pump speed pid control

[GNP 0]

I was hoping someone here might have some input on what I’m trying to figure out as I am a bit stumped at the moment.

 

I am trying to control a pump (motor) speed based on the suction pressure at the inlet of a pump. I have a 4-20ma compound pressure transducer on the suction side of the pump connected to a CJ2M analog input and a 4-20ma output from the PLC to the pump VFD.

I am using the PIDAT instruction with the transducer as the input  and the 4-20 to the VFD as the output for pump speed. I have set the SV as the maximum suction pressure to maintain.

I have to finish testing and tuning the PID but I think I have this part basically figured out.

 

My issue is that in addition to the above I have another 4-20ma input from a separate control system that “tells” the pump to run at a certain speed, which will vary regularly.

At times this control system will ask the pump to run at too high a speed causing too great a suction pressure (NPSHR > NPSHA) which in turn causes cavitation which will damage the pump and destroy the product we are pumping. The temperature (viscosity) of the products and liquid elevation vary significantly during the pumping process so we are not able to just pick a maximum speed to run.

 

So what I am struggling with is how to integrate the two control systems such that the pump will always attempt to run at the requested speed but Only if it does not exceed the SV suction pressure. If the suction pressure becomes to great the pump should slow down via the PID which in turn reduces the suction pressure.

Basically run as close as possible to the requested speed but do not exceed the maximum suction pressure.

 

I had thought at one point to keep the two speed control methods separate so the the other control system would be the primary control and compare the value from the transducer to the SV ( maximum suction pressure allowed) and turn on the PID control on when the suction pressure dictated but i think this will only cause oscillating as the PID takes over corrects the the speed which will then in turn revert back to the primary control which will then potentially increase the pump speed (and suction pressure) an so on and so on.

 

Any help on this one would be greatly appreciated.

[nednor 0]

Have you considered using a cascading control loop? Meaning one PID output controls the setpoint of the second one. That way you can limit the system to act within the suction pressure limits.

http://blog.west-cs.com/blog/how-does-cascade-control-work