PRINCIPLES OF FEEDBACK

There are different control mechanisms that can be used. Two broad control schemes
are

feedback control and

feed-forward control.

Feedback control is a control mechanism that uses information from measurements

to manipulate a variable to achieve the desired result.
Feed-forward control, also called anticipative control, is a control mechanism that
predicts the effects of measured disturbances and takes corrective action to achieve
the desired result.
Negative Feedback
By definition, negative feedback is when a change (increase/decrease) in some
variable results in an opposite change (decrease/increase) in a second variable. This
is demonstrated in Figure 3 where a loop represents a variation toward a plus that
triggers a correction toward the minus, and vice versa. Negative feedback leads to a
tight control situation whereby the corrective action taken by the controller forces the
controlled variable toward the set point, thus leading the system to oscillate around
equilibrium.

Figure 3. Negative Feedback: Maintenance of equilibrium and convergence

Positive Feedback

In a positive feedback control system, the set point and output values are added
together by the controller as the feedback is in-phase with the input. The effect of
positive (or regenerative) feedback is to increase the systems gain, ie, the overall
gain with positive feedback applied will be greater than the gain without feedback.
Its a process that occurs in a feedback loop in which the effects of a small
disturbance on a system include an increase in the magnitude of the perturbation.
That is, A produces more of B which in turn produces more of A.
positive feedback is in phase with the input, in the sense that it adds to make the
input larger. Positive feedback tends to cause system instability. When the loop gain
is positive and above 1, there will typically be exponential growth, increasing
oscillations or divergences from equilibrium.

PERFORMANCE CRITERIA
Performance criteria are various measurable parameters that indicate how good (or
bad) the control system is. These are divided into transient (moving) and steadystate (not changing) parameters.
The exact path the controlled variable takes when going from one position to the
next is called its transient response. Consider the behavior of the robot arm whose
response is shown in Figure 11.3; it is directed to move from 0 to 30, as shown by
the dashed line. This type of command (changing instantaneously from one position
to another) is called a step change. The actual response of the system is shown as
a solid line. As you can see, there is a difference between the ideal path of the arm
and the one it took. One major consideration is how fast the system picks up speed
(called rise time). The real arm simply cannot move fast enough to follow the ideal
path.
Rise time (T) is usually defined as the time it takes for the controlled variable to go
from 10 to 90% of the way to its new position. Another transient parameter is
overshoot.
Once the arm starts moving, its momentum will keep it going right on past where it
was supposed to stop. Overshoot can be reduced by the controller but usually at the
expense of a longer rise time. Settling time (Ts) refers to the time it takes for the
response to settle down to within some small percentage (typically 2-5%) of its final
value. In this case, it is the time it takes for the oscillations to die out. Rise time,
settling time, and overshoot are all related; a change in one will cause a change in
the others.
The steady-state error (ESS) of the system is simply the final position error, which
is the difference between where the controlled variable is and where it should be. In
Figure 11.3, ESS is shown as the position error after the oscillations have died out.
This error is the result of friction, loading, and feedback-sensor accuracy. A
sophisticated controller can reduce steady-state error to practically zero.

Figure 11.3; Transient response.

It is a representation of the control system giving the inter-relation between the

transfer function of various components. The block diagram is obtained after
obtaining the differetial equation and transfer function of all components of a control
system. The arrow head pointing towards the block indicates the output.
If G(s) is the transfer function; then for

G( S )

C( S )
R( S )

After obtaining the block diagram for each and every component,, all blocks are
combined to obtain a complete representation. It is then reduced to a simple form
with the help of block diagram algebra.