A distributed control system, or DCS, is essentially a control system where the control elements are geographically separated (distributed) over the control area (i.e. a plant), hence the name distributed control system.
On the other hand, a centralized control system offers a single controller at one (central) location that handles all the control functions.
In a Distributed Control System one process element (devices, group of devices, a system) is controlled by one dedicated controller, so the DCS can consist of a large number of controllers in various locations of the control area, typically connected via a high-speed network.
The controllers are able to communicate between themselves and possibly with other controllers outside the DCS like operator terminals, supervisory terminals, and so on.
In practice, the individual controllers are connected to field devices like actuators and sensors, with these controllers maintaining the transmission of collected data to other hierarchical controllers by utilizing different protocols.
Various communication protocols or field buses can be used for establishing the communications between these controllers, including but not limited to HART, Modbus, arc net, and Profibus.
With the distributed nature, DCS is more suited for large-scale applications for example in large factories and manufacturing plants where a large number of continuous control loops need to be maintained and controlled continuously.
This is due to the main advantage of the DCS – if one controller fails, then only the element/section associated with the controller will stop working while other sections of the factory can continue to operate.
As discussed above, a distributed control system has the control process distributed throughout the system instead of involving a central mechanism with a central controller.
Essentially a DCS divides the controlling tasks among multiple distributed controllers (such as PLCs).
The Distributed Control System (DCS) consists of four different interfaces:
Field Control Station
The control station(s) receive signals from sensors to track various aspects of the process (temperature, flow rate, pressure) and will perform required calculations, mainly to compare the signals from sensors with benchmark values.
Output signals after these calculations are sent to the final control element to perform the desired actions.
Distributed Interface/Control Unit
The control sub-system interface connects the distributed control system to other instruments, such as PLCs to integrate the factory/plant operation.
The local control units can be connected directly to the field devices (input sensors and output actuators), or placed in different locations and connected to the field devices via communication links.
These control units can receive and control both digital and analog inputs/outputs by utilizing I/O modules (both analog and digital). These I/O modules are extendable according to the required number of inputs and output.
Human-Machine Interface Station
The interface between the DCS and the human operator. The main function of this station is to perform central monitoring of the system and allows the human operator to provide instructions.
Typically consists of a monitor, a keyboard and mouse interface, and other typical elements of a PC station.
There can be several different HMIs in a DCS implementation, for example, one can be used only to monitor operational parameters while another is utilized for alarming purposes.
Communication Bus and Protocols
Communication buses are used to allow communications between the Human-Machine Interface station and control sub-system interface.
In practice, communication buses can include transmission cables like fiber optic or coax cables, but nowadays it can also be wireless.
The communication protocols are selected depending on the number of devices to be controlled in the DCS. The communication protocols can be Ethernet, CAN, Modbus, and so on.
In a typical DCS two or more communication protocols can be used for different areas. For example, can be used for the communication between control devices and distributed controllers, and another one between the controllers and the control stations.
In most applications, the DCS system is divided into five different levels (level 0 through 4), as we can see in the image below.
Also called the field device level, includes all the field devices involved in the DCS system like sensors, transmitters, control valves, and others.
In this level input devices (i.e. a sensor) and output devices (i.e. an actuator) are connected to the I/O units. The I/O units convert the received signals to a specially coded signal understood by the Field Bus while also converting the coded signal to 4-20 mA (digital signals).
The direct control level where the microcontroller takes data from the Field Bus to control different control functions. All control actions are performed at this level.
The Field Bus contains all the data required for each loop input and output and the Control Panel controls each loop according to this data.
This allows all the loops to be seemingly controlled at the same time, while actually the control processes are distributed and each takes milliseconds of time between each other.
The plant supervisory level.
All the data related to the control loop is displayed in this unit with a display unit. A human operator can then make adjustments via interfaces like a keyboard, mouse, and video display to adjust various processes being controlled and monitored by the DCS.
The production control level. Control engineers can implement advanced control functions at this level. Various crucial management systems in a plant like inventory control, billing, and quality control exist at this level.
The highest level of the process, the production scheduling, or group management level. Various methods can be used here, for example sending some data to a separated HQ office via satellite.
The workstation at the HQ office cannot control changes at the plant level, but will instead use updated information on production purposes for future planning.
Distributed Control Systems (DCSs) are at the moment used in a wide variety of industries like in chemical plants, electric plants, control radio, traffic control, and so on.
Basically, a DCS can be implemented in any control applications where different devices need to be controlled to potentially achieve different objectives at any given time.
In electrical industries, for example, DCS can be utilized to control the different electrical equipment, each with different desired power output and installed in different locations.
Processes where Distributed Control System might be used include, but not limited to:
While, as mentioned, a Distributed Control System can include PLCs in the control system, there are several core differences between the two controllers. You might check out our full guide on the differences between DCS and PLC control systems here.
However, below we will discuss a basic overview of the topic.
PLCs, in principle, are really good at handling repetitive and discrete control over single processes. So, they are typically used for single batch or high-speed control with their simple and low-cost, versatile design. PLCs are generic but are completely customizable.
A DCS, on the other hand, is used for continuous, complex controls with an integrated control center.
So, as a general rule of thumb, if the application requires fast control with discrete I/O, PLC is the better choice. Also worth noting is the fact that PLCs offer more granularity in I/O modules with easier maintenance.
On the other hand, for a plant-wide control involving different inputs, each with different desired outputs, DCSs are better with their built-in infrastructure.
As discussed, DCSs can also incorporate PLCs (making it a hybrid system) to control specific functions that demand speed, and also to provide better reporting.
A major difference between the DCS and PLC is the database. That is, in a DCS the engineering work like programming, reporting, and so on can be executed in a single database, while in a PLC environment different databases are required to carry out each engineering work.
Below are some of the important differences between DCSs and PLCs:
|Existing function blocks used to build custom logic||Programming||Custom logic created from high-level programming languages|
|Many complex algorithms and do not vary in different applications||Customized routines in different applications||Typically require customized routines|
|Standard libraries like function blocks and faceplates are expected.||Libraries||Standard libraries are not mandatory (typically extra features)|
|The entire system is expected to function as an integrated solution||Function||Require provisions to integrate different products into an integrated solution|
|Redundancy is commonly required||Redundancy||Redundancy typically not required and typically not cost-effective|
|Simple to advanced PID control up to Advanced Process Control||Analog Control||Simple PID only|
|Asset management will alert you before something breaks||Troubleshooting||Diagnostics, will tell you only when something is already broken|
|Designed to be easy to use but not versatile/customizable||Versatility||Designed to be versatile|
There are, in general, five different factors to consider when choosing between PLCs and DCSs in your process:
PLCs are faster and capable of doing rapid control. This makes PLCs the better option for real-time controls like firing control or safety shutdown. A DCS, on the other hand, takes much longer to process the data.
In short, if the application requires rapid response time, a PLC is the better bet.
DCSs are much more capable of carrying out complex and advanced process control capabilities including but not limited to water treatment and chemical plants.
A PLC can only integrate a few thousand I/O modules, while a DCS can handle many more I/O points and is more versatile in handling new equipment onboarding and data integration.
When advanced control is required, DCS is typically preferred, especially in applications where the plant is spread out over a large geographic area with a lot of I/O modules.
Frequent Process Changes
PLCs are used for processes that are relatively rigid and won’t change often. The DCS is the better solution when the process requires versatility and frequent adjustments.
For additional reading, be sure to check out some of these articles:
For many people, the terms electrical engineer vs electrician can seem identical, but there are…