Page 39 - Practical Control Engineering a Guide for Engineers, Managers, and Practitioners

14 C h a p te r 0 n e

the absence of disturbances, the controller drives the process variable
to new set point quickly.
Before leaving this example, we should make a few comments on
the nomenclature associated \Vith the disturbances discussed. In
Chap. 8, we will discuss how to quantitatively characterize these dis-
turbances but for the time being consider the noise riding on the tem-
perature signals in Fig. 1-15 as an example of a stochastic variation.
We perhaps can conclude the following from this example:
1. When process is subject to stochastic disturbances, feedback
controllers can 11ot "drmv straight lines."

2. Although there may be some attenuation, Disturbances In --7
Disturbances Out. As we shall see later on, the process itself
may tend to attenuate input disturbances. Controllers can aid
in the attenuation.
3. Controllers can move the process to a neighborhood of a ne\v
set point. Controllers may not be able to "draw straight lines"
but they may be able to move the mxragc value of the process
output satisfactorily near a desired set point.

Figure 1-18 gives a pictorial summary of the above comments
When confronted with set-point changes in the face of relatively
small stochastic disturbances a feedback controller can be extremely
useful. If one is so lucky to have good measurements on incoming
streams that represent disturbances to the process, feedforward
control coupled with feedback control probably is a good choice

Use fl'L'dback Use fL'L'dfon' ard
control

FrcuRE 1-18 Different approaches for different problems.