Talk:Distributed control system

History

Electronic digital based DCS goes back to circa 1974 with Honeywell (and others) launching digital DCS eg TDC200. BBC (Brown Boveri et Cie now part of ABB) also had a digital DCS. By the 1990's most large industrial plants had replaced the majority of pneumatic and hydraulic control systems. In the early 2000's most Industrial DCS's were ported to Windows (R) OS platforms after presure from cost cutting and emmbedded IT infrastructure.

Not every distributed control system is part of a manufacturing system. Big ships have them; jet planes have them, etc.

DCS and SCADA

somehow i belive that SCADA is the new word and DCS was mostly used until the early 90`s if you take a look at the homepages of different providers, all are calling their system SCADA and somewhere in the text DCS popps up.

the differences are minute and the article should be merged with SCAD

SCADA I usually think of as related to controlling something that's spread out over a wide geographic area, like a power grid. Distributed control I usually think is spread over at most a plant site, and "closes the loop" for process control - SCADA systems to my way of thinking just monitor or alarm, and leave loop control to local elements. Both terms are still in use and I think they don't entirely overlap - I'd object to merging the articles. (You can create a user ID and sign your remarks!) --Wtshymanski 22:17, 30 May 2006 (UTC)

DCS is NOT SCADA anymore than an analogue computer is a digital computer !

DCS = top down design solution to factory automation : hence Distributed control system; as it is integrated at the highest factory wide production level; it includes alarming; recording; control and monitoring complete with an integrated graphical front end (HMI or Human Machine Interface). Minimal number of diferent tools to configure; normally everything runs under a common control system (at computer level) SCADA = Supervisory Control And Data Acquisition - normally displayed using a (third party) Human Machine Interface. In any given plant (factory) there may be several diferent SCADA systems - sometimes from different suppliers. PLC - individual controls - commonly used as stand alone 'skids' that control logical sub-units of plant - frequently now attached to a SCADA front end to give central control. By default this is bottom up design.

DCS engineers frequently deal with factory wide apllications; PLC engineers frequently deal with single machine/unit control.

With the drive to more and more open standards (cf OPC ) DCS suppliers have had to make their previously be-spoke systems more and more open. At the same time the need to integrate at the lower machine level is driving PLC suppliers to offer more and more integrated systems. Resulting in an ever narrowing gap between top end PLC solutions and bottom end DCS.

The main differential therefor is that DCS requires a top down factory/plant wide design with as little interaction with low level control as possible whilest SCADA/PLC continues to be a bottom up desgin more concerned with intimate dealings with 'bits & bytes' level of machine control. Petedtm 20:49, 9 August 2006 (UTC)

SCADA can be and is increasingly a sub-system or process I/O data provider of a DCS system. The DCS system, one performing process monitoring and closed loop control (Set Point - Process Value = Error --> output), indeed will include field instrument inputs as well as SCADA system inputs. Control points may be comprised of but not limited to one type or a combination of field input/ output instrumentation, OPC (OLE for Process Control) points, SCADA points and human input. DCS systems, systems intended to monitor and control processes like petroleum refining, chemical processes and raw material production (paper, metal foil, etc), typically include hardware (controllers, SCADA, PLC or even other DCS systems) to interface with real world devices (pumps, switches, valves, fans, etc). The controllers are compact computers system in their own right but offer limited or no human access without use of a programming computer networked to the controller and acting as a terminal or station. The DCS more often today represents the front end of associated controllers, PLCs or SCADA devices and may be used to modify SCADA, PLC and controller control routines and I/O module assignments.[CDJ, 2215 GMT-5 21NOV2006] Ma ne maoj

DCS/PLC/SCADA not comparable

There appears to be much confusion between the definitions and the common usage of these terms. The architectural distinction between a global controller with complete knowledge and local controls is the most significant. Collection of historical data is entirely orthogonal to the control philosophy and should be clearly identified as such. Safety is an important aspect of the topic that is covered, but not clearly described.

PLC describes a hardware/software device that replaces hard-wired relay logic panels. The name is a trademark of Allen-Bradley, and should probably not be used in objective descriptions of programmable controllers. Such controllers can be deployed either centrally or in a distributed landscape, depending on the installation and available technology. They have traditionally (as identified in the article and the talk page) been used in discrete facility control, often (but not necessarily) in smaller installations. Generally, machines with automatic controls have a supervisory safety control defined independently of the normal operation. This may, or may not, be implemented in the same hardware.

Distributed control systems (DCS), in contrast, are characterised by large numbers of controllers responsible individually for the control of small numbers of (typically continuous process) variables with quite localised process loops across spatially distributed facilities with often safety critical aspects. Such facilities are often described in process and instrumentation (P&ID) diagrams and can be controlled without global knowledge. Safety is typically provided through low-level redundancy, although nuclear plants often require independent parallel implementations for assured operation.

SCADA was a marketing pitch applied when software-only solutions to replacements for machine and process mimics became available. The important first word is Supervisory. SCADA are not intended to control a process. They are meant to allow operators to adjust set-points (DCS-speak) or modify tag values (PLC-speak) of the underlying control system. They additionally record process history. Thus, a classic DCS like the Valmet DAMATIC incorporates controllers and SCADA functionality. This is the direct result of introducing microprocessor controllers into a process control environment with computerised control rooms. Early SCADA solutions were implemented to provide DCS level historian functionality in programmable controller landscapes that had previously not incorporated large persistent storage.

The split between centralised logic controllers and DCS has mostly blurred to the point of unrecognizability and for example the latest process control system from Siemens, PCS, utilises centralised programmable controllers as the computational element. This requires that the edge devices now independently provide the necessary safety functions that protect the process from either controller or communication failures.

The confusion results from a difference in design approach between process and discrete engineering approaches, historical architectural preferences of practitioners, technology limitations that are now largely resolved and many vested commercial interests protecting turf.

I'd suggest a clearer separation between the typical intent of DCS of controlling many independent loops familiar to process engineers and the classic discrete control provided by programmable controllers. The actual implementation is, and should be described as largely orthogonal to the intent, and SCADA relegated to its peripheral position in the discussion. I'm not offering to do this as I anticipate repeated reversion and refuse to get embroiled in the religious discussions that will follow.

--User:pekkapihlajasaari 26 December 2019 —Preceding undated comment added 14:45, 26 December 2019 (UTC)

Marketing speak

There's a little too much gee-whizzery in the article just now. It needs to be made more encyclopediac in tone. It needs more comparision with other techniques of process control. Some neutral history would always be good, as would be a block diagram. Is there now a fundamental difference between SCADA and DCS? Was there? I think of SCADA operating over dispersed geographic areas and DCS within a plant, but there's not much fundamental difference in what you see when you look at the boxes on the plant floor. --Wtshymanski 22:30, 10 September 2007 (UTC)

A Potential Explanation of the Confusion

Most of these comments are trying to explain an evolution of different systems and terminology that has taken place over three or four decades. This can be very confusing without a broad historic perspective . For example, there are Distributed Control Systems in the architecture sense in airplanes, but it would be generally agreed that aircraft do not have DCS systems. DCS systems generally refer to the particular functional distributed control system design that exist in Industrial Process plants (Oil & Gas, Refining, Chemical, Pharmaceutical, some Food & Beverage, Water & Wastewater, Utility Power, Mining, Metals, etc...).. Although PLC (Programmable Logic Controller) systems can be considered distributed, they are not DCS systems in the sense that they are generally applied to Discrete Factory Automation applications.. (Plants that make "Things" rather than "Stuff"). SCADA's history is rooted in pipeline and power distribution applications where there was a need to gather a lot of remote data and do a little bit of control with sites that were widely geographically separated. Each of these types of systems have their particular functional needs that are different. DCS systems used in hazardous explosive environments in Oil production have different requirements than PLC based systems that are controlling the manufacturing and packaging of auto parts for example. Although what you might see in the plants as Boxes that look alike, are in fact quite different in a bunch of ways. As proof of this, companies who are strong on the Process side like Emerson and Foxboro/invensys have had difficulty trying to leverage their systems into Factory Automation. Rockwell Automation, the US leader in PLCs has always wanted to leverage their strength in Factory Automation into the process side, but similarly has not been able to do that.

Having said all of this, the boundaries between these systems are becoming more gray as time goes on and technology provides more and more bridges between them --Mike (talk) 23:23, 2 April 2008 (UTC)

The Root of the differences

The primary differences between PLC control (monitored by SCADA or not) and DCS control is about the function of the processing of data. PLC scan rates are faster due to the programming residing in a single place: the central processor. While DCS is just as described: distributed control. The DCS scan rates are higher, because data is processed in various places and then sent to other places for control or monitoring. If the PLC processor fails (without redundancy) all control is lost and the PLC will set itself to a fail safe state, i.e. all outputs go to safe state. DCS has many controllers for various areas of control, hence the name distributed. A single controller failure (without redundancy) will cost the loss of part of a process but not necessarily all.

SCADA, though it's name implies control, is more often used as data aquisistion. A DCS system can act as a SCADA in relation to a PLC. The lines between DCS and SCADA are indeed more blurred than ever before. —Preceding unsigned comment added by 158.35.225.231 (talk) 12:30, 18 April 2010 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Distributed control system. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20070329065215/http://pcs.khe.siemens.com:80/index.aspx?nr=1075 to http://pcs.khe.siemens.com/index.aspx?nr=1075

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 01:12, 14 December 2016 (UTC)