PID control of level in 2 tanks -Process Question

MorphuisOGrady

Member
M
Join Date
Jul 2007
Location
Kiruna
Posts
600
Good morning,

I have a question and I don't really know what you would call this....

I have 2 tanks which I need to control the level or froth in. Each with its own level transmitter and control valve to empty it. No control of feed. By opening the control valve in tank1 releases contents into tank 2.

Using an AB Logix5000 plc and PIDE instructions this is easily achievable until...

I need to implement some sort of strategy that if there was a disturbance on feed, i.e a large inrush flow to tank 1 due to plant instability that the control valve in tank2 reacts quicker to get the level down despite the level not been high YET.

Does this make sense? I want the PID control of tank 2 to disregard it's own PV and get the level down in anticipation for a high flow thats coming based on whats happening in tank 1.

I apologize in advance if the process description is vague.
 
So each tank has level control with control valves on each tank outlet. Tank1 output feeds into Tank2. To detect the sudden change in flow into Tank1 you need to measure the flow of liquid into Tank1.
You could introduce a signals adder into the valve output signal of Tank2 and introduce the flow signal into the adder. That way when a disturbance is measured at the input to Tank1 an immediate change would be introduced into Tank2 level control valve position. After some time the system would balance out where flow into Tank1 is matched with flow out of Tank2. So the feed flow biases Tank2 level control valve. The amount of change required from the flow signal could set between 0% and 100%. ( K * Flow + Level Controller output )
 
Thanks for your input. Unfortunately having a flow transmitter is not an option. I'm told it could be done using whats called "Feed Forward"????

Have you every used this?
 
I was going to get to problems like this on my Advanced Control: Tank Level Control.
http://plctalk.net/qanda/showthread.php?t=86229&highlight=advanced

I understand what you need to do but you haven't supplied all the data for the best answer yet.

Calistodwt has it about right about adding to the control signal to tank 2's valve but I will be more precise. If the PID for valve 1 provides a signal of 80% then 80% should be added to the bias of the PID and control out to the valve for tank 2. The idea is that if the two valves are open the same amount then the in-flow to tank 2 should approximately be equal to the out-flow of tank 2 at all times.

However I am making an assumption because you haven't provided the answer. I am assuming that if both valves are open the same amount the in-flow will be equal to the out-flow. However, this will not be true if there is a different pressure drop across the valves. The pressure drop should be proportional to the tank level. I am also assuming the valves for tank 1 and 2 are the same.

The quick answer is to add the control output to valve 1 to the PID bias for tank 2.
It should get you close.
 
Peter Nachtwey said:
I was going to get to problems like this on my Advanced Control: Tank Level Control.
http://plctalk.net/qanda/showthread.php?t=86229&highlight=advanced

I understand what you need to do but you haven't supplied all the data for the best answer yet.

Calistodwt has it about right about adding to the control signal to tank 2's valve but I will be more precise. If the PID for valve 1 provides a signal of 80% then 80% should be added to the bias of the PID and control out to the valve for tank 2. The idea is that if the two valves are open the same amount then the in-flow to tank 2 should approximately be equal to the out-flow of tank 2 at all times.

However I am making an assumption because you haven't provided the answer. I am assuming that if both valves are open the same amount the in-flow will be equal to the out-flow. However, this will not be true if there is a different pressure drop across the valves. The pressure drop should be proportional to the tank level. I am also assuming the valves for tank 1 and 2 are the same.

The quick answer is to add the control output to valve 1 to the PID bias for tank 2.
It should get you close.
Click to expand...

Hi Peter,

I've got one of the best to help me! Thanks

I haven't got all the information yet so I ll come back when I do. I can't say if there is a different pressure drop across the valves but can say the tanks are at different levels. Tank 1 is higher than tank 2.

The valves are identical.

I don't entirely follow you when you say add the control output to valve bias of valve 2. If the output (CV) to valve 1is 80% I add this to the bias. Is the bias a parameter of the second PID?

Thanks
 
MorphuisOGrady said:
Tank 1 is higher than tank 2.

The valves are identical.
Click to expand...
That is good to know. The second valve will need to be opened a just a little more than the first to get the same flow. Does that make sense? The ratio is sort(level 1/level 2). This is because the flow is proportional to the square root of the pressure drop which is proportional to the levels/

I don't entirely follow you when you say add the control output to valve bias of valve 2. If the output (CV) to valve 1is 80% I add this to the bias. Is the bias a parameter of the second PID?
Thanks
Click to expand...
The PID block should have a bias. If how you can simply add the bias to the control output of the bias.

The idea is that if the second PID is controlling the second level perfectly and then the first tank starts adding more flow then the add how much the first PID must open to the second PID. However this must be scaled by the sort function I showed above.
 
The use of the flow meter to bias the control valve on Tank 2 is a type of feed forward control. That is why I suggested it. Plus your requirement was for a reaction to a disturbance on inflow to Tank 1 would give you an immediate response to the valve on Tank 2. If you say that you cannot have a flow meter then you cannot have an immediate response with your current set-up.
Peter's response would obviously give you a response over time though it can only be a reaction to changes in Tank 1 level which would occur some time after any disturbance.
This is not what you required in your original mail but under the circumstances (no flow meter ) you need to revise your requirements.
Biasing of your T2 control valve from T1 control valve needs to be considered carefully though as if for example both valves required to be 50% open then a factor would have to be applied to the influence of T1 valve otherwise T2 valve could be 100% open (50% bias plus PID output). Maybe an alternative would be a HI/LOW selector option on T2 level valve.
 
Last edited:
The control has changed now. Rather than control the level in the tank they want to control by the depth of the froth in the top of the tank.Its a floatation tank.

Can we still use the feed forward/bias concept for this application?
 
You could monitor rate of change in tank 1 and adjust the gain value of PID loop in tank 2, this is essentially what feed forward gain is.

So PID loop on Tank2 has set PID parameters and depending on rate of increase of level in tank 1 adjust the P term of PID loop in tank 2.

So normal situation tank 2 sits happy controlling at set P term, but when level in Tank 1 is increasing above a certain rate then increase the p term of Tank2 pid loop.
 
BigEndian said:
You could monitor rate of change in tank 1 and adjust the gain value of PID loop in tank 2, this is essentially what feed forward gain is.
Click to expand...

This is not a good idea as you are now introducing a variable loop gain into T2 level control. This will introduce disturbances into the loop and cause loop instability.

My previous suggestion of a maybe using a low/high selector arrangement is also not a good idea either after giving it some thought.

The best option under the set-up which you have, would be to use a bias from T1 level valve as suggested by Peter as you don't have a flow measurement available.
 
My understand of this issue,

Each Tank had Level Transmitter.

The Float Switch on the Tank no: 1 to measure froth.

The CV-1 opening depending the Level SP and PV.

There 2 questions now,

To Control Level and to Control froth in.

Not ‘or’

There had 2 type of raw in the tank.

Now the froth controlling the Level not Level Transmitter.

There nothing controls the froth e.g.

Rotate scoop for froth or spraying water system or froth breaker,

On Tank no: 1

Or overflow drain for froth to Tank no: 2

Or buffer Tank to support during inrush.

The level and froth change with incoming flow.


The present design is not include the inrush process.

When some process not includes during design then

Additional or modification mechanical part need to implement.
 

Similar Topics

S
LEVEL CONTROL BY PID
How can I connect PID Output to a valve. In ladder logic program is there any logic do I want to add between valve and PID? PV=SP What will be the...
Replies
7
Views
412
Bit_Bucket_07
Bit_Bucket_07
S
PID level control of sequential pumps
Hi, I started off my career in PLC programming doing water/wastewater on AB around 20 years ago, but then moved overseas a few years later and...
2 3 4
Replies
57
Views
12,048
drbitboy
drbitboy
R
PID level control with ratio of pumps.
Have a level control application that consists of a small chamber relative to the size of the incoming and outgoing pumps. I've done this type of...
Replies
6
Views
3,592
robertmee
R
A
PID to slow, Level Control
Hi friends I have a L24ER-QBFC1B in a Level Control that you can see in the picture. Tank1 is the process tank. Level sensor is the process...
Replies
4
Views
3,051
asteroide
A
C
Help need on Level PID Control
I am working on a CIP project. There is a tank which is about 1500L. We need to keep the water volumn inside the tank to be SP:200L. The control...
2
Replies
17
Views
10,843
daba
D
Back
Top Bottom