Advanced Control: Tank level control.

[dcooper33]

I'm interested. I've been wanting to learn more about control of integrating processes for awhile now.

 

[Peter Nachtwey]

OK, I will start soon.

 

[meghnesh]

Hi Peter !!

A Tank Level Controller is a simple Water application.
But you can add numerous controls and monitoring making the application worth watching and understanding.
1.Along with a Tank I assume that there will for sure be some pumps that mite fill the tank.Use you Level controller to start-stop these pumps based on Tank Water Level sensed by some Level Sensor.
2.Also if there is a Water Tank there will be some outflow/inflow of the water.Use this inflow/outflow as an Analog Input to your controller to synchronize your pump operation to Save Energy,to avoid Dry-Run of Pumps, to avoid Under/Over Running of the pumps or to prevent Flooding of the tank.
3.Again, if there is a water tank and pumps, depending on the water Coloumn or Level of the water in the tank, calculate the Best effecient Pumps to RUN to reduce maintenance of the pumps and increase the overall effeciency of the system.
4.If there is an Up-stream pumping set or a Down-stream Pumping set, make arrangements for Synchronizing all the Pumping sets depending on the Water Level at Each Pumping Station.
5.You can make Alarming arrangements in your controller itself to WARN and Trip(Turn-OFF)
the pumps. You can SMS the alarms to the controlling Utility or reflect the alarm on an Audible device.

I hope this helps Peter !!!

 

[mvaughan]

Always willing to learn!!

 

[Matchu04]

OK, I will start soon.

 

[Peter Nachtwey]

There hasn't been any response to my questions about writing a differential equations for the tank level control. I was wonder how interested you guys really are.

 

[ndzied1]

I sent a PM ?

 

[V0N_hydro]

Hi Peter,
I've seen your series of notes "Alvin's tank level control" and definitely lost the plot part way through so I'll get the ball rolling with the basics. My application is controlling the head pond of small hydro plants 150kW to 30 MW with surface areas from 21 m^2 to 50,000 m^2.


we know the tank level changes when flow in and flow out are not equal. The change in level depends on the surface area of the tank.

let
Qin = flow in to the tank (m^3/s)
Qout = flow out of the tank (m^3/s)
A = surface area of the tank (m^2)
z = liquid level in the tank (m)
dz/dt = change in level (m/s)

(1) dz/dt = 1/A(Qin-Qout)

If we use a PI controller to control Qout

let
K=controller gain (m^3/s/m)
zref = tank level setpoint (m)
Tr = reset time (s)

(2) Qout = K{ (z-zref) + 1/Tr∫(z-zref)dt }

substituting 2 in to 1

(3) dz/dt = 1/A (Qin - K{ (z-zref) + 1/Tr∫(z-zref)dt })

differentiating the (3) and multiplying both sides by A, the differential equation for the controlled system is

(4) A*d2z/dt = d/dt Qin - K * d/dt (z-zref) - K * 1/Tr * (z-zref)

or

(5) A * dz2/dt + K * dz/dt + K * z/Tr = d/dt Qin + K * d/dt zref + K * zref/Tr



Writing these equations in text on the forum is kind of awkward and to me makes it more difficult to understand than the notes I am copying from. Does anybody know a better way? Posting PDFs from sage/some kind of latex editor?

edit: looks like I replied to the wrong thread. Let's continue in http://www.plctalk.net/qanda/showthread.php?t=86229 instead of this "me too" thread.

 

[Peter Nachtwey]

This is the kind of interest I was looking for. Real quick it looks like you really know what you are doing. Don't worry, I will make it interesting for you. I will check tonight on the real working thread.

You are right. It is awkward writing formulas on this site. I don't think this site supports latex or anything like it. I will check that out too.
I would look at the Daum equation editor.
https://chrome.google.com/webstore/...editor/dinfmiceliiomokeofbocegmacmagjhe?hl=en

I use Mathcad. I have the problems, equations and simulations all worked out already and are already posted. You just dont have the link yet.

Please continue on the other working site.

 

[Christoph]

When you have a minute I would like to have you asses an issue we have been working on in our tank farm here. We have been unable to resolve.
Very frustrating.

 

[Peter Nachtwey]

When you have a minute I would like to have you asses an issue we have been working on in our tank farm here. We have been unable to resolve.
Very frustrating.

Start a new thread.

 

[Christoph]

Ok about a year and a half ago we installed E&H Profibus tank weight sensors. E&H PMP75's
These tanks are vented to atmosphere and are agitated with a slow moving 3 level mixer.
Tanks are carbon steel about 8 feet in diamater and 25 feet tall with welded legs that go to the floor. 4 of the tanks have severe weight changes registered by the sensors. All of the tanks show some sign of change some just more severe like 6k lbs. thru a 24 hour cycle of heating and cooling as the change seems to follow temperature. I have had E&H in to check our application and the instruments and they can find no fault. They say that it must be the properties of the chemical in the tank. Mounted a Rosemount instrument
installed with the E&H, same exact issue.