What is a DCS (distributed control system)?
What is a DCS (distributed control system)? Ever heard of a distributed control system (DCS)? Even if you haven’t, you
What is a DCS (distributed control system)?
Ever heard of a distributed control system (DCS)? Even if you haven’t, you can probably figure out what it does by its name.
In process automation, a DCS is just one of your options when it comes to control systems. In school, you hear a lot about PLCs (programmable logic controllers) and SCADA (supervisory control and data acquisition) systems. But when you have a plant with thousands of field devices, managing that many PLCs can get complex.
Thus we have the DCS, vastly more scalable than a PLC. But we’ll get to that in a bit. For now, let’s focus on understanding what exactly a DCS is.
As I said earlier, the name says it all. Now we need to talk about how it does it all. Let’s define our parameters here. A DCS connects a central controller to a number of automated local controllers all over a process facility via a high-speed communication network.
These local controllers in turn connect the central controller to your field devices. This network also connects controllers and devices to supervisory terminals, operator terminals, and historians.
Fieldbuses and standard communication protocols make all this connection possible. You can use HART, PROFIBUS, Modbus, FOUNDATION Fieldbus (FF), and others.
This system works best for large-scale processes or manufacturing plants, where you have a massive number of continuous loops to monitor and control. It also works well for systems that can’t afford downtime, as the rest of the plant can keep working even if a section goes offline.
Now that we have an overall picture of a DCS, let’s have a closer look at its architecture. We can break it down into four basic elements:
- engineering controller
- operating station
- local controller
- communication media
Let’s check them out.
The engineering controller supervises the entire DCS. You can use a PC or any other computer with some sort of dedicated engineer software installed on it.
Your configuration tools live here too. You can execute a control algorithm, create new loops, configure devices, and more. It basically acts as the brain of the whole process.
This station monitors, operates, and controls the plant parameters. Here you can set alarms, for example. The data logging takes place here as well. With this logged data, you can track trends in your control system.
Just like the engineering station, you can use nearly any computer as long as it has the right software installed, such as ABB’s DigiVis. You can also have multiple PCs with specific functions designated to them, like one for data logging and alarms and another for parameters and trends.
The local or distributed controller consists of a CPU module, fieldbus, or communication module with remote or direct connected input/outputs (I/Os.) It pulls data from field devices through an input module, analyzes the process with its algorithms, and controls the output signal via the output modules.
Your field devices connect to this unit, which has one or multiple I/O units. If you have smart devices and a fieldbus connection such as PROFIBUS, you can even skip the I/O module and connect directly to the fieldbus.
Communication media covers all the components that connect the other three elements of the DCS, like wires and cables as well as protocols. Of course, in this day and age you may have a wireless system, but you get the point, right?
The communication system carries data from one station to another, and you have a variety of protocols to choose from – PROFIBUS, Ethernet, DeviceNet, FF, etc.
You should also know that you can use more than one protocol in your DCS. You can have, for instance, I/O modules and process stations connected with PROFIBUS and Ethernet for your engineering controller and operating stations.
DCS vs PLC
OK, now we know how a DCS works, its elements and everything. But when should you use a DCS instead of a PLC system? Let’s compare the two systems so you can decide which will work best for your process.
PLCs have faster response times than DCSs, no question. If you want real-time control and actions in your process, then you want a PLC. A DCS will take much longer to process data.
Even though a PLC can handle many I/O points, a DCS can handle many thousands of I/O points. With a modular or subsystem architecture, you can add new equipment to a DCS all day long.
So if you have a large facility with parts of your process strewn throughout, then a DCS will assist you better.
Here again the DCS has an advantage over a PLC. You can slap in as many controllers and networks as you can afford. And when a problem arises, it won’t affect your monitoring and control, if you have your redundant devices set properly.
This feature makes your system more reliable, since it can keep running despite issues.
No surprise here, right? The more complex your system gets, the more likely you are to choose a DCS over a PLC. A DCS allows multiple procedures and products as well as multiple devices in your system. So if you have a lot to do, let a DCS do some – or most – of it for you!