› PLCGurus.NET Live & Interactive Forum › PLC Questions and Answers › General Automation Questions › System Startup Problem 3 PID loops
Tagged: Feed Forward
- This topic has 8 replies, 2 voices, and was last updated 4 years, 10 months ago by
Anonymous.
- AuthorPosts
- February 11, 2019 at 2:31 pm #3717
Anonymous
Karma: 0Rank: PadawanI have a startup issue in need of a solution, the problem involves 3 valves on a distillation system. The three valves are a evaporator feed water level valve, evaporator feed water temperature valve and the evaporator pressure valve. The level and pressure PID’s are fast loops the temperature is a slow loop all are tuned for steady state running of the system. The problem that is seen during startup is as follows the feed level drops, causing the level valve to open to maintain the running water level, because of the slow tuning of the temperature valve it can’t get the temperature of the water up to setpoint fast enough. The evaporator pressure valve can’t maintain the pressure due to the influx of cooler water, this causes the pressure to drop drop rapidly and shut the system down.
The old control philosophy was to set a configurable manual CV for the temperature valve which would would be held for a configurable time. This would ensure that the temperature valve would not close during startup of the system. The percents open set point varied from unit to unit and site to site depending on the feed water temperature, the set point of the timer also varied it was all pretty much a guessing game.
This philosophy also causes another problem if the site has seasonal water temperature changes if the feed water is too hot the pressure will overshoot and shut the system down.
I have started to use the feed forward of the PID to assist the temperature and pressure valve. For the pressure valve I use the percent error of the level valve as the feed forward control value. For the feed temperature I use whichever percent error is greater of the level and pressure valves. Because of the nature of the percent error once the system is stable the value is near zero and has little effect on the loop it is being used as the feed forward on.
Using the PID feed forward, assist the temperature and pressure valves greatly during startup the response time is much faster and the process variables reach the set point quickly. Unfortunately this did not eliminate the need for a configurable startup CV at sites with seasonally cold water, I ended up have to add a manual CV that is sent to the feed temperature valve during startup to kick it open to ensure the system would startup.
What I’m looking to do is get away from the guessing game of a startup CV. I do not want 2 different tuning parameters either since startup only last about 5 minutes.
Any ideas would be greatly appreciated.
February 11, 2019 at 5:13 pm #3719whiskyjim
ModeratorKarma: 262Rank: JediWould it not be possible to run the PID that controls the feed temperature valve in Override or Hand mode with a set CV, while monitoring the water temperature then switch back to Auto mode once you’ve achieved the water temperature you need for successful start up?
February 11, 2019 at 6:50 pm #3720Anonymous
Karma: 0Rank: Padawan<p style=”text-align: left;”>Something like that been tried in the past, problem is when the system sits in standby for extended periods of time or is quickly restated after having ran the temperature can be over the set point which would negate the CV hold. Another control philosophy that has been tried was limiting the close possition of the valve depending on how open the level valve was. Problem is we still end up with a guessing game of what % open CV to use for the temperature control valve, which is what I’m trying to get away from. I also have to be careful not to overheat the feed water if this is done I basically will be feeding steam instead of water into the system which causes other problems.</p>
Any experience with using ratio mode in PIDE’sFebruary 11, 2019 at 7:01 pm #3721whiskyjim
ModeratorKarma: 262Rank: JediI’m having some trouble visualizing your set up. Would it be possible to provide a sketch of the physical set up showing the valves, etc.?
February 11, 2019 at 7:21 pm #3722Anonymous
Karma: 0Rank: PadawanSee attached
February 11, 2019 at 8:20 pm #3723Anonymous
Karma: 0Rank: PadawanLet’s try again
February 11, 2019 at 9:28 pm #3724Anonymous
Karma: 0Rank: PadawanAgain
February 12, 2019 at 4:22 pm #3727whiskyjim
ModeratorKarma: 262Rank: JediJust so I’m clear on the concept, you are running water through a heat exchanger into the tank. That water is being heated to a specific temperature using steam in the heat exchanger. The level of water in the tank is being monitored so as not to let the level get below the level of a steam sparger that is located in the tank. You are also using live steam to heat the water in the tank in order to maintain a specific pressure in the tank.
Does that cover it?
February 12, 2019 at 9:24 pm #3728Anonymous
Karma: 0Rank: PadawanYou got it that is the basics of the system that I’m working with.
- AuthorPosts
- You must be logged in to reply to this topic.