The Good, The Bad, and The Ugly of Refinery Advanced Controls | RefinerLink
Cor
Cor

RL Blogs

image
The Good, The Bad, and The Ugly of Refinery Advanced Controls

By Steve Pagani

Jul 27, 2015
 

An up-front and truthful summary of the pros and cons that come with every advanced control application.

 
 

One can argue that the development of console advanced controls is the best technological breakthrough for the oil refining industry.  From safety improvement to profit improvement, it's hard to imagine running a refinery today without the use of advanced controls. 

 

However, every diamond has a flaw, so here are my views of the pros and cons of every advanced control application. 


 

 

 

Optimize Everything at Once (The Good)

 

God bless engineers. They spent years asking for a slight move on a flow here, an increase in temperature there, and a reduction in pressure here and there. Day after day they walked into refineries and tried to optimize what they could one step at a time.

 

But that was likely the frustrating part. By the time you optimize the first area and move on to the next, the first isn’t optimized anymore because someone wasn’t watching it carefully enough.


So what did those engineers do? They went out and came up with Advanced Controls; little computer programs that have a never ending quest to optimize process variables.

 

These Advanced Controls not only figured out what the optimal move was but it also executed the move. It was like you’d stuffed a thousand nagging engineers and compliant operators all into one console… and they worked together in perfect harmony!

 

If you’re an engineer and have certain parameters that need to be manipulated in order to control another parameter stop reading this and go put some advanced controls on that thing! From the very complicated to the very simple, advanced controls can make life unbelievably easy for a console operator and at the same time make an engineer giddy with pride as they watch the unit being optimized every minute.

 

But wait… nothing can be that perfect can it? No, it can’t. You got me. There is a bad side to this equation.

 

 

In Computers We Trust (The Bad)

 

So you read the first section of this and immediately rushed out and put advanced controls on everything imaginable. Great work! Your job is done. Now sit back, relax, and watch a few episodes of Modern Marvels. And while you’re at it, let your console operator know he can do the same (maybe a different show for him though).

 

That’s where the problem lies. Too many times Advanced Controls cause engineers and console operators to stop questioning. They trust the

 

 

computer because after all, how many times does a computer give you the wrong answer?

 

Well, the little computer programs are fed by humans so that means there will always be issues. 

 

It’s possible the limits on certain variables are over-constrained or the pricing set that’s driving an optimal solution is incorrect.

 

Maybe the model itself is

outdated because it hasn’t been updated after a major change in the unit. The moves that are made by the Advanced Controls should always be questioned and investigated in detail. Don’t put blind faith in that computer.

 

 

Your Advanced Controls Engineer (The Ugly)

 

Yeah I said it. Chances are that you’ve got that nerdy Advanced Controls guy lurking around the refinery somewhere. And I’m willing to bet that he or she is ugly and probably a little awkward to talk to. So what? There’s likely also some level of genius and passion for their work inside them as well.

 

Whether you’re a process engineer or a console operator, you need to go talk to that Advanced Controls Engineer.

 

Understand what the model is based on and why it’s making the moves it’s making. Talk about other possible applications that can be implemented to further optimize the process or just make things easier to control.

 

At many refineries these guys are also the most under-appreciated people there.  So make sure to say thanks every now and then too, because they play a huge role in increasing refinery profitability

 

 

 
Enjoy this content? Join our Free Newsletter    
  • Karl Kolmetz :   We previously had a good group discussion on APC - if you do not keep the analyzers working it will be mostly be the bad and the ugly

    Jul 28, 2015

  • Greg Dunnells :   Its like everything else, if the benefit is there, then the cost of maintaining the basic controls in working condition is paid for. On the other hand, if someone is not going to spend the money to keep the basic controls working, then they are probably not going to spend additional money to develop/maintain APC either, so why bother? If the applications selected are worth significant benefits, then the cost should be worth it. First step - do a cost/benefits analysis, then get management buy-in and establish a realistic budget. If they won't give you a budget, drop it and never look back........ Regarding the earlier discussion on heater effluent analyzers, I worked with a client recently on fixing their system. Most of what was wrong was incorrectly specified flowmeters, and technicians that had absolutely no idea what they were doing. We got one working, it literally took a couple thousand dollars worth of parts, and a couple of meetings to educate everyone on how the system was supposed to work, and why what they had, didn't. In the end, it was not sustainable as they had only a couple of people looking after 200 or so analyzers and conditioners. Complete waste of money. Millions and millions of dollars worth of analyzers with the effective value of scrap iron.

    Jul 28, 2015

  • Kun-Chuo Chen :   Unfortunately analyzers always have troubles and are inaccurate. It would be better to develop soft sensors for APC. This is my experience.

    Jul 28, 2015

  • Jaco Nel :   I've seen how a well built APC resulted in operator proficiency drop up to a point where they could barely control the unit themselves. The APC was on line for 400 days straight in one instance. Our challenge now is to find a balance between maximum APC on line time and maintaining operator proficiency

    Aug 02, 2015

  • Chris Hotblack :   We have a customer with the same problem - maintaining operator proficiency because their APC and RTO work so well. We believe an accurate easy to maintain OTS is the top priority - with a well designed and implemented Training Programme. He already has an OTS but it is not very accurate and is difficult to maintain.

    Aug 12, 2015

  • John van Ladesteijn :   I have seen a group of 3 Advanced control engineers, struggling for a week with the optimization program. until I (panel operator ) overheard them talking about the problem (optimizing the crude heater ) I told them the density analyzer of the fuel-gas had a problem ,and maintenance was waiting for spare parts. Communicating with panel operators can have some value. I have worked for 35 years on a refinery ,alternating as outside ,and panel operator, but mostly as panel operator , so I have controlled all kind of units without any advanced control. so slowly I learned from the beginning of advanced control until it was fully implemented. Now the new generation of panel operators have never operated a unit without advanced control, that gives a huge problem when the Advanced control program trips , they have no idea how to safely get a unit in balance . This is especially with towers which are controlled against the puking point ,and then fail just at this point, or furnaces who go to the minimum of oxygen . So learning on a simulator ,start up and then taking the Advanced control program on line is the only safe education for operators.

    May 08, 2016

  • Steve Boyle :   John, you raise a good point about console operators becoming dependent on controllers. It is scary to the point where many don't know how to manage their units when upset conditions occur. Simulators is a good way to get people trained up, but it only goes so far.

    May 10, 2016

Do NOT follow this link or you will be banned from the site!

Blog Topics

 

People You May Know