
Interval type-2 fuzzy logic systems (IT2 FLSs) already become an emerging technology in recent years. As the most popular type-reduction (TR) algorithms, Karnik-Mendel (KM) algorithms own the advantage of maintaining the uncertainties flow in systems. This paper analyzes the initialization for KM types of algorithms. Furthermore, the weighting approaches of them are also given by means of the Newton-Cotes quadrature formulas. Importantly, the reasonable initialization weighted enhanced Karnik-Mendel (RIWEKM) algorithms are provided to complete the centroid type-reduction of IT2 FLSs. Three computer simulation experiments illustrate that, the proposed RIWEKM algorithms own both smaller absolute errors and faster convergence speeds in contrast to the EKM and RIEKM algorithms.
Citation: Yang Chen, Jiaxiu Yang, Chenxi Li. Design of reasonable initialization weighted enhanced Karnik-Mendel algorithms for centroid type-reduction of interval type-2 fuzzy logic systems[J]. AIMS Mathematics, 2022, 7(6): 9846-9870. doi: 10.3934/math.2022549
[1] | Yanting Xiao, Yifan Shi . Robust estimation for varying-coefficient partially nonlinear model with nonignorable missing response. AIMS Mathematics, 2023, 8(12): 29849-29871. doi: 10.3934/math.20231526 |
[2] | Heng Liu, Xia Cui . Adaptive estimation for spatially varying coefficient models. AIMS Mathematics, 2023, 8(6): 13923-13942. doi: 10.3934/math.2023713 |
[3] | Yanping Liu, Juliang Yin . B-spline estimation in varying coefficient models with correlated errors. AIMS Mathematics, 2022, 7(3): 3509-3523. doi: 10.3934/math.2022195 |
[4] | Anum Iftikhar, Hongbo Shi, Saddam Hussain, Ather Qayyum, M. El-Morshedy, Sanaa Al-Marzouki . Estimation of finite population mean in presence of maximum and minimum values under systematic sampling scheme. AIMS Mathematics, 2022, 7(6): 9825-9834. doi: 10.3934/math.2022547 |
[5] | Sanaa Al-Marzouki, Christophe Chesneau, Sohail Akhtar, Jamal Abdul Nasir, Sohaib Ahmad, Sardar Hussain, Farrukh Jamal, Mohammed Elgarhy, M. El-Morshedy . Estimation of finite population mean under PPS in presence of maximum and minimum values. AIMS Mathematics, 2021, 6(5): 5397-5409. doi: 10.3934/math.2021318 |
[6] | Jieqiong Lu, Peixin Zhao, Xiaoshuang Zhou . Orthogonality based modal empirical likelihood inferences for partially nonlinear models. AIMS Mathematics, 2024, 9(7): 18117-18133. doi: 10.3934/math.2024884 |
[7] | Zawar Hussain, Atif Akbar, Mohammed M. A. Almazah, A. Y. Al-Rezami, Fuad S. Al-Duais . Diagnostic power of some graphical methods in geometric regression model addressing cervical cancer data. AIMS Mathematics, 2024, 9(2): 4057-4075. doi: 10.3934/math.2024198 |
[8] | Peng Lai, Wenxin Tian, Yanqiu Zhou . Semi-supervised estimation for the varying coefficient regression model. AIMS Mathematics, 2024, 9(1): 55-72. doi: 10.3934/math.2024004 |
[9] | Dayang Dai, Dabuxilatu Wang . A generalized Liu-type estimator for logistic partial linear regression model with multicollinearity. AIMS Mathematics, 2023, 8(5): 11851-11874. doi: 10.3934/math.2023600 |
[10] | Abdullah Mohammed Alomair, Weineng Zhu, Usman Shahzad, Fawaz Khaled Alarfaj . Non-parametric calibration estimation of distribution function under stratified random sampling. AIMS Mathematics, 2025, 10(2): 4457-4472. doi: 10.3934/math.2025205 |
Interval type-2 fuzzy logic systems (IT2 FLSs) already become an emerging technology in recent years. As the most popular type-reduction (TR) algorithms, Karnik-Mendel (KM) algorithms own the advantage of maintaining the uncertainties flow in systems. This paper analyzes the initialization for KM types of algorithms. Furthermore, the weighting approaches of them are also given by means of the Newton-Cotes quadrature formulas. Importantly, the reasonable initialization weighted enhanced Karnik-Mendel (RIWEKM) algorithms are provided to complete the centroid type-reduction of IT2 FLSs. Three computer simulation experiments illustrate that, the proposed RIWEKM algorithms own both smaller absolute errors and faster convergence speeds in contrast to the EKM and RIEKM algorithms.
Recently, neural networks such as the Hopfield neural network, cellular neural network, Cohen-Grossberg neural network and bidirectional associative neural network have attracted much attention due to their significant promise for many practical applications[1,2,3,4,5,6]. For example, in the field of signal processing, based on the theoretical results of discrete-time high-order switched neural networks synchronization, the encryption and decryption scheme of multi-channel audio signal design has good security [3]. In the field of fault diagnosis, by using deep parameter-free reconstruction-classification networks with parameter-free adaptively rectified linear units, the fault characteristics of vibration signals under the same fault state under different operating conditions can be better captured [5]. In the field of image processing, the theoretical results of random synchronization control framework based on semi-Markov switching quaternion-valued neural networks can be effectively applied to image encryption [6]. In particular, stability properties of the neural networks play a significant role in their design for solving practical problems. There are some different factors to influence the stability of neural networks. In particular, time delay is often unavoidable. Moreover, for the neural networks with time delay, there exists many techniques to reduce the conservatism of stability conditions, such as the improved bounding technique [7], free-weighting matrix theory [8], integral inequality technique [9] and so on.
It is well known that many dynamical systems may switch in different unpredictable modes, such as random failures [10]. As an effective tool, the Markovian jump process can be used to model these switching systems. Until now, there existed many literatures to investigate the stability, stabilization and observation for Markovian jump systems. For example, the authors investigated the adaptive sliding mode control problem of nonlinear Markovian jump systems with partly unknown transition probabilities, and obtained some conditions to guarantee the stochastic stability of the closed-loop system in [11]. The authors in [12] discussed the realization of H∞ finite-time control for a class of uncertain stochastic time-delay systems with unmeasured states through sliding mode control, and obtained some conditions to guarantee the system state was stabilized within a limited time interval. In [13], the authors established some passivity analysis criteria for Markovian jump singularly perturbed systems with partially unknown probabilities by using the hidden Markov model, and presented a unified controller design method to ensure the passivity of the system. Much more literatures can be found in [14,15,16].
In recent years, sliding mode control has become an effective robust control method because it is insensitive to model uncertainties, parameter variations and external disturbances. So, the sliding mode control has been used for lots of physical systems, such as robot manipulators, automotive engines and power systems. There are some existing results for linear or nonlinear systems[17,18,19,20,21,22]. For example, the authors in [17] investigated the adaptive sliding mode control issue for switched nonlinear systems with matched and mismatched uncertainties, designed the switched adaptive sliding mode control law and estimated the upper bound parameters of the matched uncertainty. The authors in [18] considered the synchronization of delayed chaotic neural networks with unknown disturbance via the observer-based sliding mode control, where the sliding surface involves an integral structure and a discontinuous controller.
In addition, synchronization has received considerable attention from various research fields, such as secure communication [3], engineering [23] and chemistry [24]. Synchronization phenomenons can be observed in many real systems such as neural systems, lasers and electronic circuits [25,26,27,28]. For example, the authors in [25] dealt with chaos synchronization for master-slave piecewise linear systems and provided some new sufficient conditions by using a Lyapunov approach and the so-called S-procedure. In [26], the authors designed a proportional-derivative (PD) controller for the master-slave synchronization of chaotic Lurie systems and presented a new synchronization criterion based on Lyapunov functions with a quadratic form of states and nonlinear functions of the systems. In [27], the authors studied the finite-time lag synchronization issue of master-slave complex networks with unknown signal propagation delays by the linear and adaptive error state feedback approaches and discovered that the setting time was related to initial values.
Motivated by the above discussion, we will consider the master-slave synchronization for uncertain neural networks with time-delay by using the sliding mode control method. The main contributions of our paper are as follows: (1) The master-slave synchronization for uncertain Markov jump neural networks with time-delay is considered by using the sliding mode control method, (2) the effect of time-delay and uncertainty for the neural networks stability are considered, and the uncertain parts in the neural networks only need be bounded other than any structure condition and (3) the provided sliding mode controller is very general.
The rest of this letter is organized as follows. In Section 2, the considered neural network model and some preliminaries are given. In Section 3, some sufficient conditions are proposed by the sliding mode control. In Section 4, a numerical example is provided to illustrate the effectiveness of the method proposed. In the last section, conclusions are presented.
Notation:
Rn: the n-dimensional Euclidean space; Rn×m: the set of all n×m real matrices; ||⋅||: the two-norm of a vector; ||⋅||1: its one-norm; In: the n order unit matrix; λmin(H): the minimum eigenvalues of matrix H; the notation X>Y, where X,Y are symmetric matrices, meaning that X−Y is a positive definite symmetric matrix. For a given matrix A∈Rn×n, AT denotes its transpose. sign(⋅) is the sign function. ∗ in a symmetric matrix denotes the symmetric terms.
Let (Ω,F,Ft,P) be a probability space related to an increasing family {Ft}t≥0 of the σ-algebras Ft⊂F, where Ω is the sample space. F is the σ-algebras of the sample space and P is the probability measure defined on F.
Consider the uncertain time delay neural networks with the Markovian jump defined on the probability space (Ω,F,Ft,P) as follows:
M:{˙xm(t)=−A(r(t))xm(t)+[B(r(t))+ΔB(r(t))]f(xm(t))+[C(r(t))+ΔC(r(t))]g(xm(t−d(t)))+J,xm(t)=φ(t),t∈[−d,0], | (2.1) |
where xm(t)=(xm1(t),xm2(t),...,xmn(t))T∈Rn is the state of neuron networks and A(r(t))∈Rn×n and B(r(t))∈Rn×m and C(r(t))∈Rn×m are coefficient matrices. ΔB(r(t)) and ΔC(r(t))∈Rn×m denote the system's uncertain parts and satisfy
[ΔB(r(t))ΔC(r(t))]=M(r(t))W(t)[N1(r(t))N2(r(t))], |
where matrix W(t) satisfies WT(t)W(t)≤I, M(r(t)),N1(r(t)),N2(r(t)) are some known matrices with appropriate dimensions. d(t) represents time delay and satisfies 0≤d(t)≤d and ˙d(t)≤μ≤1. f(xm(t)) and g(xm(t−d(t))) are the neuron activation functions, and J∈Rn is a constant vector. {r(t),t≥0} is a finite state Markov jumping process and represents the switching process among different modes, which takes values in a state space L={1,2,...,l}, and l is the number of modes. Let ∏=[πij]l×l denote the transition rate matrix, where the mode transition probabilities are
Pr{r(t+Δt)=j|r(t)=i}={πijΔt+o(Δt),i≠j,1+πiiΔt+o(Δt),i=j, | (2.2) |
where Δt>0 and limΔt→0o(Δt)Δt=0, πij satisfies πij>0 with i≠j and πii=−∑lj=1,j≠iπij for each mode i.
In order to be notional convenience, for the i-th mode, system (2.1) can be rewritten as
{˙xm(t)=−Aixm(t)+(Bi+ΔBi)f(xm(t))+(Ci+ΔCi)g(xm(t−d(t)))+J,xm(t)=φ(t),t∈[−d,0]. | (2.3) |
Let (2.3) be the master system, then the slave system is
S:{˙xs(t)=−A(r(t))xs(t)+B(r(t))f(xs(t))+C(r(t))g(xs(t−d(t)))+J+u(t),xs(t)=ψ(t),t∈[−d,0]. | (2.4) |
Definition 1. [29] Master system (2.3) and slave system (2.4) are said to be asymptotic synchronization if
limt→∞||xm(t)−xs(t)||=0 |
for any initial conditions.
In this paper, our objective is to design a suit controller u(t) such that the master system (2.3) and slave system (2.4) are in synchronization by using the sliding mode control method.
To the end, writing the state error e(t)=xs(t)−xm(t), and the corresponding state error system can be described by
{˙e(t)=−Aie(t)+BiF(e(t))+CiG(e(t−d(t)))−ΔBif(xm(t))−ΔCig(xm(t−d(t)))+u(t),e(t)=ψ(t)−φ(t),t∈[−d,0], | (2.5) |
where F(e(t))=f(xs(t))−f(xm(t)) and G(e(t−d(t)))=g(xs(t−d(t)))−g(xm(t−d(t))).
First, we take the sliding mode surface as
σi(t)=e(t)+∫t0[(Ai+Ki)e(θ)−BiF(e(θ))−CiG(e(θ−d(θ)))]dθ. | (2.6) |
Thus, the derivative of σi(t) is
˙σi(t)=˙e(t)+(Ai+Ki)e(t)−BiF(e(t))−CiG(e(t−d(t)))=Kie(t)−ΔBif(xm(t))−ΔCig(xm(t−d(t)))+u(t), | (2.7) |
where Ki(1≤i≤l) are some unknown matrices to be determined later. When the state trajectories reach the sliding mode surface, then ˙σi(t)=0 and σi(t)=0. So, we obtain the equivalent controller
ueq(t)=−Kie(t)+ΔBif(xm(t))+ΔCig(xm(t−d(t))). | (2.8) |
Substituting (2.8) into (2.5), we have
{˙e(t)=−(Ai+Ki)e(t)+BiF(e(t))+CiG(e(t−d(t))),e(t)=ψ(t)−φ(t),t∈[−d,0]. | (2.9) |
In this paper, the following assumptions for the neuron activation functions are needed.
Assumption 1. [30] Assume that each component of the nonlinear function f(⋅) and g(⋅) are continuous and bounded and satisfy
η−k≤fk(z1)−fk(z2)z1−z2≤η+k, |
θ−k≤gk(z1)−gk(z2)z1−z2≤θ+k,∀k=1,2,...,n, |
for any z1,z2∈R, where η−k>0,η+k>0,θ−k>0,θ+k>0 are some known positive constants.
From Assumption 1, it is easy to get the following inequalities
[eT(t)FT(e(t))][Θ1HΘ2H∗H][e(t)F(e(t))]≤0 | (2.10) |
and
[eT(t)GT(e(t−d(t)))][Θ3HΘ4H∗H][e(t)G(e(t−d(t)))]≤0, | (2.11) |
where H=diag{h1,h2,...,hn} is a positive definite diagonal matrix,
Θ1=diag{η−1η+1,η−2η+2,...,η−nη+n},Θ2=diag{−η−1+η+12,−η−2+η+22,...,−η−n+η+n2}, |
Θ3=diag{θ−1θ+1,θ−2θ+2,...,θ−nθ+n},Θ4=diag{−θ−1+θ+12,−θ−2+θ+22,...,−θ−n+θ+n2}. |
Assumption 2.[31] Assume that each component of the nonlinear functions f(⋅) and g(⋅) are bounded, which means that there exists positive scalars Bf and Bg such that
||fk(⋅)||≤Bf,||gk(⋅)||≤Bg |
for k=1,2,...,n.
Remark 1. In fact, the activation functions of neural networks are usually bounded. For example, the Logistic Sigmoid function h1(x)={1+e−ax}−1 and the threshold value function
h2(x)={1,x≥0,−1,x<0, |
and so on.
Throughout the paper, we need the following lemmas.
Lemma 1.[32] For any positive definite symmetric matrix W∈Rn×n and scalar τ>0, there is
∫tt−τxT(s)dsW∫tt−τx(s)ds≤τ∫tt−τxT(s)Wx(s)ds. |
Lemma 2.[33] The linear matrix inequality
[S11S12ST12S22]<0 |
is equivalent to the following condition
S22<0,S11−S12S−122ST12<0, |
where S11 and S22 are symmetric matrices.
Now, we will analyze the synchronization condition and construct the sliding mode controller.
Theorem 1. Under Assumption 1, if there exists positive definite symmetric matrices Pi,Wi,Ri∈Rn×n such that
Φ1i=[Φ1i,1101dRiPiBi−Θ2HPiCi−Θ4H−ATiRi−KTiRi∗−(1−μ)Wi0000∗∗−1dRi000∗∗∗−H0BTiRi∗∗∗∗−HCTiRi∗∗∗∗∗−1dRi]<0 | (3.1) |
for i=1,2,...,l, then the master system (2.1) and slave system (2.4) are in synchronization, where
Φ1i,11=−PiAi−ATiPi−PiKi−KTiPi+l∑j=1,j≠iπij(Pj−Pi)+Wi−1dRi−Θ1H−Θ3H. |
Proof. Constructing the following Lyapunov function
V(t)=eT(t)Pie(t)+∫tt−d(t)eT(θ)Wie(θ)dθ+∫0−d∫tt+s˙eT(θ)Ri˙e(θ)dθds. |
The derivative of V(t) along with the trajectories of system (2.9) is
˙V(t)=2eT(t)Pi˙e(t)+eT(t)∑lj=1πijPje(t)+eT(t)Wie(t)−(1−˙d(t))eT(t−d(t))Wie(t−d(t))+d˙eT(t)Ri˙e(t)−∫tt−d˙eT(θ)Ri˙e(θ)dθ≤2eT(t)Pi[−(Ai+Ki)e(t)+BiF(e(t))+CiG(e(t−d(t)))]+eT(t)∑lj=1,j≠iπij(Pj−Pi)e(t)+eT(t)Wie(t)−(1−μ)eT(t−d(t))Wie(t−d(t))+d[−(Ai+Ki)e(t)+BiF(e(t))+CiG(e(t−d(t)))]TRi⋅[−(Ai+Ki)e(t)+BiF(e(t))+CiG(e(t−d(t)))]−1d[e(t)−e(t−d)]TRi[e(t)−e(t−d)]. |
Letting ξ(t)=(eT(t),eT(t−d(t)),eT(t−d),FT(e(t))andGT(e(t−d(t)))T, one yields
˙V(t)≤ξT(t)[Φ2i+dΨTRiΨ]ξ(t), | (3.2) |
where
Φ2i=[Φ2i,1101dRiPiBiPiCi∗−(1−μ)Wi000∗∗−1dRi00∗∗∗00∗∗∗∗0], |
Ψ=[−Ai−Ki00BiCi], |
Φ2i,11=−PiAi−ATiPi−PiKi−KTiPi+l∑j=1,j≠iπij(Pj−Pi)+Wi−1dRi. |
It is noted that
ξT(t)[Φ2i+dΨTRiΨ]ξ(t)−[eT(t)FT(e(t))][Θ1HΘ2H∗H][e(t)F(e(t))]−[eT(t)GT(e(t−d(t)))][Θ3HΘ4H∗H][e(t)G(e(t−d(t)))]=ξT(t)[Φ3i+dΨTRiΨ]ξ(t), | (3.3) |
where
Φ3i=[Φ1i,1101dRiPiBi−Θ2HPiCi−Θ4H∗−(1−μ)Wi000∗∗−1dRi00∗∗∗−H0∗∗∗∗−H]. |
By using Lemma 2, we know that Φ3i+dΨTRiΨ<0 is equivalent to Φ1i<0. Thus, one obtains ˙V(t)<0 and system (2.9) is asymptotically stable, which shows that the master system (2.1) and slave system (2.4) are in synchronization. The proof is completed.
Theorem 2. Under Assumption 2 and the action of controller
u(t)=−Kie(t)−ζ(t)sign(σi(t)), | (3.4) |
master system (2.1) and slave system (2.4) are in synchronization, where
ζ(t)=Bf⋅||Mi||⋅||N1i||+Bg⋅||Mi||⋅||N2i||+α, |
and α>0 is a positive scalar.
Proof. Constructing the following Lyapunov function
U(t)=12σTi(t)σi(t), |
then
˙U(t)=σTi(t)˙σi(t)=σTi(t)[Kie(t)−ΔBif(xm(t))−ΔCig(xm(t−d(t)))−Kie(t)−ζ(t)sign(σi(t))]≤||σi(t)||[||ΔBi||⋅||f(xm(t))||+||ΔCi||⋅||g(xm(t−d(t)))||]−ζ(t)||σi(t)||. | (3.5) |
Because of
||ΔBi||⋅||f(xm(t))||+||ΔCi||⋅||g(xm(t−d(t)))||≤||Mi||[Bf⋅||N1i||+Bg⋅||N2i||], |
then
˙U(t)≤−α||σi(t)||=−α√2U(t). |
Thus, the state trajectories can attain the sliding mode surface in the finite time interval [0,T∗], where T∗≤√2U(0)2α. The proof is completed.
Remark 2. Compared with the uncertain system in [15], we especially considered the effect of time delay. From Theorems 1 and 2, we see that the designed sliding mode controller can realize the master-slave synchronization of the Markov jump neural networks.
Remark 3. The control gain Ki can be obtained from matrix inequality (3.1). However, (3.1) is not a linear matrix inequality. In order to solve it, we can take Pi=Ri.
From Theorem 1, we can obtain the following useful corollary as r(t)=1, which means that system (2.1) only has one mode.
Corollary 1. Under Assumption 1, if there exist positive definite symmetric matrices P,W,R∈Rn×n such that
Φ1=[Φ1,1101dRPB−Θ2HPC−Θ4H−ATR−KTR∗−(1−μ)W0000∗∗−1dR000∗∗∗−H0BTR∗∗∗∗−HCTR∗∗∗∗∗−1dR]<0, | (3.6) |
then the master system
{˙xm(t)=−Axm(t)+(B+ΔB)f(xm(t))+(C+ΔC)g(xm(t−d(t)))+J,xm(t)=φ(t),t∈[−d,0] | (3.7) |
and slave system
{˙xs(t)=−Axs(t)+Bf(xs(t))+Cg(xs(t−d(t)))+J+u(t),xs(t)=ψ(t),t∈[−d,0] | (3.8) |
are in synchronization, where
Φ1,11=−PA−ATP−PK−KTP+W−1dR−Θ1H−Θ3H. |
Consider master system (2.1) and slave system (2.4) with the following parameters
A1=[42105−20−26],A2=[5120700−18],B1=[0.20.60.20.2−0.200.2−0.1−0.1], |
B2=[0.20.50.30.2−0.400.3−0.1−0.2],C1=[0.50.4−0.30.50.10.10.50.10.5],C2=[0.8−0.30.30.10.20.2−0.60.30.2], |
Π=[−221−1],M1=[−0.20.50.7],M2=[0.30.8−0.4], |
N11=[0.1−0.2−0.1],N12=[0.05−0.1−0.03], |
N21=[−0.40.10.3],N22=[−0.10.020.07],μ=0.5,d=2, |
f(x(t))=[tanh(0.5x1(t)),tanh(0.4x2(t)),tanh(0.6x3(t))]T, |
g(x(t))=[tanh(0.3x1(t)),tanh(0.4x2(t)),tanh(0.2x3(t))]T, |
J1=0.5I3,J2=0.3I3,Θ1=Θ3=I3,Θ2=Θ4=−I3,w(t)=sin(t). |
By using the linear matrix inequality (LMI) toolbox in the MATLAB, we obtain the following solutions of inequality (3.1):
P1=[23.4405−7.37377.0525−7.373765.8658−20.94927.0525−20.949240.9154],P2=[21.2128−6.29397.4587−6.293961.5282−19.50707.4587−19.507038.4827], |
W1=[10.4452−3.21871.4795−3.218724.3005−7.12561.4795−7.125615.5240],W2=[5.7913−1.22002.4543−1.220016.4749−4.94762.4543−4.947610.7346], |
K1=[0.13660.0093−0.01880.00930.05500.0266−0.01880.02660.0902],K2=[0.13660.0093−0.01880.00930.05500.0266−0.01880.02660.0902], |
H=[76.637400059.824500045.2654]. |
For the initial values xm(0)=(−25,5,25)T and xs(0)=(30,−5,−25)T, Figure 1 is the Markovian jump process in different modes. Figure 2 is the state trajectories of error system (2.5), which shows that the error system is convergent and the master system synchronizes with the slave system. Figure 3 is the curve of the sliding mode surface, which shows that the state trajectories can arrive at the surface in a finite time interval.
In this paper, the master-slave synchronization for uncertain neural networks with time delay by using the sliding mode control method has been studied. An integral sliding mode surface and sliding mode controller was designed. Moreover, the state trajectories of the neural networks can reach the sliding mode surface in finite time under the action of the controller. Sufficient conditions in terms of linear matrix inequalities were presented to guarantee the neural networks asymptotical stability. Finally, an example was provided to illustrate the validity of the proposed design method. In the future, we will consider how to solve some physical problems by applying the obtained theoretical results.
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
This work is supported by the National Natural Science Foundation (NNSF) of China under Grant 11602134.
The authors declare that they have no conflicts of interest.
[1] | J. M. Mendel, Uncertain rule-based fuzzy systems, Springer International Publishing, 2017. https://doi.org/10.1007/978-3-319-51370-6 |
[2] |
H. Hagras, C. Wagner, Towards the wide spread use of type-2 fuzzy logic systems in real world applications, IEEE Comput. Intell. M., 7 (2012), 14–24. https://doi.org/10.1109/MCI.2012.2200621 doi: 10.1109/MCI.2012.2200621
![]() |
[3] |
J. M. Mendel, Type-2 fuzzy sets and systems: An overview, IEEE Comput. Intell. M., 2 (2007), 20–29. https://doi.org/10.1109/MCI.2007.380672 doi: 10.1109/MCI.2007.380672
![]() |
[4] |
M. H. F. Zarandi, B. Rezaee, I. B. Turksen, E. Neshat, A type-2 fuzzy rule-based expert system model for stock price analysis, Expert Syst. Appl., 36 (2009), 139–154. https://doi.org/10.1016/j.eswa.2007.09.034 doi: 10.1016/j.eswa.2007.09.034
![]() |
[5] |
Y. Chen, D. Z. Wang, S. C. Tong, Forecasting studies by designing Mamdani interval type-2 fuzzy logic systems: with combination of BP algorithms and KM algorithms, Neurocomputing, 174 (2016), 1133–1146. https://doi.org/10.1016/j.neucom.2015.10.032 doi: 10.1016/j.neucom.2015.10.032
![]() |
[6] |
A. Khosravi, S. Nahavandi, Load forecasting using interval type-2 fuzzy logic systems: Optimal type reduction, IEEE T. Ind. Electron., 10 (2014), 1055–1063. https://doi.org/10.1109/TII.2013.2285650 doi: 10.1109/TII.2013.2285650
![]() |
[7] |
M. Biglarbegian, W. W. Melek, J. M. Mendel, Design of novel interval type-2 fuzzy controllers for modular and reconfigurable robots: theory and experiments, IEEE T. Ind. Electron., 58 (2011), 1371–1384. https://doi.org/10.1109/TIE.2010.2049718 doi: 10.1109/TIE.2010.2049718
![]() |
[8] |
D. Z. Wang, Y. Chen, Study on permanent magnetic drive forecasting by designing Takagi Sugeno Kang type interval type-2 fuzzy logic systems, T. I. Meas. Control, 40 (2018), 2011–2023. https://doi.org/10.1177/0142331217694682 doi: 10.1177/0142331217694682
![]() |
[9] |
R. S. Rama, P. Latha, An effective torque ripple reduction for permanent magnet synchronous motor using ant colony optimization, Int. J. Fuzzy Syst., 17 (2015), 577–584. https://doi.org/10.1007/s40815-015-0077-5 doi: 10.1007/s40815-015-0077-5
![]() |
[10] |
O. Linda, M. Manic, Interval type-2 voter design for fault tolerant systems, Inform. Sci., 181 (2011), 2933–2950. https://doi.org/10.1016/j.ins.2011.03.008 doi: 10.1016/j.ins.2011.03.008
![]() |
[11] |
A. Niewiadomski, On finity, countability, cardinalities, and cylindric extensions of type-2 fuzzy sets in linguistic summarization of databases, IEEE T. Fuzzy Syst., 18 (2010), 532–545. https://doi.org/10.1109/TFUZZ.2010.2042719 doi: 10.1109/TFUZZ.2010.2042719
![]() |
[12] |
D. R. Wu, J. M. Mendel, Uncertainty measures for interval type-2 fuzzy sets, Inform. Sci., 177 (2007), 5378–2393. https://doi.org/10.1016/j.ins.2007.07.012 doi: 10.1016/j.ins.2007.07.012
![]() |
[13] |
A. Khosravi, S. Nahavandi, D. Creighton, D. Srinivasan, Interval type-2 fuzzy logic systems for load forecasting: a comparative study, IEEE T. Power Syst., 27 (2012), 1274–1282. https://doi.org/10.1109/TPWRS.2011.2181981 doi: 10.1109/TPWRS.2011.2181981
![]() |
[14] |
Y. Chen, Study on weighted Nagar-Bardini algorithms for centroid type-reduction of interval type-2 fuzzy logic systems, J. Intell. Fuzzy Syst., 34 (2018), 2417–2428. https://doi.org/10.3233/JIFS-171669 doi: 10.3233/JIFS-171669
![]() |
[15] |
J. M. Mendel, On KM algorithms for solving type-2 fuzzy sets problems, IEEE T. Fuzzy Syst., 21 (2013), 426–446. https://doi.org/10.1109/TFUZZ.2012.2227488 doi: 10.1109/TFUZZ.2012.2227488
![]() |
[16] |
T. Kumbasar, Revisiting Karnik-Mendel algorithms in the framework of linear fractional programming, Int. J. Approx. Reason., 82 (2017), 1–21. https://doi.org/10.1016/j.ijar.2016.11.019 doi: 10.1016/j.ijar.2016.11.019
![]() |
[17] |
J. M. Mendel, F. L. Liu, Super-exponential convergence of the Karnik-Mendel algorithms for computing the centroid of an interval type-2 fuzzy set, IEEE T. Fuzzy Syst., 15 (2007), 309–320. https://doi.org/10.1109/TFUZZ.2006.882463 doi: 10.1109/TFUZZ.2006.882463
![]() |
[18] |
D. R. Wu, J. M. Mendel, Perceptual reasoning for perceptual computing: a similarity based approach, IEEE T. Fuzzy Syst., 17 (2009), 1397–1411. https://doi.org/10.1109/TFUZZ.2009.2032652 doi: 10.1109/TFUZZ.2009.2032652
![]() |
[19] |
D. R. Wu, J. M. Mendel, Enhanced Karnik-Mendel algorithms, IEEE T. Fuzzy Syst., 17 (2009), 923–934. https://doi.org/10.1109/TFUZZ.2008.924329 doi: 10.1109/TFUZZ.2008.924329
![]() |
[20] |
X. W. Liu, J. M. Mendel, D. R. Wu, Study on enhanced Karnik-Mendel algorithms: initialization explanations and computation improvements, Inform. Sci., 184 (2012), 75–91. https://doi.org/10.1016/j.ins.2011.07.042 doi: 10.1016/j.ins.2011.07.042
![]() |
[21] |
Y. Chen, Study on sampling-based discrete noniterative algorithms for centroid type-reduction of interval type-2 fuzzy logic systems, Soft Comput., 24 (2020), 11819–11828. https://doi.org/10.1007/s00500-020-04998-2 doi: 10.1007/s00500-020-04998-2
![]() |
[22] |
O. Castillo, P. Melin, E. Ontiveros, C. Peraza, P. Ochoa, F. Valdez, et al., A high-speed interval type 2 fuzzy system approach for dynamic parameter adaptation in metaheuristics, Eng. Appl. Artif. Intell., 85 (2019), 666–680. https://doi.org/10.1016/j.engappai.2019.07.020 doi: 10.1016/j.engappai.2019.07.020
![]() |
[23] |
G. M. Méndez, M. D. L. A. Hernandez, Hybrid learning mechanism for interval A2-C1 type-2 non-singleton type-2 Takagi-Sugeno-Kang fuzzy logic systems, Inform. Sci., 220 (2013), 149–169. https://doi.org/10.1016/j.ins.2012.01.024 doi: 10.1016/j.ins.2012.01.024
![]() |
[24] |
J. M. Mendel, Type-2 fuzzy sets and systems: an overview, IEEE Comput. Intell. Mag., 2 (2007), 20–29. https://doi.org/10.1109/MCI.2007.380672 doi: 10.1109/MCI.2007.380672
![]() |
[25] |
Y. Chen, D. Z. Wang, Study on centroid type-reduction of general type-2 fuzzy logic systems with weighted enhanced Karnik-Mendel algorithms, Soft Comput., 22 (2018), 1361–1380. https://doi.org/10.1007/s00500-017-2938-3 doi: 10.1007/s00500-017-2938-3
![]() |
[26] |
Y. Chen, D. Z. Wang, Study on centroid type-reduction of general type-2 fuzzy logic systems with weighted Nie-Tan algorithms, Soft Comput., 22 (2018), 7659–7678. https://doi.org/10.1007/s00500-018-3551-9 doi: 10.1007/s00500-018-3551-9
![]() |
[27] |
D. R. Wu, Approaches for reducing the computational cost of interval type-2 fuzzy logic systems: overview and comparisons, IEEE T. Fuzzy Syst., 21 (2013), 80–99. https://doi.org/10.1109/TFUZZ.2012.2201728 doi: 10.1109/TFUZZ.2012.2201728
![]() |
[28] | T. Wang, Y. Chen, S. C. Tong, Fuzzy reasoning models and algorithms on type-2 fuzzy sets, Int. J. Innov. Comput. Inform. Control, 24 (2008), 2451–2460. |
[29] |
J. W. Li, R. John, S. Coupland, G. Kendall, On Nie-Tan operator and type-reduction of interval type-2 fuzzy sets, IEEE T. Fuzzy Syst., 26 (2018), 1036–1039. https://doi.org/10.1109/TFUZZ.2017.2666842 doi: 10.1109/TFUZZ.2017.2666842
![]() |
[30] |
S. Greenfield, F. Chiclana, Accuracy and complexity evaluation of defuzzification strategies for the discretised interval type-2 fuzzy set, Int. J. Approx. Reason., 54 (2013), 1013–1033. https://doi.org/10.1016/j.ijar.2013.04.013 doi: 10.1016/j.ijar.2013.04.013
![]() |
[31] |
E. Ontiveros-Robles, P. Melin, O. Castillo, New methodology to approximate type-reduction based on a continuous root-finding karnik mendel algorithm, Algorithms, 10 (2017), 77–96. https://doi.org/10.3390/a10030077 doi: 10.3390/a10030077
![]() |
[32] |
Y. Chen, J. X. Wu, J. Lan, Study on reasonable initialization enhanced Karnik-Mendel algorithms for centroid type-reduction of interval type-2 fuzzy logic systems, AIMS Math., 5 (2020), 6149–6168. https://doi.org/10.3934/math.2020395 doi: 10.3934/math.2020395
![]() |
[33] |
J. M. Mendel, X. W. Liu, Simplified interval type-2 fuzzy logic systems, IEEE T. Fuzzy Syst., 21 (2013), 1056–1069. https://doi.org/10.1109/TFUZZ.2013.2241771 doi: 10.1109/TFUZZ.2013.2241771
![]() |
[34] |
M. A. Khanesar, A. Jalalian, O. Kaynak, H. Gao, Improving the speed of center of sets type reduction in interval type-2 fuzzy systems by eliminating the need for sorting, IEEE T. Fuzzy Syst., 25 (2017), 1193–1206. https://doi.org/10.1109/TFUZZ.2016.2602392 doi: 10.1109/TFUZZ.2016.2602392
![]() |
[35] |
J. M. Mendel, General type-2 fuzzy logic systems made simple: A tutorial, IEEE T. Fuzzy Syst., 22 (2014), 1162–1182. https://doi.org/10.1109/TFUZZ.2013.2286414 doi: 10.1109/TFUZZ.2013.2286414
![]() |
[36] |
C. H. Hsu, C. F. Juang, Evolutionary robot wall-following control using type-2 fuzzy controller with species-de-activated continuous ACO, IEEE T. Fuzzy Syst., 21 (2013), 100–112. https://doi.org/10.1109/TFUZZ.2012.2202665 doi: 10.1109/TFUZZ.2012.2202665
![]() |
[37] |
Y. Chen, D. Z. Wang, W. Ning, Forecasting by TSK general type-2 fuzzy logic systems optimized with genetic algorithms, Optim. Control Appl. Method., 39 (2018), 393–409. https://doi.org/10.1002/oca.2353 doi: 10.1002/oca.2353
![]() |
[38] |
F. Gaxiola, P. Melin, F. Valdez, J. R. Castro, O. Castillo, Optimization of type-2 fuzzy weights in backpropagation learning for neural networks using GAs and PSO, Appl. Soft Comput., 38 (2016), 860–871. https://doi.org/10.1016/j.asoc.2015.10.027 doi: 10.1016/j.asoc.2015.10.027
![]() |
[39] |
Y. Chen, D. Z. Wang, Forecasting by general type-2 fuzzy logic systems optimized with QPSO algorithms, Int. J. Control Autom. Syst., 15 (2017), 2950–2958. https://doi.org/10.1007/s12555-017-0793-0 doi: 10.1007/s12555-017-0793-0
![]() |
[40] |
Y. Maldonado, O. Castillo, P. Melin, Particle swarm optimization of interval type-2 fuzzy systems for FPGA applications, Appl. Soft Comput., 13 (2013), 496–508. https://doi.org/10.1016/j.asoc.2012.08.032 doi: 10.1016/j.asoc.2012.08.032
![]() |
[41] |
E. Ontiveros-Robles, P. Melin, O. Castillo, Comparative analysis of noise robustness of type 2 fuzzy logic controllers, Kybernetika, 54 (2018), 175–201. https://doi.org/10.14736/kyb-2018-1-0175 doi: 10.14736/kyb-2018-1-0175
![]() |
[42] |
L. Cervantes, O. Castillo, Type-2 fuzzy logic aggregation of multiple fuzzy controllers for airplane flight control, Inform. Sci., 324 (2015), 247–256. https://doi.org/10.1016/j.ins.2015.06.047 doi: 10.1016/j.ins.2015.06.047
![]() |
[43] |
O. Castillo, L. Amador-Angulo, J. R. Castro, M. Garcia-Valdez, A comparative study of type-1 fuzzy logic systems, interval type-2 fuzzy logic systems and generalized type-2 fuzzy logic systems in control problems, Inform. Sci., 354 (2016), 257–274. https://doi.org/10.1016/j.ins.2016.03.026 doi: 10.1016/j.ins.2016.03.026
![]() |
[44] |
C. W. Tao, J. S. Taur, C. W. Chang, Y. H. Chang, Simplified type-2 fuzzy sliding controller for wing rocket system, Fuzzy Set. Syst., 207 (2012), 111–129. https://doi.org/10.1016/j.fss.2012.02.015 doi: 10.1016/j.fss.2012.02.015
![]() |
[45] |
D. R. Wu, J. M. Mendel, Recommendations on designing practical interval type-2 fuzzy systems, Eng. Appl. Artif. Intell., 85 (2019), 182–193. https://doi.org/10.1016/j.engappai.2019.06.012 doi: 10.1016/j.engappai.2019.06.012
![]() |
[46] |
S. C. Tong, Y. M. Li, Observer-based adaptive fuzzy backstepping control of uncertain pure-feedback systems, Sci. China Inform. Sci., 57 (2014), 1–14. https://doi.org/10.1007/s11432-013-5043-y doi: 10.1007/s11432-013-5043-y
![]() |
[47] |
M. Deveci, I. Z. Akyurt, S. Yavuz, GIS-based interval type-2 fuzzy set for public bread factory site selection, J. Enterp. Inform. Manag., 31 (2018), 820–847. https://doi.org/10.1108/JEIM-02-2018-0029 doi: 10.1108/JEIM-02-2018-0029
![]() |
[48] |
S. C. Tong, Y. M. Li, Robust adaptive fuzzy backstepping output feedback tracking control for nonlinear system with dynamic uncertainties, Sci. China Inform. Sci., 53 (2010), 307–324. https://doi.org/10.1007/s11432-010-0031-y doi: 10.1007/s11432-010-0031-y
![]() |
[49] | F. Y. Wang, H. Mo, Some fundamental issues on type-2 fuzzy sets, Acta Autom. Sin., 43 (2017), 1114–1141. |
[50] |
H. Mo, F. Y. Wang, M. Zhou, R. Li, Z. Xiao, Footprint of uncertainty for type-2 fuzzy sets, Inform. Sci., 272 (2014), 96–110. https://doi.org/10.1016/j.ins.2014.02.092 doi: 10.1016/j.ins.2014.02.092
![]() |
[51] |
Y. Chen, J. X. Yang, Study on center-of-sets type-reduction of interval type-2 fuzzy logic systems with noniterative algorithms, J. Intell. Fuzzy Syst., 40 (2021), 11099–11106. https://doi.org/10.3233/JIFS-202264 doi: 10.3233/JIFS-202264
![]() |
[52] |
X. Tao, J. Yi, Z. Pu, T. Xiong, Robust adaptive tracking control for hypersonic vehicle based on interval type-2 fuzzy logic system and small-gain approach, IEEE T. Cybernetics, 51 (2021), 2504–2517. https://doi.org/10.1109/TCYB.2019.2927309 doi: 10.1109/TCYB.2019.2927309
![]() |
[53] |
Y. Chen, J. X. Yang, Design of back propagation optimized Nagar-Bardini structure-based interval type-2 fuzzy logic systems for fuzzy identification, T. I. Meas. Control, 43 (2021), 2780–2787. https://doi.org/10.1177/01423312211006635 doi: 10.1177/01423312211006635
![]() |
[54] | L. Wu, F. Qian, L. Wang, X. Ma, An improved type-reduction algorithm for general type-2 fuzzy sets, Inform. Sci., 2022. |
[55] |
Y. Chen, Study on weighted-based noniterative algorithms for computing the centroids of general type-2 fuzzy sets, Int. J. Fuzzy Syst., 24 (2022), 587–606. https://doi.org/10.1007/s40815-021-01166-y doi: 10.1007/s40815-021-01166-y
![]() |
[56] |
C. Chen, D. Wu, J. M. Garibaldi, R. I. John, J. Twycross, J. M. Mendel, A comprehensive study of the efficiency of type-reduction algorithms, IEEE T. Fuzzy Syst., 29 (2020), 1556–1566. https://doi.org/10.1109/TFUZZ.2020.2981002 doi: 10.1109/TFUZZ.2020.2981002
![]() |