Stabilization Algorithm for Interval Type-2 TSK Fuzzy Logic
Control Systems with Bounded Time-Varying Delay
SUN Xun, ZHANG Huaguang, WANG Yingchun
College of Information Science and Engineering, Northeastern University, Shenyang, Liaoning, 110819, P. R. China
E-mail: sunxun2014@gmail.com; hgzhang@ieee.org; wangyingchun@ise.neu.edu.cn
Abstract: This paper studies the problem of stabilization for an interval type-2 Takagi-Sugeno-Kang fuzzy logic control system
(IT2 TSK FLCS) with time varying delay. A type-1 fuzzy system (T1 FS) is used for modeling the plant, then an interval type-2
fuzzy system (IT2 FS) is used for constructing the controller which has been realized better abilities to handle uncertainties than
its T1 counterpart. A less-redundant Lyapunov-Krasovskii function and a useful linear matrix inequality (LMI) are constructed,
which make the closed-loop system stabilization analysed in a mathematical way. It provides a novel way to deal with the
stabilization problem of IT2 TSK FLCS with time varying delay.
Key Words: TSK, LMI, Time varying delay, interval type-2 fuzzy system, TSK model
1 Introduction
Type-1 fuzzy sets (T1 FSs) were first introduced in 1965
[1] and type-2 fuzzy sets (T2 FSs) were proposed in 1975
[2], both by L.A.Zadeh. A T2 FS is considered as an exten-
sion of a T1 FS. The difference between a T1 FS and a T2 FS
is the membership function (MF). The MF of a T1 FS is two-
dimensional, while the MF of a T2 FS is three-dimensional
with the primary MF and the secondary MF, so that when
all sources of uncertainties disappear a T2 MF reduces to a
T1 MF. The primary membership value is a crisp number in
[0,1] while the secondary membership value is a fuzzy set in
[0,1]. The union of the primary MF is called footprint of un-
certainty (FOU) and the domain of the secondary MF is the
primary membership. A T2 FS provides a better representa-
tion of uncertainty in the most of applications. So far, there
have been many successful applications employing T2 fuzzy
systems in the real world, such as computing with words [3],
modeling and control [4], patten recognition [5], data clas-
sification [6], and decision making [7]. Today an IT2 FS
whose secondary MFs equal one is the most widely used T2
FS because it is computationally simple to use.
Today the three most popular fuzzy logic models used by
engineers are Mamdani [8], TSK [9]-[10] and fuzzy hyper-
bolic model [11]-[12]. There are characterized by IF-THEN
rules, and have the same antecedent structures. They dif-
fer in the consequent structures. The consequent of a Mam-
dani rule is a fuzzy set, whereas the consequent of a TSK
or a fuzzy hyperbolic model rule is a function. Liang and
Mendel [13] discussed three possible structure of T2 TSK
FS: systems with T2 antecedents and crisp (T0) consequents,
T1 antecedents and T1 consequents, and T2 antecedents and
T1 consequents. Later Mendel and his colleagues refer to
them as A2-C0, A1-C1, and A2-C1, respectively. This pa-
per here only focuses on A2-C0 IT2 TSK FS. Much effort
has been directed toward finding a suitable controller using
TSK FLCS in order to guarantee robust stability, for exam-
ple, [14]-[16].
This work was supported by the National Natural Science Founda-
tion of China (61034005, 61433004), and the National High Technology
Research and Development Program of China (2012AA040104) and IAPI
Fundamental Research Funds 2013ZCX14. This work was supported also
by the development project of key laboratory of Liaoning province.
The time-delay nonlinear systems are very common
in many kinds of industrial fields, which arises from
biology systems, economics intrinsically, and chemical
processes[17]-[20]. It is a straightforward idea to extend the
TSK model to the nonlinear systems with time-delay. So,
there are some models using TSK FLCS with time-delay
representing nonlinear system, such as [21]-[24]. Recently
some articles discuss the stability of IT2 FLCS, for example,
[25] addresses the design problem of static output feedback
controllers for nonlinear interval time-delay systems by us-
ing the Takagi-Sugeno fuzzy model. In addition, [26]-[27]
investigated stability analysis of IT2 TS FLCS. To the best
of the authors’ knowledge, few use IT2 TSK FLCs with time
varying delay in the literature.
Motivated by the above discussion, this paper investigates
the stability and the stabilization of IT2 TSK fuzzy time-
varying delay systems. We study the problem of stabilization
for an IT2 TSK FLCS with time-varying delay. We choose
a T1 FS for modeling the plant and an IT2 FS for construct-
ing the controller. It has been realized that T2 FS has bet-
ter abilities to handle uncertainties than T1 FS. A new less-
redundant Lyapunov-Krasovskii function and a useful LMI
are constructed, which make the closed-loop system stabi-
lization analysed in a mathematical way. The article pro-
vides a novel way to deal with the stabilization problem of
IT2 TSK FLCS with time varying delay.
This paper is organized as follows. First, we will intro-
duce the problem in section 2. Next, controller design which
based on a a novel Lyapunov-Krasovskii function is given in
section 3. Finally, the problem of the fuzzy controller design
is converted into LMI feasibility problem, by solved the LMI
we can get controller parameters.
2 System Description
In this section, we first give a nonlinear time varying de-
lay system represented by T1 TSK fuzzy system with time
delays.
Plant rule i:
IF f
1
(t) is F
i1
,f
2
(t) is F
i2
and ··· and f
q
(t) is F
iq
Then
˙x(t)=A
i
x(t)+A
id
x(t − d(t)) + B
i
u(t), (1)
x(t)=φ(t),t∈ [−h, 0],
Proceedings of the 34th Chinese Control Conference
Jul
28-30, 2015, Han
zhou, China
321