DCS Systems for Balance of Plant

C

Thread Starter

Chraico

Hello,

I'm a trainee, and i'm working on studying different solutions for instrumentation and control for BOP (balance of plant) and compare between them.

I'm wondering:
- Which DCS systems are better for BOP?
- What about the use of PLC?
- What are the BOP's requirements to consider?

Thank you!
Achraf
 
Achraf

These are good questions, but before you can answer what is best, you need to define what you system is:

1) Are you controlling a full power plant with various sections or are you operating only a small section of the facility?

2) How many I/O count?

3) What are you interfacing to? Are you just interacting with one other control system? will you be interacting with many? RTU stations? Grid control?

4) Do you expect to interact with operators (i.e. do you need HMIs)?

Once you know these answers, you will know more if you need a PLC or DCS based system.

Then, you need to get into the nitty gritty of requirements. This is tough, but you need to consider how and what you are interacting with...protocols that you will need, and possibly any HMI software requirements.

I worked for GE's Automation business for 9 years, so I am biased to the MarkVIe. It's flexible and has great interface abilities to a lot of other systems. However there are some downfalls with the package.

Happy to help out if you need more information

Max
 
Hello Max,

First i want to thank you for your reply and for helping me with your experience.

My project is to study the different control/instrumentation solutions to have a global view about "what to use and in which case?" This concerns DCS systems as well as communication protocols,

Well, my first goal is to do it just for the BOP part of the plant (balance of plant). I was told that actually Mark VIe isn't a really good solution for it, and I'm looking for information about the other DCS systems like Emerson, ABB...

And for your questions:

1) Are you controlling a full power plant with various sections or are you operating only a small section of the facility?

For the moment, just the BOP part.

2) How many I/O count?

This is the most important requirement that I have to consider. I would like to study different cases.

3) What are you interfacing to? Are you just interacting with one other control system? will you be interacting with many? RTU stations? Grid control?

Again I have to study each case here. Most likely more that one control system. For example Mark VIe for ST (steam turbine) and other DCS system (or systems) for BOP and Boiler.

4) Do you expect to interact with operators (i.e. do you need HMIs)?

I think so yes,

Thank you!
Achraf
 
R

RocketTester

>- What about the use of PLC?

Look at Beckhoff Automation, which we use. They have both traditional PLC inputs and more scientific type (100 kSps, 32-bit A/D, accelerometer, resistance, strain bridge). Cost per channel is much less than A-B or Siemens and wiring is simpler, channel configuration flexible, and I/O can be distributed around the plant w/ Ethernet cables (Cu or fiber). We had ~200 I/O channels in one application.

Re HMI, one off-the-shelf tool is Indusoft which we used in one control room, or WonderWare if you want to spend more. If you want to save plus allow customization, search for free AdvancedHMI by Archie on PLCtalk.org. It is Visual Basic.net source code with snappy graphics, which you can use as-is or customize. Interface examples for A-B, Beckhoff, Siemens...
 
Achraf,

again, I will preface that I am biased for the Mark VIe. However, if it was my plant, I would want the entire plant to be 1 control system...if this was possible. There are benefits to having 1 supplier, 1 platform of training, 1 set of spares, & 1 set of HMI screens for a facility. That being said, I would be extremely biased to Mark VIe. The entire systems fits seamlessly together. You do not have to recreate screens, and it should be cheaper at the end of the day. Note: you will pay more money for integration costs and it will take longer to install 2 or more systems.

That being said, here are some other big competitors:
* Emerson Ovation: In the US, they have a very good install base, so there is a lot of information out there accessible to you and your team. Their technology is great for slower process (<100ms) control and their customer service is pretty decent. Because they are mainly DCS, they have developed a lot of interface protocols.

* Siemens S7 platform: Good blend of PLC and DCS. If you have small BOP, this is good for 100-800 IO. Very Flexible. They follow the PLC footprint, so you could buy the hardware yourself and design the system, or you could outsource to an integrator.

*ABB Freelance DCS: I dont know much about this, but ABB has a strong hold on the market as well. Their SCADA package, Symphony is pretty good. You can look into that as well.

Hope this helps. It is not too much detailed information, but gives you are start. As other have said, there are tons of automation platforms out there, the ones I have highlighted are in the top 10.

Max
 
I don't know anything about the control systems on your turbine generators but I imagine they are very fast.

Typical DCS controllers are quite slow, I have seen scan times over 1 second, for control loops they don't need to be fast.

Interlocking on the other hand has to be fast, for that reason it used to be common to have a combination of DCS and PLC with the DCS handling graphics and alarming.
Fisher Delta V is a very good system used throughout the world.
 
I think Max provides a thoughtful and very balanced response here: in both what one should consider when evaluating systems, and the potential control system suppliers that could meet your requirements.

While I don't wish to introduce my Emerson bias, I do want to add that Ovation has specialized cards for handling faster applications like turbine control and generator excitation control.

Our website does not go into deep technical detail, but it does do a fair job summarizing the breadth of applications that Ovation supports as a single-platform solution for power plants. www.emerson.com/ovation

I'm sure the other supplier websites offer similar summaries.
 
Achraf

Depending on the Needs, Foxboro-Schneider did most of BOP before some of these other Companies new what it was. They can Interface with anyone. I have their DCS on 1 Coal Unit and 1 on a HRSG Combined Cycle Unit connected to a GE MKVI Unit. Also on some older equipment that still runs occasionally. Cyber Security, Maintained Website. Siemens is German and their standard answer is this is the way we do it in Europe. Delta V is a lot of Visual Basic Code. Emerson is Westinghouse and they sold their Gas Turbine business to Siemens. All DCS Systems will be able to do anything you want. Just make sure you know what you want and is the DCS Customizing capable (Graphics, Blocks, etc..) Some use Templates only. So you have many decisions to make. Talk to Foxboro-Schneider or Emerson and then talk to the Local I/C Tech's using it. Other site Visits will help also. See what they have. They will give you the real lowdown as opposed to a Sales Person who will tell you they can do anything you want. They will promise you the Moon and usually deliver just the Craters.

Good Luck and Happy Controlling...!!!!
 
B

Bob Peterson

There is little difference in the capability of DCS and PLC systems these days. Pick what makes sense for you. I will say this about DCS systems. It appears that the hardware and software prices are insane compared to PLC. The system integrators that work on DCS systems also charge an enormous amount of money compared to the system integrators that typically work on PLC systems.

These days DCS systems will talk to just about any PLC system so you can still use the DCS console as the plant alarming system.

I don't have any real wisdom for you beyond it is about cost and configurability. PlCs tend to be easier to do certain types of control and DCS seems to do other types of control a little better. PlCs are better if you have speed requirements.
 
One last piece of advice...Look at the ENTIRE cost of the project. Don't just compare PLCs with DCS. Many people compare just the hardware costs when evaluating systems. There are a lot more 'gotchyas' when only looking at hardware pricing. i.e.

* PLCs usually require marshaling and DCS systems sometimes don't. the marshaling and cabling work can add a lot.

* There are often hidden costs for add on point counts or server connections.

* Integration work/costs can vary per system and per integrator. Make sure this is captured somehow in the integrator's scope of work.

* also evaluate competency level of the integrator & the specific person working on the project. I don't care if the hardware is expensive. as long as the worker is smart and willing to learn.

Max
 
With all of the different combinations of upgrading DCS/turbine controls/PLCs we opted to use a firm, RoviSys, to help us with the upgrading decision making process. We provided them with the information they required. After that, they did all of the leg work.

They brought in vendors to demo products that fit our needs. They gathered pricing, options and bundled it up into a nice package that we can select from. They also provided pricing options for doing upgrades over an extended period vs. all at once.

We are considering several different levels of replacement. Here are a few examples we are considering.
1. Update existing system to latest version
2. Replace BOP and/or turbine controls with two different platforms
3. Combine DCS & turbine controls on one common platform
4. Update PLCs to latest versions
5. Replace PLCs with remote IO
 
Top