MPE Mathematical Problems in Engineering 1563-5147 1024-123X Hindawi Publishing Corporation 10.1155/2015/514583 514583 Research Article Fast Sampling Control of Singularly Perturbed Systems with Actuator Saturation and L2 Disturbance Miao Yanzi Ma Lei Ma Xiaoping Zhou Linna Liu Wanquan School of Information and Electrical Engineering China University of Mining and Technology Xuzhou 221116 China cumt.edu.cn 2015 1692015 2015 27 04 2015 03 08 2015 1692015 2015 Copyright © 2015 Yanzi Miao et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

We will consider the problem of fast sampling control for singularly perturbed systems subject to actuator saturation and L2 disturbance. A sufficient condition for the existence of a state feedback controller is proposed. Under this controller, the boundedness of the trajectories in the presence of L2 disturbances is guaranteed for any singular perturbation parameter less than or equal to a predefined upper bound. To improve the capacity of disturbance tolerance and disturbance rejection, two convex optimization problems are formulated. Finally, a numerical example is presented to demonstrate the effectiveness of the main results of this paper.

1. Introduction

Many practical physical systems consist of subsystems operating on different time scales. Applying the normal control methods to these systems usually lead to ill-conditioned numerical problems. To overcome the numerical problem, singular perturbation theory was introduced to the field of control system and widely used in practice . In this framework, a multiple time-scale system is modeled by a singularly perturbed system (SPS) with a small positive parameter such that the degree of separation between fast and slow modes can be determined. For example, in a power system, the flux linkages of the rotor windings are fast variables while the emf behind transient reactance, the generator’s rotor angle in radians, and the actual rotor speed are the slow ones. Thus the power system can be modeled as a SPS, where the torque of the winding represents the perturbation parameter ε .

Since most of the modern control systems are implemented by computer, sampling control of SPSs has been widely investigated. There are three sampling modes for SPSs: multirate, slow, and fast sampling mode. Under multirate sampling mode, the slow and fast states are measured at different sampling rate. In , a multirate sampling model predictive control method is proposed for large-scale nonlinear uncertain systems. In , a multirate sampling composite controller is designed. Slow sampling control is usually under the assumption that the fast subsystem is stable [5, 6]. Slow sampling discrete-time SPS is considered in [7, 8]. If the fast subsystem is not stable, fast sampling control is necessary [9, 10]. Fast sampling control of SPSs with disturbance is considered in . In , H controller design method together with a stability bound optimization method is proposed, and a less conservative method is proposed in . But the above achievements do not take into account the actuator saturation, which is common in practice. It is known that actuator saturation may force the systems to work in the open-loop and thus destroy stability of control systems . Thus many research efforts have been devoted to analysis and design of control systems with actuator saturation . Recently, SPSs with actuator saturation are considered. Besides basin of attraction, stability bound is also an important stability index for SPSs with actuator saturation. In , state feedback controller is designed and the basin of attraction is estimated. The obtained results guarantee the existence of the stability bound but can not present an estimate of the bound. Many results have been proposed for estimating or enlarging the stability bound of the SPSs without actuator saturation . In , continuous-time SPS with actuator saturation is considered and a state feedback controller is designed to achieve a desired stability bound while the basin of attraction is optimized. To the best knowledge of the authors, the fast sampling control problem of the SPSs with actuator saturation and disturbance has not been considered.

This paper focuses on fast sampling control of SPSs subject to actuator saturation and L2 disturbance. First, a state feedback controller design method is proposed such that the trajectories of the closed-loop SPSs starting from a bounded set remain bounded for any allowable singular perturbation parameter and L2 disturbance. Then, a method to enlarge the capacity of disturbance tolerance is proposed in terms of linear matrix inequalities (LMIs). Furthermore, a convex optimization problem is formulated to optimize the disturbance rejection. Finally, an example is given to show the effectiveness of the proposed results.

The rest of this paper is organized as follows: Section 2 provides the problems under consideration. Section 3 gives the main results of this paper. In Section 4, an example is presented to demonstrate the proposed approaches. Section 5 makes a conclusion of the paper.

Notations. The superscript T stands for matrix transpose. For a matrix M, the notation M-T denotes the transpose of the inverse matrix of M and M(i) denotes ith row of M. Let QRn×n be a positive definite matrix. An ellipsoid Ω(Q,ρ) is defined as Ω(Q,ρ){ηRnηTQηρ}.

2. Problem Formulation

The fast sampling model of SPSs with actuator saturation and L2 disturbance is described by the following compact form: (1)xk+1=Aεxk+Bεsatuk+Eεwk,zk=Cxk,where (2)xk=x1kx2kRn,Aε=I+εA11εA12A21A22,Bε=εB1B2,Eε=εE1E2,C=C1C2.ε represents the singular perturbation parameter and is assumed to be available for controller design in this paper. x1Rn1 and x2Rn2 are the state variables, u(k) is the control input, z(k) is the output of the system, A11, A12, A21, A22, B1, B2 and E1, E2 are constant matrices of appropriate dimensions and sat(·) is a componentwise saturation map RmRm defined by sat(uj)=signujmin{1,uj}, j=1,2,,m, and w(k) is the L2 disturbance which belongs to (3)Wα2w:R+Rq:k=0+wTkwkα,where α is a positive number.

Under the state feedback controller (4)uk=Kεxk,the closed-loop system can be described by (5)xk+1=Aεxk+BεsatKεxk+Eεwk,zk=Cxk.

The problems under consideration are as follows.

Problem 1.

Given ε0>0 and α>0, design a state feedback controller (4), such that, for any ε(0,ε0] and wWα2, all the trajectories of the closed-loop system starting from inside in Ω(P-1(ε),1) will remain inside of Ω(P-1(ε),1+ηα) with η>0 and P(ε)>0 to be determined.

Problem 2.

Given ε0>0, design a state feedback controller (4) to maximize α, such that for any ε(0,ε0] and wWα2 all the trajectories of the closed-loop system starting from the origin are bounded.

Problem 3.

Given ε0>0 and α>0, design a state feedback controller (4) to minimize γ>0, such that, for any ε(0,ε0] and wWα2, the restricted L2 gain from output to disturbance of the closed-loop system with zero initial condition is less than γ under the control of (4).

Remark 4.

The upper bound ε0 characterizes the robustness of the system performance with respect to ε. The disturbance tolerance bound α describes the largest disturbance that can be tolerated. The disturbance rejection γ means the largest ratio between the L2 norms of the output and the disturbance.

There are two common tools to handle the saturation nonlinearities, one is the sector bound approach , and the other is the convex hull approach [13, 18]. The latter is adopted in this paper since it has been shown to be less conservative than the former . The related preliminaries are recalled in the following.

For a given matrix KRm×n, denote the ith row of matrix K as ki and define (6)LK=ηRn:kiη1,i1,m.

Let D be the set of m×m diagonal matrices whose diagonal elements are either 1 or 0. There are 2m elements in D. Suppose these elements of D are labeled as Di, i[1,2m]. Denote Di-=I-Di. Clearly, Di-D if DiD.

Lemma 5 (see [<xref ref-type="bibr" rid="B13">13</xref>]).

Let K,HRm×n. Then, for any xL(H), it holds that (7)satKxcoDiKx+Di-Hx,i1,2m,where co stands for the convex hull.

3. Main Results 3.1. Controller Design

This subsection will present a solution to Problem 1.

Theorem 6.

Given ε0>0 and α>0, if there exist symmetric matrices P11Rn1n1, P22Rn2n2 and matrices P12Rn1n2, Z1Rmn1, Z2Rmn2, Y1Rmn1, Y2Rmn2, as well as a positive scalar η satisfying (8)P22ϕ1Tϕ2T0-ϕ3-ϕ3TP12-ϕ4TE1P22E2ηI>0,i1,2m,(9)ε0P11ε0P12ε0ϕ3Tε0ϕ4T0P22ε0P12TA11T+ϕ1Tε0P12TA21T+ϕ2T0-ϕ3-ϕ3TP12-ϕ4TE1P22E2ηI>0,i1,2m,(10)P22Y2rTY2r11+ηα>0,r=1,2,,m,(11)ε0P11ε0P12ε0Y1rTε0P12TP22Y2rTε0Y1rTY2r11+ηα>0,r=1,2,,m,(12)ϕ1=A12P22+B1DiZ2+B1Di-Y2,ϕ2=A22P22+B2DiZ2+B2Di-Y2,ϕ3=A11P11+A12P12T+B1DiZ1+B1Di-Y1,ϕ4=A21P11+B2DiZ1+B2Di-Y1+A22P12T,i1,2m,then, for any ε(0,ε0] and wWα2, all of the trajectories of the closed-loop system (5) starting within Ω(P-1(ε),1) will remain inside of Ω(P-1(ε),1+ηα) with P(ε)=εP11εP12εP12TP22>0.

And the state feedback controller gain matrix is given by(13)Kε=Z1Z2P11εP12P12TP22-1.

Proof.

From (10) and (11), it follows that, for any ε(0,ε0], (14)εP11εP12εY1rTεP12TP22Y2rTεY1rTY2r11+ηα>0,r=1,2,,m.

Let Zε=εZ1Z2, Yε=εY1Y2. Then (14) is equivalent to (15)PεYεrTYεr11+ηα>0,r=1,2,,m,which implies that (16)P-1εHεrTHεr11+ηα>0,r=1,2,,m,where Hε=YεP-1(ε).

By Schur complement, it follows from (16) that (17)P-1ε>HεrT1+ηαHεr,r=1,2,,m,which implies that Ω(P-1(ε),1+ηα)L(Hε), ε(0,ε0]. Then by Lemma 5, we have (18)Aεxk+BεsatKεxkcoAεxk+BεDiKε+Di-Hεxk,ε0,ε0,i1,2m.

From (8) and (9), it follows that(19)εP11εP12εϕ3Tεϕ4T0P22εP12TA11T+ϕ1TεP12TA21T+ϕ2T0-ϕ3-ϕ3TP12-ϕ4TE1P22E2ηI>0,ε0,ε0,i1,2m.

Before and after multiplying (19) by (20)I00000I00000εI00000I00000Iand its transpose, respectively, we have (21)εP11εP12ε2ϕ3Tεϕ4T0P22ε2P12TA11T+εϕ1TεP12TA21T+ϕ2T0ε2-ϕ3-ϕ3TεP12-εϕ4TεE1P22E2ηI>0,ε0,ε0,i1,2m.

Before and after multiplying (21) by the matrix (22)I00000I000I0I00000I00000Iand its transpose, respectively, we have(23)εP11εP12εP11+ε2ϕ3Tεϕ4T0P22εP12T+ε2P12TA11T+εϕ1TεP12TA21T+ϕ2T0εP11εP12εE1P22E2ηI>0,ε0,ε0,i1,2m.

Taking into account the definitions of Aε, Bε, Eε, and P(ε), we can rewrite (23) as (24)PεAεPε+BεDiZε+BεDi-YεT0PεEεηI>0,ε0,ε0,i1,2m.

Define (25)Kε=ZεP-1ε.

Before and after multiplying (24) by diag(P-1(ε),P-1(ε),I), we have (26)P-1εAε+BεDiKε+BεDi-HεTP-1ε0P-1εP-1εEεηI>0,ε0,ε0,i1,2m.

By Schur complement, inequality (26) is equivalent to (27)Ξ1Ξ2Ξ2TΞ3-ηI<0,where(28)Ξ1=Aε+BεDiKε+BεDi-HεTP-1εAε+BεDiKε+BεDi-Hε-P-1ε,Ξ2=Aε+BεDiKε+BεDi-HεTP-1εEε,Ξ3=EεTP-1εEε,ε0,ε0,i1,2m.

Define an ε-dependent Lyapunov function:(29)Vx=xTP-1εx.

Calculating the difference of V(k) along the trajectories of the closed-loop system (5), and using (18), we have (30)ΔVx=Vxk+1-Vxk=Aεxk+Bεsatu+EεwkTP-1εAεxk+Bεsatu+Eεwk-xTkP-1εxkmaxi1,2mAεxk+BεDiKε+Di-Hεxk+EεwkTP-1εAεxk+BεDiKε+Di-Hεxk+Eεwk-xTkP-1εxk=maxi1,2mxTkAε+BεDiKε+Di-1HεTP-1εAε+BεDiKε+Di-1Hε-P-1εxk+xTkAε+BεDiKε+Di-1HεTP-1εEεwk+wTkEεTP-1εAε+BεDiKε+Di-1Hεxk+wTkEεTP-1εEεwk,xkΩP-1ε,1+ηα,ε0,ε0,wWα2.

It follows that (31)ΔVx-ηwTkwk=ξTkΞ1Ξ2Ξ2TΞ3-ηIξk0,xkΩP-1ε,1+ηα,ε0,ε0,wWα2,with (32)ξk=xkwk.

Then we have (33)ΔVxηwTkwk,xkΩP-1ε,1+ηα,ε0,ε0,wWα2.

Summing up both sides of (33) from 0 to m, we can get (34)Vxm+1Vx0+ηk=0mwTkwkVx0+αη,xkΩP-1ε,1+ηα,ε0,ε0,wWα2,which shows that when x(0)Ω(P-1(ε),1), we can get V(x(m+1))1+ηα, that is, x(m+1)Ω(P-1(ε),1+ηα).

And, according to (25), the controller gain matrix is (35)Kε=ZεP-1ε=εZ1Z2εP11εP12εP12TP22-1=Z1Z2P11εP12P12TP22-1.

Remark 7.

According to (33), when the disturbance w(k)=0, it holds that ΔV(x)<0, x0, which means that the closed-loop system (5) is locally asymptotically stable. In this case, the ellipsoid Ω(P-1(ε),1) is an estimation of the basin of attraction of the closed-loop system. In addition, when ε is small enough, an ε-independent gain matrix can be computed by (36)K=limε0+Z1Z2P11εP12P12TP22-1=Z1Z2P110P12TP22-1=Z1-Z2P22-1P12TP11-1Z2P22T.

3.2. Disturbance Tolerance

The ability of the closed-loop system to tolerate the disturbance is characterized by α. Based on Theorem 6, we have the following corollary which can be used to maximize α.

Corollary 8.

Given ε0>0 and α>0, if there exist symmetric matrices P11Rn1n1, P22Rn2n2 and matrices P12Rn1n2, Z1Rmn1, Z2Rmn2, Y1Rmn1, Y2Rmn2 satisfying(37)P22ϕ1Tϕ2T0-ϕ3-ϕ3TP12-ϕ4TE1P22E2I>0,i1,2m,(38)ε0P11ε0P12ε0ϕ3Tε0ϕ4T0P22ε0P12TA11T+ϕ1Tε0P12TA21T+ϕ2T0-ϕ3-ϕ3TP12-ϕ4TE1P22E2I>0,i1,2m,(39)P22Y2rY2rT1α>0,(40)ε0P11ε0P12ε0Y1rε0P12TP22Y2rε0Y1rTY2rT1α>0,where ϕ1, ϕ2, ϕ3, ϕ4 are defined in Theorem 6, then all the trajectories of the closed-loop system (5) starting from the origin will still remain inside of Ω(P-1(ε),α). And the state feedback controller gain matrix is given by (41)Kε=Z1Z2P11εP12P12TP22-1.

Proof.

According to (39) and (40), for any ε(0,ε0], we have (42)εP11εP12εY1rεP12TP22Y2rεY1rTY2rT1α>0,which is equivalent to (43)P-1εHεrTHεr1α>0,r=1,2,,m.

By Schur complement, it follows from (43) that (44)P-1ε>HεrTαHεr,r=1,2,,m;thus Ω(P-1(ε),α)L(Hε), ε(0,ε0].

Similarly to (34) in the proof for Theorem 6, for the trajectories start from the origin, we have (45)Vxm+1V0+k=0mwTkwkV0+α=α.

This complete the proof.

From Corollary 8, the bigger α means the better disturbance tolerance ability. To get the best disturbance tolerance ability we formulate the following optimization problem:(46)maxP11,P12,P22,Z1,Z2,Y1,Y2αs.t.37,38,39,40,where Ω(S,1) represents the initial condition set.

Let μ=1/α. Then inequalities (39) and (40) can be rewritten as (47)P22Y2rTY2rμ>0,r=1,2,,m,ε0P11ε0P12ε0Y1rTε0P12TP22Y2rTε0Y1rTY2rμ>0,r=1,2,,m.

Then the optimization problem (46) can be converted into (48)minP11,P12,P22,Z1,Z2,Y1,Y2μs.t.37,38,47.

Solving the optimization problem (48) yields the minimal value μ. Then the largest disturbance that can be tolerated by the closed-loop system at zero initial condition is bounded by α=1/μ. Therefore, Problem 2 can be solved by the optimization problem (48).

3.3. Disturbance Rejection

Problem 3 will be considered in this subsection.

Theorem 9.

Given ε0>0, α>0, and γ>0, if there exist symmetric matrices P11Rn1n1, P22Rn2n2 and matrices P12Rn1n2, Z1Rmn1, Z2Rmn2, Y1Rmn1, Y2Rmn2 satisfying (49)P22ϕ1Tϕ2T0P22C2T-ϕ3-ϕ3TP12-ϕ4TE1-ϕ5TP22E20γ2I0I>0,(50)ε0P11ε0P12ε0ϕ3Tε0ϕ4T0ε0ϕ5TP22ε0P12TA11T+ϕ1Tε0P12TA21T+ϕ2T0ε0P12TC1T+P22C2T-ϕ3-ϕ3TP12-ϕ4TE1-ϕ5TP22E20γ2I0I>0,i1,2m,(51)P22Y2Y2T1α>0,(52)ε0P11ε0P12ε0Y1ε0P12TP22Y2ε0Y1TY2T1α>0,where(53)ϕ1=A12P22+B1DiZ2+B1Di-Y2,ϕ2=A22P22+B2DiZ2+B2Di-Y2,ϕ3=A11P11+A12P12T+B1DiZ1+B1Di-Y1,ϕ4=A21P11+B2DiZ1+B2Di-Y1+A22P12T,i1,2m,ϕ5=C1P11+C2P12,then the L2 gain from w to z of the closed-loop system (5) with x(0)=0 is less than γ. And the state feedback controller gain matrix is given by (54)Kε=Z1Z2P11εP12P12TP22-1.

Proof.

Let Pε=εP11εP12εP12TP22>0, Zε=εZ1Z2, Yε=εY1Y2, Hε=YεP-1ε, Kε=ZεP-1(ε).

Similarly to the proof for Theorem 6, it follows from (49) and (50) that (55)PεAεPε+BεDiZε+BεDi-YεT0PεCTPεEε0γ2I0I>0,ε0,ε0,i1,2m.

Applying the Schur complement formula to (55), we have (56)Ξ1+CTCΞ2Ξ2TΞ3-γ2I<0,where(57)Ξ1=Aε+BεDiKε+BεDi-HεTP-1εAε+BεDiKε+BεDi-Hε-P-1ε,Ξ2=Aε+BεDiKε+BεDi-HεTP-1εEε,Ξ3=EεTP-1εEε,ε0,ε0,i1,2m.

From (51) and (52), for any ε(0,ε0], it follows that (58)εP11εP12εY1εP12TP22Y2εY1TY2T1α>0.

From the proof for Theorem 6, we can get (59)P-1ε>HεrTαHεr,r=1,2,,m,which shows that Ω(P-1(ε),α)L(Hε),ε(0,ε0].

Define an ε-dependent Lyapunov function (60)Vx=xTP-1εx.

Similarly to proof for Theorem 6, it follows from (56) that (61)ΔVxk+zTkzk-γ2wTkwk<0,xΩP-1ε,α.

Then, summing up left and right of (61), respectively, with x(0)=0, yields that (62)Vxm+1-0+k=0mzTkzk-γ2wTkwk<0,which implies (63)k=0mzTkzk<γ2k=0jwTkwk.Thus the L2 gain from w to z of the closed-loop system with x(0)=0 is less than γ. This completes the proof.

By Theorem 9, the minimal L2 gain can be obtained by solving the following optimization problem: (64)minP11,P12,P22,Z1,Z2,Y1,Y2γ2s.t.49,50,51,52.

Remark 10.

As mentioned in Section 1, discretization of a continuous-time SPS can lead to different discrete-time models depending on the sampling rate. Since the structure of fast sampling models is different from that of slow sampling models, it is not easy to generalize the proposed results to slow sampling control of SPSs, as will be considered in our future work.

4. Examples

This section will illustrate the proposed results by an example.

Consider an inverted pendulum system controlled by DC motor via a gear train. The model, which was first established in , is described by (65)x˙1t=x2t,x˙2t=glsinx1t+NKmml2x3t,Lax˙3t=-KbNx2t-Rax3t+ut+wt,where x1(t)=θp(t) denotes the the angle (rad) of the pendulum from the vertical upward, x2(t)=θ˙p(t), x3(t)=Ia(t) denotes the current of the motor, u(t) is the control input voltage, w(t) is the disturbance, Km is the motor torque constant, Kb is the back emf constant, N is the gear ratio, and La is the inductance which is usually a small positive constant. The parameters for the plant are as follows: g=9.8m/s2, N=10, l=1m, m=1kg, Km=0.1 Nm/A, Kb=0.1 Vs/rad, Ra=1 Ω, and La=0.05 H and the input voltage is required to satisfy u1. Note that La represents the singular perturbation parameter of the system. Substituting the parameters into (65) and linearizing the equations, we have (66)x˙1t=x2t,x˙2t=9.8x1t+x3t,εx˙3t=-x2t-x3t+ut+wt,where ε=La.

The equilibrium point of system (66), that is, xe=000T, corresponds to the upright rest position of the inverted pendulum. We will design a controller to balance the pendulum around its upright rest position.

According to , we choose the sampling period as Tf=αfε, where αf=0.1, ε=0.05. Then the fast sampling discrete-time model of system (66) is in the form of (1) with (67)A11=00.10.98-0.0048,A12=00.0952,A22=0.9048,B1=00.0048,B2=0.0952,E1=00.0048,E2=0.0952,C1=11,C2=1.

Solving the LMIs of Theorem 6 with ε0=0.05, α=0.2, and η=1, we have (68)P11=23.4260-52.6506-52.6506157.2814,P12=-92.204068.0912,P22=92.9218,Z1=-152.3819187.0950,Z2=-491.8613,Y1=0.1739-0.1772,Y2=-0.1620.

Choosing ε=0.05(0,ε0], then we have (69)Pε=1.1713-2.6325-4.6102-2.63257.86413.4046-4.61023.404692.9218,Zε=-7.61919.3548-491.8613,Yε=0.0087-0.0089-0.1620.

Then we can calculate the state feedback controller gain: (70)Kε=-147.2017-43.3206-11.0093.

Solving the optimization problem (48) with ε0=0.05, then we get (71)P11=0.0097-0.0277-0.02770.1043,P12=-0.0122-0.0551,P22=0.0710,Z1=-0.07480.0910,Z2=-0.0492,Y1=-0.0442-0.0299,Y2=0.0133.

Choosing ε=0.05(0,ε0], we have (72)Pε=0.0005-0.0014-0.0006-0.00140.0052-0.0028-0.0006-0.00280.0710,Zε=-0.00370.0045-0.0492,Yε=-0.0022-0.00150.0133,Kε=-37.7656-9.9051-1.4045,and μ=0.0641, which means the capacity of disturbance tolerance of the system is α=15.5997.

To improve the ability of disturbance rejection, solving the optimization problem (64) with ε0=0.05, α=15, we have (73)P11=0.0045-0.0133-0.01330.0522,P12=-0.03380.0513,P22=0.0888,Z1=-0.04790.0363,Z2=-0.6921,Y1=-0.0238-0.0405,Y2=-0.0004.

Choosing ε=0.05(0,ε0], we have (74)Pε=0.0002-0.0007-0.0017-0.00070.00260.0026-0.00170.00260.0888,Zε=-0.00240.0018-0.6921,Yε=-0.0012-0.0020-0.0004,Kε=-234.9799-48.4215-10.8701,and γ=1.0369. Thus the L2 gain from w to z of the closed-loop system with x(0)=0 is less than 1.0369.

Define a piecewise function as follows: (75)vk=0.3if  1k1000if  k>100.

Given α=15<15.5997 and ε0=0.05, it is easy to see that the disturbance w(k)=v(k)Wα2. As shown in Figure 1, under controller (4), with (76)Kε=-234.9799-48.4215-10.8701the trajectory of the closed-loop system with the disturbance ω(k)=v(k) starting from the origin is bounded and converges to the origin when the disturbance disappears.

The ellipsoid Ω(P-1(ε),α) and the trajectory starting from the origin.

The projection of trajectory to x1x2

The projection of trajectory to x1x3

The projection of trajectory to x2x3

The trajectory starting from the origin

5. Conclusion

This paper investigated the problem of fast sampling control for singularly perturbed systems subject to actuator saturation and L2 disturbance. A state feedback controller method was proposed such that all the trajectories of the closed-loop system starting from a bounded set will remain bounded for any singular perturbation parameter less than or equal to a predefined upper bound. Convex optimization problems were formulated to optimize the ability of disturbance tolerance and disturbance rejection, respectively. The presented example has illustrated the significance and validity of the proposed approaches.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (61374043, 61303183), the Jiangsu Provincial Natural Science Foundation of China (BK20130204, BK20130205), the China Postdoctoral Science Foundation Funded Project (2013M530278, 2014T70558), and the Fundamental Research Funds for the Central Universities (2013QNA50, 2013RC10, and 2013RC12).

Kokotovic P. V. Khalil H. K. O'Reilly J. Singular Perturbation Methods in Control: Analysis and Design 1986 New York, NY, USA Academic Ramírez-Betancour R. Fuerte-Esquivel C. R. Van Cutsem T. A two-time scale simulation for dynamic analysis of power systems Electric Power Systems Research 2012 83 1 185 195 10.1016/j.epsr.2011.11.007 2-s2.0-84355161936 Heidarinejad M. Liu J. de la Peña D. M. Davis J. F. Christofides P. D. Multirate Lyapunov-based distributed model predictive control of nonlinear uncertain systems Journal of Process Control 2011 21 9 1231 1242 10.1016/j.jprocont.2011.07.016 2-s2.0-80052557361 Chen X. Heidarinejad M. Liu J. de la Peña D. M. Christofides P. D. Model predictive control of nonlinear singularly perturbed systems: application to a large-scale process network Journal of Process Control 2011 21 9 1296 1305 10.1016/j.jprocont.2011.07.004 2-s2.0-80052560409 Rao K. A. Naidu D. S. Singular perturbation method applied to the open-loop discrete optimial control problem Optimal Control Applications and Methods 1982 3 2 121 131 2-s2.0-0020113747 Kando H. Iwazumi T. Sub-optimal control of discrete regulation problems via time-scale decomposition International Journal of Control 1983 37 6 1323 1347 10.1080/00207178308933048 2-s2.0-0020766746 Dong J. Yang G.-H. Robust H control for standard discrete-time singularly perturbed systems IET Control Theory & Applications 2007 1 4 1141 1148 10.1049/iet-cta:20060234 2-s2.0-34250816866 Wang G. Zhang Q. Sreeram V. H control for discrete-time singularly perturbed systems with two Markov processes Journal of the Franklin Institute 2010 347 5 836 847 10.1016/j.jfranklin.2010.03.007 2-s2.0-77955229334 Blankenship G. Singularly perturbed defference equations in optimal control problems IEEE Transactions on Automatic Control 1981 26 4 911 917 2-s2.0-0019598160 Litkouhi B. Khalil H. Infinite-time regulations for singularly perturbed defference equations International Journal of Control 1984 39 3 587 598 10.1080/00207178408933190 Dong J. Yang G.-H. H control for fast sampling discrete-time singularly perturbed systems Automatica 2008 44 5 1385 1393 10.1016/j.automatica.2007.10.010 2-s2.0-41949111849 Xu S. Feng G. New results on H control of discrete singularly perturbed systems Automatica 2009 45 10 2339 2343 10.1016/j.automatica.2009.06.011 2-s2.0-70049097547 Hu T. Lin Z. Control Systems with Actuator Saturation: Analysis and Design 2001 Boston, Mass, USA Birkhäuser Zuo Z. Ho D. W. C. Wang Y. Fault tolerant control for singular systems with actuator saturation and nonlinear perturbation Automatica 2010 46 3 569 576 10.1016/j.automatica.2010.01.024 2-s2.0-77549087681 Wang Q. Zhou B. Wen C. Duan G.-R. Output feedback gain scheduled control of actuator saturated linear systems with applications to the spacecraft rendezvous Journal of the Franklin Institute 2014 351 11 5015 5033 10.1016/j.jfranklin.2014.08.009 MR3267035 2-s2.0-84908065196 Zhou B. Wang Q. Lin Z. Duan G.-R. Gain scheduled control of linear systems subject to actuator saturation with application to spacecraft rendezvous IEEE Transactions on Control Systems Technology 2014 22 5 2031 2038 10.1109/tcst.2013.2296044 2-s2.0-84905264807 Zhou B. Gao H. Lin Z. Duan G.-R. Stabilization of linear systems with distributed input delay and input saturation Automatica 2012 48 5 712 714 10.1016/j.automatica.2012.02.007 ZBL06072646 2-s2.0-84859777128 Hu T. Lin Z. Chen B. M. An analysis and design method for linear systems subject to actuator saturation and disturbance Automatica 2002 38 2 351 359 10.1016/S0005-1098(01)00209-6 ZBL0991.93044 2-s2.0-0036466753 Cao Y.-Y. Lin Z. Ward D. G. An antiwindup approach to enlarging domain of attraction for linear systems subject to actuator saturation IEEE Transactions on Automatic Control 2002 47 1 140 145 10.1109/9.981734 2-s2.0-0036256863 Liu P.-L. Stabilization of singularly perturbed multiple-time-delay systems with a saturating actuator International Journal of Systems Science 2001 32 8 1041 1045 10.1080/00207720010015672 ZBL1023.93051 2-s2.0-0035437153 Garcia G. Tarbouriech S. Control of singularly perturbed systems by bounded control 5 Proceedings of the American Control Conference June 2003 Denver, Colo, USA IEEE 4482 4487 10.1109/ACC.2003.1240547 2-s2.0-0142138290 Xin H.-H. Wu D. Gan D.-Q. Qiu J.-J. A method for estimating the stability region of singular perturbation systems with saturation nonlinearities Acta Automatica Sinica 2008 34 12 1549 1555 10.3724/sp.j.1004.2008.01549 2-s2.0-58149229769 Xin H. Gan D. Huang M. Wang K. Estimating the stability region of singular perturbation power systems with saturation nonlinearities: an linear matrix inequality-based method IET Control Theory & Applications 2010 4 3 351 361 10.1049/iet-cta.2008.0331 2-s2.0-77749273805 Lizarraga I. Tarbouriech S. Garcia G. Control of singularly perturbed systems under actuator saturation Proceedings of the 16th Triennial World Congress of International Federation of Automatic Control (IFAC '05) July 2005 Prague, Czech Republic 243 248 2-s2.0-79960732271 Feng W. Characterization and computation for the bound ε in linear time-invariant singularly perturbed systems Systems & Control Letters 1988 11 3 195 202 Sen S. Datta K. B. Stability bounds of singularity perturbed systems IEEE Transactions on Automatic Control 1993 38 2 302 304 10.1109/9.250478 2-s2.0-0027544731 Cao L. Schwartz H. M. Complementary results on the stability bounds of singularly perturbed systems IEEE Transactions on Automatic Control 2004 49 11 2017 2021 10.1109/TAC.2004.837546 2-s2.0-9744234480 Li T.-H. S. Li J.-H. Stabilization bound of discrete two-time-scale systems Systems & Control Letters 1992 18 6 479 489 10.1016/0167-6911(92)90052-t 2-s2.0-0026883896 Liu W. Q. Paskota M. Sreeram V. Teo K. L. Improvement on stability bounds for singularly perturbed systems via state feedback International Journal of Systems Science 1997 28 6 571 578 10.1080/00207729708929418 ZBL0891.34055 2-s2.0-0031166340 Chiou J.-S. Kung F.-C. Li T.-H. S. An infinite ε-bound stabilization design for a class of singularly perturbed systems IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications 1999 46 12 1507 1510 10.1109/81.809557 2-s2.0-0033450590 Yang G.-H. Dong J. X. Control synthesis of singularly perturbed fuzzy systems IEEE Transactions on Fuzzy Systems 2008 16 3 615 629 10.1109/tfuzz.2007.905911 2-s2.0-45549087781 Yang C. Zhang Q. Multiobjective control for T-S fuzzy singularly perturbed systems IEEE Transactions on Fuzzy Systems 2009 17 1 104 115 10.1109/tfuzz.2008.2005404 2-s2.0-60549101932 Yang C. Y. Sun J. Ma X. P. Stabilization bound of singularly perturbed systems subject to actuator saturation Automatica 2013 49 2 457 462 10.1016/j.automatica.2012.11.004 MR3004711 2-s2.0-84872051498 Hindi H. Boyd S. Analysis of linear systems with saturating using convex optimization Preceedings of the 37th IEEE Conference on Decision and Control 1998 Tampa, Fla, USA 903 908 Zak S. H. Maccarley C. A. State-feedback control of non-linear systems International Journal of Control 1986 43 5 1497 1514 10.1080/00207178608933554 2-s2.0-0022715120 Kando H. Iwazumi T. Multirate digital control design of an optimal regulartor via singular perturbation theory International Journal of Control 1986 44 6 1555 1578