Study Questions

Apply Your Knowledge

You are asked to evaluate the following control design for the packed bed reactor. The models have errors indicated. Would you recommend a feedforward controller?

Yes         No

The valve to AC-2 model

gains, dead times and time constants have errors of approximately +/- 25%

AC-2(SP) to AC-1 model

gains, dead times and time constants have errors of approximately +/- 25%











You have implemented a feedforward controller without feedback and obtained the behavior shown below. What would you change (fine tune)?   

       (  q = dead time)


We should be able to diagnose performance and continually improve.


Tune qff, the feedforward controller dead time

Tune Tlg, the feedforward controller lag time


Tune Tld, the feedforward controller lead time

Tune Kff, the feedforward controller gain










If feedback control were not implemented (perhaps because a real-time sensor is not available) would you recommend a
valve positioner for this feedforward-only design?

Yes         No












Suppose that the packed-bed reactor situation were different: A2 does not vary significantly and the feed temperature, T1, is a significant disturbance. Would you recommend feedforward control?

Yes         No











If the blended product quality were measured by A100 how could this be used to
improve the control performance for the initial design show below?

What could cause poor performance of the design below?


include in feed forward.

remove feedforward and use feed back instead.


have feedback from A100 manipulate the ratio. 













Table 15.2 in the textbook includes the statement "
Feedforward does not affect the stability of the system."  What assumption is required for this statement to be true?

We want to understand limits to useful guidelines.



The feedback dynamics are open-loop stable.

The plant is linear.


The models used in the feedforward controller match the plant perfectly.

The feedforward controller is stable.












How are the results of feedforward and feedback controllers combined in linear designs?

We gain great advantages from combining feedback and feedforward!



The two signals are added.

The two signals are multiplied.


The two signals are subtracted.

The two signals are not combined.













In what order are feedforward and feedback controllers tuned?

We should think about implementation strategies.


Tune the feedforward first.

Tune the feedback first.


Tune either one first.

Tune both simultaneously.