printer to print graphics,you should first find out whether it is dot matrix or page scanning format.Most Hp printers are page scanned.To interpret PCL instructions,the software printers can print graphics on the operating station screen.If it is not an HP printer,explain other instructions.
