16
Fuzzy Control
16.1
INTRODUCTION TO INTELLIGENT CONTROL
In the last two decades, a new approach to control has gained considerable attention.
This new approach is called intelligent control (to dist
15
Robust Control
15.1
INTRODUCTION
Control engineers are always aware that any design of a controller based on a xed
plant model (e.g., transfer function or state space) is very often unrealistic. Th
Adaptive Control
625
part B q1 of Bm q1 while Eq. (14.4-16) should be valid, otherwise there is no
m
solution to the design problem.
It is necessary to establish conditions under which a solution for
Adaptive Control
613
In what follows, we seek to nd a control law uk, which drives the ltered
"
plant-model error ek d q1 ek d to zero. Using Eqs (14.3-21) and
(14.3-22), we have
q1 ek d q1 yk d q1 ym
Robust Control
Figure 15.7
649
Checking robust stability with a multiplicative uncertainty, for Example 15.3.1.
15.3.2 Robust Stability with an Inverse Multiplicative Uncertainty
In this subsection a
Robust Control
661
criterion, e.g., the Routh criterion. Applying the Routh criterion in the four
Kharitonovs polynomials, we obtain the following four Routh tables:
1.
2.
3.
4.
Polynomial p1 s:
1
s5
The Z-Transform
Figure B.3
709
The unit gate sequence g k k k1 k k2 .
Figure B.4 shows the graphical representation of rk k0 .
5 The Exponential Sequence
The exponential sequence is designated by gk a
Fuzzy Control
697
It is evident here that only rules 1, 2, 4, and 5 have a nonzero contribution, the
remainder playing no part in the nal decision. Furthermore, rule 1 is seen to be
dominant, while ru
Fuzzy Control
Figure 16.13
685
Graphical forms of the rules 69 of Table 16.1.
Next, determine the value of sk , dened as follows:
1
n
o
sk min k1 ; k2 minf0:5; 0:2g 0:2
1
e
e
16:8-1
This completes the
System Identification
21.
601
E Walter. Identication of State Space Models. Berlin: Springer Verlag, 1982.
Articles
22.
23.
KJ Astrom, PE Eykhoff. System identicationa survey. Automatica 7:123162, 197