Abstract: This paper compares DCS and PLC in terms of historical development, technical characteristics, and applicable fields to illustrate the characteristics and development trends of DCS and PLC technologies. Keywords: DCS; PLC; NT6000 In the field of thermal power plant automation, DCS and PLC are two completely different yet closely related concepts. Both DCS and PLC are products of the combination of computer technology and industrial control technology. Thermal power plants use DCS for their main control systems, while PLCs are mainly used in auxiliary workshops. Both DCS and PLC have operator stations providing human-machine interaction, rely on computer-based controllers to complete control calculations, exchange data with primary components and actuators through I/O cards, and possess a network communication system. Given their similarities, why are DCS and PLC so different, and how do we choose between them in engineering practice? This paper provides an overview of their historical development, technical characteristics, and development directions, hoping to offer some guidance for thermal engineering professionals. The DCS case uses the NT6000 from Keyuan as an example, striving for detailed and clear explanations. 1. Historical Development and Core Concepts of DCS and PLC DCS stands for Total Distributed Control System. It refers to the dispersion of hazards and the centralization of data. It entered the market in the mid-1970s, completing analog control and replacing analog control instruments that primarily relied on PID calculations. The concept of DCS was first proposed by instrument manufacturers, and it was mainly used in the chemical industry at the time. PLC, on the other hand, was successfully developed in the late 1960s and is called a Programmable Logic Controller (PLC). It was mainly used in the automotive manufacturing industry. The design principles of DCS and PLC differ significantly. PLC evolved from the principle of relay control, and in the 1970s, PLCs only had on/off logic control. It stores instructions for performing logical operations, sequential control, timing, counting, and calculations; and controls various machines or production processes through digital input and output operations. The user-written control program expresses the technological requirements of the production process. This program is stored in the PLC's user program memory, and during runtime, it executes the stored program line by line to complete the required operations. DCS was developed based on operational amplifiers. All functions and relationships between process variables are designed as function blocks. In the mid-1970s, DCS only supported analog control. The main difference between DCS and PLC controllers lies in their handling of digital and analog operations; even with some later overlap, distinctions remain. After the 1980s, PLCs added control loop algorithms in addition to logic operations, but completing complex calculations remained difficult. PLCs use ladder diagram programming, making analog operations less intuitive and more cumbersome. However, they offer a faster solution to logic problems. DCS, on the other hand, uses function blocks to encapsulate analog and logic operations, providing a clear expression for both logical and complex analog operations. However, its logic operation expression efficiency is lower than that of PLCs. The historical differences between DCS and PLC are significant, greatly influencing their subsequent development. However, the most significant factor influencing their later development is not the difference in their origin technology, but rather the difference in their original concepts. The core concept of DCS (Distributed Control System) is a computer-controlled system that disperses hazards and centralizes data. Therefore, the development of DCS involves continuously utilizing the latest advancements in computer, communication, and control technologies to build a complete distributed control system. DCS provides users with a complete, safe, reliable, efficient, and flexible solution for industrial control. The core concept of PLC (Programmable Logic Controller) is to replace relays, executing sequential control functions such as logic, timing, and counting, and establishing flexible programmable control devices. Therefore, the main theme of PLC development is continuously improving various performance indicators to provide users with a complete and flexible control device. DCS is a system, while PLC is a device; this is the fundamental difference in their concepts. This difference has a profound impact, permeating every aspect of technology and economics. 2. Technical Characteristics and Mutual Penetration of DCS and PLC Different conceptual foundations and different development paths result in distinct technical characteristics for DCS and PLC. Technological development is not closed; mutual learning and penetration are always present in the development process. 2.1 Control Processing Capacity We know that a PLC controller can often handle thousands of I/O points (up to 8000+ I/Os). DCS controllers typically handle only a few hundred I/O points (no more than 500). Does this mean DCS developers are technically incompetent? Probably not. From the perspective of distributed systems, centralized control is not allowed. A controller with too many I/O points is useless in practical applications. DCS developers don't need to develop drivers for controllers with many I/O points; their main focus is on providing system reliability and flexibility. PLCs are different. As independent, flexible control devices, stronger I/O handling capabilities naturally indicate higher technical levels. The overall application level of the control system is primarily the concern of engineers and users, not the core objective of PLC manufacturers. Another indicator of control processing capability is processing speed. In people's minds, PLCs are much faster than DCSs. From a certain perspective, this is indeed the case. PLCs are highly efficient at executing logic operations, executing 1K logic programs in less than 1 millisecond, and their control cycle (taking DI input directly to DO output as an example) can be controlled within 50ms; while DCS uses the same method for handling logic and analog operations, its control cycle is often over 100ms. When comparing using PID algorithms, we can see that a PLC executes a PID calculation in a few milliseconds, while the T2550 controller of the NT6000DCS takes about one millisecond to calculate a PID. This indicates that the actual computing power of PLCs and DCSs is comparable, with some DCS models even being more powerful. The difference in control cycle time is mainly related to the controller's scheduling design. Large PLCs often use a secondary CPU to handle analog quantity calculations, while the main CPU performs high-speed digital quantity calculations. Therefore, even if the analog quantity calculation speed is average, the speed performance in digital quantity control is still excellent. DCS, however, processes both digital and analog quantity calculations at the same speed, resulting in a less than ideal control cycle time. Newer DCS controllers have adopted designs similar to those of large PLCs, significantly improving their control cycle performance. Taking the T2550 controller of the NT6000DCS as an example, the controller can be configured with four tasks of different priorities, with a minimum calculation cycle of 10ms. With high-speed I/O cards, the control cycle can reach 15-20ms. Analog quantity calculations are then assigned to other tasks with longer cycles. 2.2 Data Communication Exchange Data communication exchange mainly refers to the control system network and its data exchange methods. DCS has an inherent advantage in this aspect. The "distribution" of a distributed system is mainly reflected in independent controllers, while the "centralization" is mainly reflected in human-machine interface devices with complete data. The network is what connects the distributed and centralized components into a distributed system. Therefore, from the early stages of DCS development, networking has become a core technology direction for DCS manufacturers. Redundancy technology and narrowband transmission technology were among the first technologies developed or successfully applied by DCS manufacturers. PLCs are mainly designed as independent devices, and their "network" is actually serial communication. The development and widespread application of industrial Ethernet technology has, in form, leveled the gap between DCS and PLC networks. Superficially, many DCS and PLCs use industrial Ethernet, but the essential differences still exist. Take MODBUS-TCP, used by many PLCs, as an example. MODBUS is a serial communication protocol, not a network, which is undisputed; but is MODBUS-TCP a network? Many people have this question. A closer analysis reveals that MODBUS-TCP is a communication method that loads the MODBUS communication protocol onto the Ethernet TCP protocol. While it possesses the appearance of a network, it remains a master-slave management approach with a data table transmission structure. In contrast, a DCS, taking the NT6000's ELIN network as an example, although also based on industrial Ethernet, utilizes a masterless tokenless LIN network protocol accumulated over nearly 30 years, with long-term successful applications on 1M OLIN, 2.5M, and 20M ARCNET. On the ELIN network, all stations are equal, and there is no master management station. Moreover, data communication is based on structured data at the module level, offering data management capabilities far superior to the data table method. Taking the PID module as an example, its basic data includes PV, SP, and OP. Using a data table transmission method, you must first define the data addresses of PV, SP, and OP as 01, 02, and 03. Other stations also receive data in a data table format, but what data is 01? What data is 02? These can only be determined through the data definition table. The traditional data table management method is cumbersome and error-prone. Managing tens of thousands of data points in a large system using this method, spreading them out across data tables, is extremely daunting. The NT6000DCS, however, employs a structured management system based on modules. It treats a PID as a module; to access its PV value, the module is accessed first, managed in the form of PID.PV. This categorizes and centralizes all the spread-out data into small boxes, managing them by module.component, significantly improving management efficiency. The problem of PLC data communication exchange stems primarily from the fact that PLCs have long been developed as independent devices without a system concept; moreover, they are mainly used in small control systems, where the problems are not readily apparent, resulting in slow development. While some large PLCs have improved in this area, reaching the level of a DCS will require a considerable amount of time. 2.3 Configuration and Maintenance Functions: Configuration and maintenance functions include logic configuration, download and modification, operation and debugging, and remote diagnostics. Early PLCs primarily used ladder diagrams, while DCS primarily used module function diagrams. After years of development, the International Electrotechnical Commission (IEC) has specified five programming languages ​​through the IEC 1131-3 standard. Currently, mainstream DCS and PLCs all claim compliance with this standard and support several or all of these programming languages. Considering development efficiency and program readability, modular function charts (MFLACs) and sequential function charts are increasingly becoming the primary programming methods, while ladder logic and structured text have become development tools for custom modules. Large PLCs are increasingly resembling DCSs in their configuration methods, and the gap is gradually narrowing, while small PLCs still primarily use ladder logic. DCSs, after years of development, have accumulated a large number of advanced algorithm modules. For example, the NT6000 has device-level modules that centrally complete basic device control and fault alarm functions within a single module. Network communication is also based on these modules, greatly improving software development efficiency. A single device-level module is equivalent to 0.5K of ladder logic quantities; for a PLC to achieve the same function, it would be much more cumbersome. PLCs lack solutions for downloading, modifying, running, debugging, and remote diagnostics. In contrast, DCSs are designed from the perspective of system needs from the outset and have accumulated years of comprehensive solutions. Taking the NT6000DCS as an example, the system allows for both online modification and downloading of control strategies without affecting normal system operation. The NT6000DCS has comprehensive virtual DCS functionality, enabling not only verification of configuration logic but also the construction of a complete virtual DCS connected to the model for system simulation and debugging. The NT6000DCS features robust safety measures and provides a remote debugging solution based on a wide area network. 2.4 Hardware Packaging Structure: PLCs typically use a large-base rack with enclosed I/O modules. This enclosed structure improves the reliability of the I/O modules, providing resistance to radio frequency, electrostatic discharge, and damage. PLC modules typically have 8, 16, or 32 I/O points. Most DCS systems use a 19-inch standard chassis with plug-in I/O modules, which are exposed. Each module typically has 8 or 16 I/O points; 32-point modules are rarely used. The DCS structure stems from its primary application in controlling large objects. The 19-inch standard chassis facilitates dense deployment, and the fewer I/O points are due to the requirement for dispersion. PLCs, with their large baseplate racks and enclosed modular structure, offer greater flexibility in management and configuration, and higher reliability for individual devices. Therefore, many DCS systems have adopted the structural advantages of PLCs, using similar packaging structures, such as metal casings for I/A modules and conductive plastic casings for NETWORK-6000+. 2.5 Human-Machine Interface Devices In the early days, the human-machine interface device (HMI) of a DCS system was a dedicated device provided by the DCS manufacturer. PLC manufacturers generally did not provide HMI devices, which were often implemented by engineers using general-purpose monitoring software (such as iFixit, InTouch, and KingSCADA). DCS-integrated HMI devices often had more specialized functions and better stability, but they were also more expensive. With the rapid development of PC technology, some general-purpose monitoring software has developed rapidly, and its functions and performance have gradually surpassed the dedicated devices provided by DCS manufacturers. Therefore, many DCS manufacturers have gradually abandoned dedicated human-machine interface devices and instead adopted general-purpose monitoring software, just like PLCs. DCS manufacturers' use of general-purpose monitoring software is not simply a matter of assembling components; rather, it involves collaborative development, preserving and inheriting their years of accumulated network communication and system self-diagnostic technologies in the form of dedicated software packages. For example, the NT6000 initially used the T1000 human-machine interface system based on a dedicated operating system, while currently it primarily uses the T3500 human-machine interface system based on FIX/IFIX or INTOUCH. The LINPOLL network communication package within this system is developed and integrated. 3. Market Situation and Development Direction of DCS and PLCs In the field of thermal automation, the main plant control system almost invariably uses DCS. PLCs are used only in auxiliary workshops. The main reason for this is that early DCS systems were very expensive. It was believed that auxiliary workshop operations could be intermittent, reliability requirements were not very high, and analog control requirements were less stringent. From a cost-reduction perspective, PLCs were often chosen to build the control system. The control systems for boilers, steam turbines, and generators require long-term stable and reliable operation, and their signals contain a significant proportion of analog signals. From a performance perspective, this necessitates the choice of expensive DCS (Distributed Control Systems). Furthermore, analyzing the market competition for main plant DCS and auxiliary workshop control systems reveals an interesting phenomenon. Competition for main plant DCS often occurs between suppliers or agents of different brands, leading to fierce competition and continuous price reductions. In contrast, competition for auxiliary workshop control systems is typically among engineering companies using the same brand of PLCs. The entry barrier is lower, resulting in even fiercer competition, but PLC price reductions are not as significant as for DCS. This is primarily because DCS manufacturers directly participate in the competition, continuously reducing equipment manufacturing and engineering implementation costs under immense market pressure. PLC manufacturers, however, do not directly compete, limiting the scope for engineering companies to lower their own limited engineering costs. Currently, the price difference between DCS and high-end PLCs is no longer significant, and the continued widespread use of PLCs in auxiliary workshops is a result of market inertia. With the continuous expansion of installed capacity in domestic power plants and the advancement of power system reform, the requirements for auxiliary workshop control are also constantly increasing. In this context, the integration of DCS systems into auxiliary workshop control has become a trend. The NT6000DCS, due to its comprehensive technical and economic advantages, has played and will continue to play an increasingly important role in auxiliary workshop control. PLCs, widely used in auxiliary workshops, will not disappear from the historical stage of thermal automation. Unprecedented competitive pressure will prompt PLC manufacturers to converge with DCS standards technically and make greater efforts in pricing. The result of market competition will bring greater benefits to users. 4. Conclusion: As products of the combination of computer technology and control technology, both DCS and PLC have made their own contributions to improving the level of thermal automation in thermal power plants. Due to the significant similarities in their applications, their respective technical or price advantages will directly affect their market position at different times. The market reaction will also be reflected, quickly or slowly, in their respective technological developments and price adjustments. Overall, the integration and promotion of DCS and PLC technologies will be the mainstream of competition, while the continuous improvement in cost-effectiveness will also be the main theme of development.