Event-triggered Control for Networked Control Systems and Its Application

基於事件觸發機制的網絡化控制研究及其應用

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
  • Min XIE (Supervisor)
  • Junlin Xiong (External person) (External Supervisor)
Award date30 Jul 2021

Abstract

Theoretical research on networked control systems (NCSs) has attracted great attention. In addition, the research on NCSs based on event-triggered schemes (ETSs) was hot in the last decade. It has been proven that ETSs can reduce the number of transmitted packets while maintaining control performance of NCSs. However, there are some limitations in the existing results: 1. There is still room for improvement in the event-triggered conditions, and most of the trigger parameters are static. How to further reduce transmission rate and how to design an adaptive ETS are worth studying. 2. Most of the results focus on the ideal network case, (that is, the case without cyber attacks). It is of practical significance to investigate the security control problem with malicious attacks; 3. Most of the results are only applicable to wireless networks, and the issue of event-triggered control for NCSs over cognitive radio (CR) networks has not been investigated. It is more practical to consider the event-triggered control over CR networks.

To overcome the limitations of the existing results, it is necessary to improve the existing ETSs and consider the security control problem. To solve the limitations, the main results of this thesis are as follows:

1. For a class of continuous time linear time-invariant NCSs, the stability problem is investigated based on a new ETS. By using both the relative error and the absolute error information, a new ETS is proposed to reduce transmission rate. Next, sufficient conditions are derived for the stability of NCSs and the controller design by using the common Lyapunov functional method. The conditions are described by a set of linear matrix inequalities (LMIs). Furthermore, an optimization algorithm is proposed to design the optimal trigger parameters and controller gain simultaneously. Finally, an inverted pendulum is performed to verify the effectiveness of the proposed method.

2. For a class of NCSs with disturbances, the stability problem and L∞ performance problem are investigated based on an adaptive ETS. To guarantee the system stability, it is only required that the Lyapunov function tends to zero when the system operation time approaches to infinite. The derivative of Lyapunov function can be nonnegative in some time intervals. Motivated by above, the dynamic trigger parameter is designed. The adaptive trigger parameter is adjusted in real time according to the designed algorithm. Moreover, the dynamic and the static trigger parameter are set as the upper and the lower bound of the adaptive trigger parameter, respectively. Sufficient conditions are derived for the stability and L∞ performance by using the Lyapunov functional method. Finally, a two-area power system is used to show that the proposed adaptive ETS can balance the relationship between system performance and transmission rate.

3. For a class of NCSs with time-varying delays and deception attacks, the stability problem is investigated based on an improved ETS. The closed-loop system based on the improved ETS is modelled as a switched system. Sufficient conditions are derived for the mean square stability of NCSs. Moreover, a co-design method is given to obtain both controller gain and trigger parameters.

4. For a class of NCSs with deception attacks, the stability problem and L∞ performance problem are investigated based on a dynamic switching ETS. The closed-loop system based on dynamic switching ETS is modelled as a switched system. By using the common Lyapunov functional method, sufficient conditions are derived to guarantee the mean square stability and L∞ performance. Finally, the effectiveness of the proposed method is verified by an unmanned aerial vehicle (UAV) system.

5. For a class of NCSs with time-varying delays, the stability problem and H∞ performance problem are investigated based on an ETS over CR networks. An ETS is proposed in the presence of CR networks. The CR channel is modelled as an ON-OFF switch with sojourn time, and the sojourn time is arbitrary but bounded. The event-triggered system is modelled as a switched delay system. Then, the common Lyapunov functional method and the time-delay system method are used. Based on the above methods, the system stability is investigated, and the H∞ performance controller is designed. Finally, the effectiveness of the proposed method is verified by an UAV system.

    Research areas

  • Cognitive radio networks, deception attacks, event-triggered control, Lyapunov functional, networked control systems, time-varying delays