Information Technology Reference
In-Depth Information
l
ν
l
ν
l
ν
l
ν
∂μ
i (
u ν , α
i )
= ∂μ
i (
u ν , α
i )
u ν
x q .
(A.36)
x q
u
ν
l
ν
l
ν
∂μ
i (
u ν , α
i )
can be obtained directly from the definition of the membership
u
ν
function. Thus,
x k
k = 0
n
c k ν ˜
n
n
u
c k ν
c k ν ˜
x k
ν
x q =
=
x q ˜
x k +
x q ,
(A.37)
x q
k
=
0
k
=
0
ergo,
n
u
c k ν
ν
x q =
x q ˜
x k +
c q ν .
(A.38)
k
=
0
Substituting ( A.22 )into( A.38 ):
∂ω
N ν
r
ν
s
c k ν
c k ν )
x q ω
ν (
n
u ν
r
,
s
=
1
x q =
c q ν +
x k .
˜
(A.39)
N ν
2
r = 1 ω
k
=
0
r
ν
Finally, substituting ( A.31 ) and ( A.34 )into( A.26 ), the expression that calculates
the derivative of the degree of activation of the rules of the plant in a closed-loop
fuzzy control system is obtained:
m
l
l
n
w i (
j = 1 μ
∂μ
qi (
x q , σ
qi )
k = 1 , k = q μ
x
,
u
)
l
l
l
l
=
ji (
u j , α
ji )
ki (
x k , σ
ki )
x q
x q
∂μ
n
m
m
l
v
l
v
i (
u
v , α
i )
,
l
l
l
l
+
1 μ
ki (
x k , σ
ki )
μ
ji (
u j , α
ji )
x q
k
=
v =
1
j
=
1
,
j
= v
(A.40)
l
ν
l
ν
where ∂μ
i (
u ν , α
i )
is calculated using ( A.39 ) over ( A.36 ) and the derivatives of
x q
the membership functions can be obtained directly from the definition of the function
itself.
 
Search WWH ::




Custom Search