What Is DCS? (Distributed Control System)

A Distributed Control System, or DCS, is a system that is designed to control complex industrial processes.

By the end of the article, you will know:

• What a DCS is and why it’s used in the process industry,
• The key differences between a DCS and a PLC control system, and
• The main components that make up a DCS.

Let’s first clarify, for the purpose of this lesson, what we mean by DCS.

What is a Distributed Control System?

Over the years, the term DCS has evolved from the original description for the acronym as a “Distributed Control System” to the use of the term “Decentralized Control System,” and they seem to be somewhat interchangeable nowadays.

Regardless of which description is used, we are discussing a structure that, at the high-level view, is a system that coordinates and supervises an entire plant of many varying processes.

What is process control?

A Process is the transformation of raw materials into end products through operations like mixing, heating, cooling, filtering, and storing. These raw materials are typically in liquid, gas, or slurry form, and they travel through pipes to specifically engineered vessels where operations are carried out.

A process is monitored by instrumentation. These are sensors that check temperatures, pressures, flow rates, and other important variables to verify that the conditions are correct for the process to work as designed.

Process control involves maintaining setpoints for process variables. If any of the sensors detect a deviation from a setpoint, then a process controller responds by adjusting actuators like pumps, valves, and heaters to bring conditions back to the required range.

Modern industrial processes are controlled by a Distributed Control System, or DCS.

An Overview of DCS and PLC Systems

Briefly, as a point of the historical review, when PLCs were invented, they were really good at handling single processes.

A PLC control system is typically used for high-speed, discrete, repetitive, and deterministic control. Think of an automotive assembly line where there is a fixed sequence that involves turning outputs on or off, and the control needs to execute in precise, predictable cycles.

A PLC system is built up from components based on an application’s requirements. An engineer selects a processor, IO modules, network equipment, and an HMI and integrates them into a working system.

PLCs are typically programmed using the Ladder Diagram programming language.

The advent of the DCS was for controlling many autonomous controllers that handled many continuous operations, mainly using analog control.

Over time and through innovation, the lines between the two systems have blurred, but each still has some principal differences.

PLCs, traditionally, were used for single batch or high-speed control, have a relatively simple, low-cost design, and are the core of the system. Their design is flexible and generic but completely customizable.

Task processing times are typically very fast; operators usually interact with and control the system using some type of graphical display, such as SCADA.

A DCS is designed for the complex analog control that is required to keep continuous or batch processes under control.

Since they are used for continuous and batch processes they are typically found in chemical manufacturing, pharmaceutical manufacturing, and oil and gas refineries. In these environments, the control system needs to monitor analog variables like temperature, flow, and pressure and control hundreds or thousands of control loops simultaneously.

Because a DCS needs to have precise, characterized performance, it is purchased as a complete system from a single vendor.

Rather than programming in Ladder Diagram, engineers configured DCS control strategies using a library of function blocks that allow fast, repeatable, and precise engineering of a system.

Processing times are somewhat slower. Operators interact with the control system via an integrated graphical display.

DCS also have a claim that when safety is a top priority, it is the most reliable system.

The reason for this is because the manufacturer supplies both the control and supervisory equipment as an integrated package, the risks of integration errors are greatly reduced.

There are indeed scenarios in which a PLC system would be the best option, such as smaller-sized processes where you could employ redundant components to negate the possibility of process shutdowns.

Without redundancy, you risk production halts due to the nature of a single processor controlling an entire plant.

Redundancy may be deployed in either the PLC or DCS applications.

Just as there are circumstances for a PLC system, the use of DCS would be for larger, more complex processes that require a lot of interaction between many processors.

Now that we’ve touched on some of the differences in the systems, let’s focus on the DCS and some of its components.

The DCS is a process-oriented system that uses closed-loop control.

Basic Components of a Typical DCS Network

Operator Stations

A typical plant starts with a centralized operator control center typically called Operator Stations.

Operator Stations, in a DCS, are the heart of the system.

This is where the operator can observe the operations of the plant, view process warnings and alarms, monitor production, and more.

Servers, Archiving and Engineering Computers

The next level of components may contain servers, archiving computers, and engineering stations.

Communications with the Operator Station level is typically Industrial Ethernet.

Servers are used for the collection of data at the processor level.

They are responsible for the data that moves between the Operator Station and the processors on the plant floor.

Archiving computers are used for storing historical data that may be used for trends or compliance.

This historical data is typically stored in a historian; a historian is a time-series database that records tag values over time.

This data can be recalled for product quality tracking, safety audits, process optimization, and regulatory reporting.

Engineering stations are used for creating the projects on which the processes run. This includes;

– Hardware configurations

– Logic for tasks

– Graphical displays for operator interaction

– Administration of all of those tasks through installed software packages.

This is the station that is used to download the projects to the processors and the graphical displays.

Centralized tag database

Although controllers are distributed across the plant, all data is stored in a centralized tag database. This database acts as a single source of truth for every piece of data in the system.

This tag database can be accessed by workstations on the network for monitoring and troubleshooting.

DCS Controllers

At the next level, you have the master controllers that supervise the individual processors as well as I/O modules.

These controllers are also responsible for providing the data to the servers, which in turn, supply the data for the graphical interface.

Industrial Ethernet is typically used for communication with the previous level. Fiber Optic may be used here when Ethernet cabling runs would be too long.

At this level, the processor executes the logic and does what it needs to do in order to control the process.

At this level, the processor executes the logic and does what it needs to do in order to control the process.

In the process industry, most control strategies are built from a standardized library of predefined function blocks. This library approach speeds up the engineering process and reduces the risk of errors in a project.

Field Devices

The next level is the field device level. Communications between this level and the processor level can be nearly any type that may be compatible with the components.

Those include Industrial Ethernet, Profibus DP, EtherCAT, Fiber Optic, or other proprietary communication protocols.

Components at this level would be devices such as transmitters, switches, valves, motors, remote or distributed I/O, etc.

Modern instrumentation devices often do more than sending a single process variable back to a controller. They also send diagnostic data that helps operators to detect equipment issues before they become failures.

Redundant hardware

Some processes are so critical that they cannot stop even if a fault occurs. For example, an unplanned shutdown of a refinery can cost hundreds of thousands of dollars per hour.

To ensure that a process can continue to run even if a fault occurs, many DCSes are built with redundant hardware including redundant controllers, redundant power supplies, and redundant network configurations. By leveraging redundant hardware, plants can ensure that a single fault won’t cause the whole process to stop.

Summary

In short, both PLCs and a DCS have their place in the market today.

PLCs would work best in a small production environment where component failure if no redundant system is deployed, is a small risk for the application, the budget is restricted, or the tasks and I/O count are minimal.

A DCS would be better used in an environment where there are large I/O counts with many continuous processes, a processor failure in one section of the plant is not a problem for production, or risk assessment has determined that an integrated package would be the best option.

The line between the two systems is nearly invisible nowadays and it may not be long before there is no differentiation at all.

Now you know what a Distributed Control System is. Specifically, you learned:

• What process control is and why a DCS is used for process control,
• The main components that make up a DCS, and
• The differences between a DCS and PLC control system.

Modern PLC and DCS systems share many of the same capabilities. For this reason, selecting the right architecture should be based on the requirements of the application rather than relying only on traditional technology categories.

PlantPAx, Rockwell Automation’s DCS, is implemented using Logix 5000 PLCs. This demonstrates how modern PLC technology can be used as part of a DCS architecture.

To learn more, check out our articles on SCADA systems and the differences between SCADA and DCS control systems.