It picks the lowest of the two signals from PIC and LIC and passes it to FIC to operate on. To make it work as intended, the action on PIC must be reversed. If level is low, more feed is needed, and CV opens. If pressure is low, CV needs to close to raise pressure. Opposite action is needed on the two signals.
The sketch with the valve failure position added would also clarify the control details. Also, my initial view would desire more controller scheme definition.
I don't like the scheme as shown if both level and pressure controllers were simply PID (proportional, integral, derivative). With a simple PID scheme, the outputs from both pressure and level controllers would be fighting for the flow setpoint rate. A control algorithm with both high and low setpoint brackets (no output change when within limits) on each pressure and level would seem like an improvement. I have not played with the recent DCS algorithms but setpoint bracket configuration would exist.
The simple PID schemes could also likely work. The setpoint brackets could make the flow control flatter. The horizontal vessel could narrow the level setpoint scheme such as between 40% and 60% of the level measurement. Also determine whether the feed is liquid, two-phase or vapor that is condensed. Such issues could impact the pressure scheme in an interesting manner.
Consider overrides via a safety control system. Normally the operator can select the controller on automatic mode that ignores the level and pressure controllers + the low output selector. Trip the pump on low level. Trip the inlet valve on high level. ...
The turn-down ratio for this drum may be required to considered.
The bigger turn-down ratio of the drum, the large control range control valve may be required, and the smaller turn-down ratio to the drum, the smaller control range control valve may be required.
