Non-fragile H2 reliable control for switched linear systems with actuator faults
Highlights
► A new method of describing the continuous fault matrix is adopted. ► A new method settling non-fragile reliable control is proposed. ► H2 control and pole assignment are combined together. ► The state-dependent switching law is designed for H2 control.
Introduction
Switched systems, a vital class of hybrid systems, have received more and more attentions during the past decade [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22]. The basic developments of switched systems are located in three basic problems [1]: asymptotical stability of the switched system with arbitrary switching, stability for certain useful classes of switching sequences, and construction of asymptotically stabilizing switching signals for a switched system. Almost all the control problems of switched systems are developed on the basis of the basic problems. control problem of switched systems is solved in [2]. Based on the multiple storage function approach, the passivity theory and stability theory of switched systems are provided for the first time in [3]. The work in [4] can be extended to switched systems. filtering [5], [6], [7], stabilization [8], [9], [10], fault-tolerant control [11], [12] and fault recoverability analysis [13] for switched systems have received good developments. By means of linear matrix inequality (LMI) approach and average dwell time technique, the fault detection problem is solved [14] for a discrete-time switched linear system with time-varying delays. The robust fault detection problem is settled [15] for a class of uncertain switched linear time-delay systems with arbitrary switching signal. The delay-depend exponential control problems for continuous- and discrete-time switched systems are settled in [16] and [17], respectively. Based on reduced-order observer, the delay-deoendent control is concerned in [18]. For a class of discrete-time nonlinear switched systems, the exponential stability and stabilization problems are solved in [19], by using LMI approach and Lyapunov functional technique. Based on switching generalized gramians, the switching Petrov–Galerkin projection is determined for the model reduction of switched systems [20]. By using observer technique, the state estimation and sliding mode control (SMC) problems are settled for a class of uncertain switched systems [21]. As another class of hybird systems, Markovian system is also concerned in the SMC area, recently [22], [23], [24]. For a class of discrete-time linear networked control systems with Markovian jump, the stabilization problem is investigated by using interactive LMI approach [22]. By using the proportional and derivative sliding mode observer technique, the fault-tolerant control problem for a class of nonlinear Markovian jump systems with sensor fault is concerned in [23]. For a class of Markovian jump singular time-delay systems, the problems of L2 control and SMC are solved in [24] by using LMI approach.
Only a few results are devoted to fault tolerant control of switched linear systems. The finite-time stability of switched systems with actuator discrete faults is concerned in [25], by using average dwell-time approach. For mixed-fault model, the integrity of the switched systems with actuator faults is analysed in [26]. The issue of fault estimator design for switched systems is discussed in [27], by using LMI approach.
H2 control is an important problem in the control field. For non-switched systems, H2 control has been developed well [28]. But it is rarely reported for switched systems [29], [30], so is non-fragile control. Non-fragile control is to design a controller which is not sensitive to the disturbance in the feedback. There are many research results about non-fragile control, e.g. [31], [32], [33] and the references therein.
Now we focus on pole assignment. Pole assignment is closely related to . If all the poles of the system matrix lie in a specified region , the system is so-called [34]. The point is that the region must be a subregion of the asymptotically stable region of the system. The universal pole assignment is proposed in [34] for the first time, and it also promotes the development of pole assignment greatly.
Based on the statement above, we can see that there is still much room for mobility development in developing H2 control for switched systems. In many practical systems, such as power systems and flight control systems, an important measure of peak values that produced by finite energy content signals is H2 norm, so that H2 control is widely concerned. However, the non-fragile H2 control problem of switched linear systems with actuator faults has not been solved. For a class of switched linear systems with actuator faults, the non-fragile H2 reliable control is concerned in this paper.
This paper is organized as follows. Section 2 formulates the main purpose of this paper and prepares some useful theories. Section 3 proposes a new approach of settling a class of matrix multiply phenomenon, and then provides an approach to design the switching signal and the non-fragile reliable controller which can guarantee that the closed-loop system is of a specified H2 performance and all the closed-loop poles lie in a specified circular disk. Finally, the effectiveness and practicality of this method are shown by a practical example in Section 4.
Section snippets
Problem statement and preliminaries
Consider the following switched linear system with actuator faults:where is the state; is the control input vector of actuator fault for the i-th mode of the system; is the exogenous disturbance input; is the regulated output. is a switching rule to be designed. Furthermore, implies that the i-th subsystem is activated, and , , , and
Main results
Lemma 4 Let B, E, X and Y be matrices of appropriate dimensions, and H be an uncertain matrix such that , and be a diagonal uncertain matrix satisfying . Then, there exist a positive definite diagonal matrix U and a positive scalar such thatand Proof It follows from Lemma 1 that there exists a positive scalar such that Obviously, there exists a positive definite diagonal matrix U satisfies the
Example
Consider the linear switched system of the system (1) with N=3 in the fault-free case as follows [36]:
The parameters of perturbation of the controller (2) are given by
Conclusions
Non-fragile H2 reliable control with circular disk pole constraints problem has been solved for switched systems with actuator faults based on state feedback. In terms of LMI, we have designed the switched state feedback controller which can guarantee the H2 performance is smaller than a specified scalar for all uncertainties in the state feedback and admissible actuator faults. The effectiveness of the proposed result has been shown in the given example.
Our future work is to extend the method
Acknowledgement
The work is supported by the National Natural Science Foundations of China under Grant Nos. 61034005, 61104116 and 61273171, the Doctoral Fund of Ministry of Education of China under Grant No. 20113218110011, and A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions.
References (37)
- et al.
On stability, and control for switched systems
Automatica
(2008) - et al.
Passivity and stability of switched systemsa multiple storage function method
Systems & Control Letters
(2008) - et al.
Robust filtering for jumping systems with mode-dependent delays
Signal Processing
(2006) Delay-dependent filtering of a class of switched discrete-time state delay systems
Signal Processing
(2008)- et al.
Asynchronous filtering of discrete-time switched systems
Signal Processing
(2012) - et al.
Stability of switched positive linear systems with average dwell time switching
Automatica
(2012) - et al.
Delay-dependent fault detection for switched linear systems with time-varying delays—the average dwell time approach
Signal Processing
(2011) - et al.
Weighted H-infinity model reduction for linear switched systems with time-varying delay
Automatica
(2009) Stabilization of Markovian jump linear system over networks with random communication delay
Automatica
(2009)- et al.
Sliding mode control with bounded gain performance of Markovian jump singular time-delay systems
Automatica
(2012)