Mathematical Problems in Engineering
Compared with our existing result in our previous result
[], the established model is more general since the lower
and upper boundaries cannot be known.
Remark 3. Usually, besides quantization, signals transmitted
over the channels also assume discretization on time, such as
in [, ]. To focus on the control design for the mismatch
problem of quantizer sensitivity, only signal quantization is
considered in this paper. Similar discusses were also made in
[, , ].
emainaimofthispaperistoformdecentralized
adaptive integral sliding mode control strategy with L
2
performance for the nonlinear large-scale system () subject
to the quantization mismatch as described in ().
Now considering
𝑖
(
𝑖
())in () and the ratio in (),we
have
𝑖
𝑖
(
)
=
𝑖𝑑
(
)
𝑖
𝑖
(
)
𝑖𝑐
(
)
=
𝑖𝑑
(
)
𝑖𝑐
(
)
𝑖𝑐
(
)
𝑖
𝑖
(
)
𝑖𝑐
(
)
=
𝑖
(
)
𝑖𝑐
(
)
𝑖
𝑖
(
)
𝑖𝑐
(
)
=
𝑖
(
)
𝑖
(
)
+
𝜏
𝑖𝑐
(
)
,
()
where
𝜏
𝑖𝑐
() :=
𝜏
𝑖𝑐
(𝑡)
(
𝑖
()) −
𝑖
() and
𝜏
𝑖𝑐
(𝑡)
(
𝑖
()) :=
𝑖𝑐
()
𝑖
(
𝑖
()/
𝑖𝑐
()).And
𝜏
𝑖𝑐
()satises
𝜏
𝑖𝑐
(
)
≤∇
𝑖
𝑖𝑐
, ()
with ∇
𝑖
=
𝑖
/2;
𝑖
is the dimension of the vector
𝑖
.
Denition 4. L
2
Gain Performance.Givenpositivescalars
𝑖
,
the nonlinear large-scale system in ()-() with
𝑖
() = 0is
said to be robust stable with a bounded L
2
gain performance;
if it is robust stable for
𝑖
() = 0, and, under zero initial
condition, for nonzero
𝑖
()∈L
2
[0,∞),itholdsthat
𝑁
𝑖=1
∞
0
𝑇
𝑖
(
)
𝑖
(
)
<
𝑁
𝑖=1
2
𝑖
∞
0
𝑇
𝑖
(
)
𝑖
(
)
.
()
Remark 5. Denote that
(
)
=
𝑇
1
(
)
,
𝑇
2
(
)
,...,
𝑇
𝑁
(
)
𝑇
,
(
)
=
𝑇
1
(
)
,
𝑇
2
(
)
,...,
𝑇
𝑁
(
)
𝑇
,
()
and letting
𝑖
=,=1,2,...,,onecaneasilyobtainthat
∞
0
𝑇
(
)
(
)
<
2
∞
0
𝑇
(
)
(
)
.
()
isistheusuallyconsideredL
2
gain performance for
simple linear system as in [].
For the considered problem, the following lemma will be
used.
Lemma 6 (see []). Given a symmetric matrix Θ and
matrices , with appropriate dimensions, then
Θ+Ξ
(
)
+
𝑇
Ξ
𝑇
(
)
𝑇
<0
()
for all Ξ()satisfying Ξ
𝑇
()Ξ()≤,ifandonlyifthereexistsa
scalar >0 such that the following inequality holds:
Θ+
𝑇
+
−1
𝑇
<0.
()
3. Main Results
In this paper, the following integral-type sliding surface
function is considered:
𝑖
(
)
=
𝑖
𝑖
(
)
−
𝑡
0
𝑖
𝑖
−
𝑖
𝑖
𝑖
𝑖
𝑖
,
()
where
𝑖
and
𝑖
aresystemmatricesdenedinlarge-scale
system ().
𝑖
∈ R
𝑚
𝑖
×𝑛
𝑖
and
𝑖
∈ R
𝑚
𝑖
×𝑛
𝑖
are real matrices
to be designed. In particular, the matrix
𝑖
is selected such
that
𝑖
𝑖
is nonsingular. Without loss of generality, as done
in [, ],
𝑖
is designed to be
𝑖
=
+
𝑖
, the pseudoinverse of
𝑖
for the convenience of the proof.
By the theory of sliding mode control [, ], when
the trajectory of large-scale system () is kept on the sliding
dynamics, it has
𝑖
() = 0and
𝑖
() = 0. us, by taking
𝑖
() = 0, one can obtain the equivalent control for the th
subsystem of large-scale system ():
𝑖eq
𝑖
(
)
=−
𝑖
𝑖
−1
𝑖
𝑖
(
)
𝑖
(
)
+
𝑖
𝑖
𝑖
,,+
𝑖
𝑖
𝑖
𝑖
(
)
+
𝑖
𝑖
𝑖
(
)
=−
𝑖
𝑖
−1
𝑖
𝑖
(
)
𝑖
(
)
−
𝑖
,,−
𝑖
𝑖
(
)
−
𝑖
𝑖
−1
𝑖
𝑖
𝑖
(
)
.
()
Substituting () into (),onecangetthat
𝑖
(
)
=
𝑖
+
𝑖
(
)
𝑖
(
)
+
𝑖
−
𝑖
𝑖
−1
𝑖
𝑖
(
)
𝑖
(
)
−
𝑖
,,
−
𝑖
𝑖
(
)
−
𝑖
𝑖
−1
𝑖
𝑖
𝑖
(
)
+
𝑖
𝑖
,,
+
𝑖
𝑖
(
)
=
𝑖
−
𝑖
𝑖
+−
𝑖
𝑖
𝑖
−1
𝑖
𝑖
(
)
𝑖
(
)
+
−
𝑖
𝑖
𝑖
−1
𝑖
𝑖
𝑖
(
)
.
()
Denoting
𝑖
=−
𝑖
(
𝑖
𝑖
)
−1
𝑖
,
𝑖
=
𝑖
−
𝑖
𝑖
,
𝑖
() =
𝑖
𝑖
(),and
𝑖
=
𝑖
𝑖
,onecanseethat() can be
represented as
𝑖
(
)
=
𝑖
+
𝑖
(
)
𝑖
(
)
+
𝑖
𝑖
(
)
, ()
where
𝑖
()=
𝑖
𝑖
()=
𝑖
𝑖
𝑖
()
𝑖
.
We have the following theorem for the sliding mode
dynamics in ().