Event-based Finite-time Boundedness of Discrete-time Network Systems
- PDF / 1,406,805 Bytes
- 10 Pages / 594.77 x 793.026 pts Page_size
- 69 Downloads / 189 Views
ISSN:1598-6446 eISSN:2005-4092 http://www.springer.com/12555
Event-based Finite-time Boundedness of Discrete-time Network Systems Yingqi Zhang*, Miaojun Zhan, Yan Shi, and Caixia Liu Abstract: This paper deals with the event-based finite-time H∞ control problem of discrete-time network control systems with norm bounded input disturbances and nonlinear stochastic functions. A network-induced delay stochastic model is first constructed by event-triggered approach. Utilizing stochastic analysis and event-triggered schemes, conditions on stochastic finite-time (FT) boundedness and stochastic H∞ FT boundedness are then derived for the network model. Subsequently, an event-based finite-time controller and an event-triggered matrix are co-designed to ensure that the stochastic model is stochastically FT bounded or stochastically H∞ FT bounded by utilizing matrix decomposition scheme. All derived criteria can be solved in terms of convex optimal method, and numerical examples demonstrate the validity of obtained results as well. Keywords: Event-based control, network control systems, network-induced delay, stochasticaly finite-time boundedness.
1.
INTRODUCTION
Network control systems have extensively been used due to their amounts of virtues such as low costs of installation and maintenance, system flexibility and high reliability. Therefore, much work has been investigated for the system performance and stability analysis in [1–3]. In order to overcome the challenges of limited bandwidth (data dropout, data disordering and network-induced delays), nowadays, there are two kinds of strategy to tackle the challenges: quantization scheme and event-triggered technique. To reduce the data size, the quantization strategy is adopted. For instance, the stabilization issues of linear time-invariant systems with saturating quantized measurements are investigated by proposing a new control design method in terms of changing the sensitivity of the quantizer in [4,5]. To decrease the data transmission frequency of the network signals, the event-triggered control method is used, in which the control task is executed according to a predefined event constraint other than a fixed time interval. Thus, many appealing results have been found, such as state feedback control in [6, 7], optimal state estimation in [8], output feedback control in [9], H∞ control in [10, 11], event-triggered fault detection design in [12, 13], sliding mode control in [14], dissipative filter design in
[15, 16], and so forth. On another frontier, the transient performance of the systems in finite-time (FT) intervals can be more important than classical Lyapunov stability during an infinite time. To tackle the system transient behavior, FT stability was introduced in [17, 18], and the concept of FT boundedness was also proposed in [19]. Then, the results related to FT control were extended to optimal FT guaranteed cost control [20, 21], H∞ FT control [22, 23], finite-time positiveness and distributed control [24, 25], FT reliable control and state estimation [2
Data Loading...
