I'm trying to wrap my head around what feels like should be a straightforward proposition.

We have established reasonable control over a water pumping system using PLC-based PID reservoir level control driving VFD speed.

There is a possibility that if the Pumps were run at full power, that the discharge pressure would exceed safe operating level.
I am trying to figure out how to incorporate pressure-limitation control.
As I read the description of how Cascade control works, I am not sure whether that method is appropriate.
The other tactic which occurs to me is to run a pressure PID loop with the setpoint set to PMax.
when PV exceeds SP, CV would go negative and would be scaled and added to inhibit the primary control.
When PV is less than SP, the positive CV would be set to zero such that the primary control loop
would only respond to over-pressure and would do so quickly so that the over-pressure pulse would be suppressed / spread out.

I also realize that ad-hoc arm-wavery solutions likely induce unanticipated instability issues.

Is there a term for this kind of pressure-limitation control ?
Is it a form of Cascade Control or is there some other term for this ?

A related question is how to deal with worst-case scenarios in control strategies.

Consider a general situation where "normal" reservoir inflows are within control criteria.
How does one best accommodate the "100 year flood event" ?
I suppose the best-case scenario is to run the pump at maximum speed - with over-pressure and under-level control interlocks ?

Best regards

 

Typically you would implement two separate controllers . LC with the level SP and a PC with the high pressure as its SP. Both outputs go to a selector block which selects the lowest value and outputs the lowest value to the motor VSD.
The PC will sit at 100% open until the pressure starts to raise. The PC PID will move toward 0% and override the LC PID if its output is higher.

Hope this makes sense.