1. In terms of development, DCS has evolved from traditional instrument panel monitoring systems. Therefore, DCS is inherently more focused on instrument control. For example, the YOKOGAWA CS3000 DCS system we use does not even have a limit on the number of PIDs (PID, proportional differential integral algorithm, is the standard algorithm for closed-loop control of regulating valves and frequency converters, and usually the number of PID determines the number of regulating valves that can be used). PLC has evolved from traditional relay circuits, and the original PLC did not even have the ability to handle analog signals. Therefore, PLC emphasized logical operation ability from the beginning.
2. In terms of system scalability and compatibility, there are numerous control products in the market, both DCS and PLC, which are produced and sold by many manufacturers. For PLC systems, there is generally no or very little demand for expansion, as PLC systems are generally designed for equipment use. Generally speaking, PLCs also rarely have compatibility requirements, such as the requirement for resource sharing between two or more systems, which is also very difficult for PLCs. Moreover, PLCs generally adopt dedicated network structures, such as Siemens’ MPI total linear network, and even adding an operator station is not easy or costly. During the development process of DCS, various manufacturers have their own systems. However, most DCS systems, such as Yokogawa YOKOGAWA, Honeywell, ABB, and so on, although the communication protocols within the system (process level) are not the same, the network platforms at the operation level all choose Ethernet networks and use standard or modified TCP/IP protocols. This provides convenient scalability. In this type of network, both the controller and the computer exist as a node, and as long as the network reaches its destination, the number of nodes can be increased or decreased and the position of nodes can be arranged arbitrarily. In addition, based on open protocols such as OPC and DDE in Windows systems, various systems can also communicate conveniently to achieve resource sharing.
3. In terms of databases, DCS generally provides a unified database. In other words, once a data exists in the database in a DCS system, it can be referenced in any situation, such as in configuration software, monitoring software, trend charts, reports… However, the database of PLC systems is usually not unified, and configuration software, monitoring software, and even archiving software have their own databases. Why is it often said that Siemens S7 400 is called DCS only when it reaches 414 or above? Because Siemens PCS7 systems only use a unified database, and PCS7 requires controllers to be at least S7 414-3 or higher.
4. In terms of time scheduling, PLC programs generally cannot run according to pre-set cycle times. PLC programs are executed from start to finish and then from scratch. Some new PLCs have been improved, but there are still limitations on the number of task cycles. DCS can also set task cycles. For example, fast tasks, etc. Similarly, for the sampling of sensors, the change time of pressure sensors is very short. We can use a task cycle of 200ms for sampling, while the lag time of temperature sensors is large. We can use a task cycle of 2s for sampling. In this way, DCS can schedule the resources of the controller reasonably.
5. From the perspective of network structure: Generally speaking, DCS commonly uses a two-layer network structure, with one layer being a process level network. Most DCS uses their own bus protocols, such as Yokogawa’s Modbus, Siemens and ABB’s Profibus, ABB’s CAN bus, etc. These protocols are built on the basis of standard serial transmission protocols RS232 or RS485. On site IO modules, especially analog sampling data (machine code, 213/scan cycle), are very large, and there are many interference factors on site. Therefore, network standards with high data throughput and strong anti-interference ability should be adopted. The bus structure based on RS485 serial asynchronous communication method meets the requirements of on-site communication. The sampling data of IO is converted by the CPU into shaped or solid data and transmitted on the operation level network (second layer network). Therefore, operation level networks can adopt numbers
According to the network standards of moderate throughput, fast transmission speed, and convenient connection, as the operation level network is generally arranged in the control room, the requirements for anti-interference are relatively low. Therefore, using standard Ethernet is the best choice. TCP/IP protocol is a standard Ethernet protocol that typically uses a communication speed of 100Mbit/s. The working task of PLC system is relatively simple, so the amount of data that needs to be transmitted is generally not too large, so the common PLC system is a one layer network structure. The process level network and the operation level network can either be merged together, or the process level network can be simplified into internal connections between modules. PLC does not or rarely uses Ethernet.
6. In terms of the scale of application objects, PLC is generally used in small self-control places, such as equipment control or control and interlocking of a small amount of analog signals, while large-scale applications are generally DCS. Of course, this concept is not very accurate, but it is very intuitive. Traditionally, we refer to systems larger than 600 points as DCS, and systems smaller than this scale as PLC. Our heat pump, QCS, and horizontal product control systems are generally referred to as PLCs.
What are the similarities between DCS and PLC?
After discussing so much about the differences between PLC and DCS, we should recognize that with the development of PLC and DCS to this day, they are actually moving closer to each other. Strictly speaking, nowadays, PLC and DCS cannot be cut in one go, and the concept between them is often blurred. Now, let’s discuss the similarities (similarities) between each other.
1. In terms of functionality, PLCs already have analog control capabilities, and some PLC systems even have strong analog processing capabilities, such as Yokogawa FA-MA3, Siemens S7 400, ABB’s Control Logix, and Schneider’s Quantum system. And DCS also has strong logical processing capabilities, for example, we have achieved all possible process interlocks and equipment linkage start stop on CS3000.
2. In terms of system structure, the basic structure of PLC and DCS is the same. PLC has been fully ported to computer system control today, and traditional programmers have long been phased out. Small scale PLCs generally use touch screens, while large-scale PLCs fully utilize computer systems. Like DCS, the controller and IO station use a fieldbus (usually based on RS485 or RS232 asynchronous serial communication protocol). If there is no expansion requirement between the controller and the computer, that is, if only one computer is used, this bus will also be used for communication. But if there is more than one computer used, the system structure will be the same as DCS, and the upper computer platform will use Ethernet structure. This is one of the reasons why the concept of DCS is blurred after the large-scale PLC.
3. The development direction of PLC and DCS: Miniaturized PLCs will move towards more specialized usage perspectives, such as more targeted functions and application environments. The boundary between large PLCs and DCS gradually fades until they are fully integrated. DCS will continue to develop in the direction of FCS. The core of FCS, in addition to more decentralized control systems, is particularly important in terms of instrumentation. The application of FCS in foreign countries has developed to the instrument level. The control system only needs to handle signal acquisition and provide human-machine interface and logic control. The control of the entire analog quantity is dispersed to the field instruments, and there is no need for traditional cable connections between the instruments and the control system. The entire instrument system is connected using a field bus. At present, Yokogawa has used FCS in the CNOOC Shell petrochemical project in China, and the instrument level uses intelligent instruments such as EJX, which has the world’s most advanced control level.