Event-triggered fault detection filtering for discrete-time Markovian jump systems

2018 ◽  
Vol 152 ◽  
pp. 384-391 ◽  
Author(s):  
Bingna Qiao ◽  
Xiaojie Su ◽  
Renfeng Jia ◽  
Yan Shi ◽  
Magdi S. Mahmoud



2019 ◽  
Vol 356 (5) ◽  
pp. 3045-3071 ◽  
Author(s):  
Guoliang Wang ◽  
Zhiqiang Li ◽  
Xin Miao ◽  
Qingling Zhang ◽  
Chunyu Yang


Author(s):  
Dunke Lu ◽  
Xinyu Zhang ◽  
Yunsheng Lan ◽  
Xiaohua Wang

This paper addresses the fault detection and isolation for a class of discrete-time Markovian jump system with generally bounded transition probabilities. The proposed method is mainly based on a bank of [Formula: see text] robust observers, which can detect the actuator fault occurrence and, even more, can locate those subjected to the fault. Furthermore, the performance and the robustness of the designed [Formula: see text] robust observer are disseminated respect to disturbances and measurement noises. In particular, we propose the zonotop-based dynamic threshold to predict the ranges of the residual, which shows less conservatism in comparison to traditional methods. Finally, pertinent simulations are performed to corroborate the potential and superiority of the proposed method. In general, the proposed method offers effective fault detection and isolation scheme for Markovian jump systems, which can detect the fault timely and accurately with lowering the rate of fault missing report to a great extent, thus having practical value.



Author(s):  
Mengmeng Liu ◽  
Jinyong Yu ◽  
Yiming Sun ◽  
Jiahao Li

In this paper, the adaptive event-triggered fault detection (FD) problem is investigated for Markovian jump systems (MJSs) with network time-delays and data package dropouts. First, a novel event-triggered communication strategy with an adaptive threshold is introduced to screen the sampled signals and reduce the data releasing frequency. Consequently, the limited communication resources and network bandwidth are saved as much as possible. Second, since communication links between the plant and filter are considered to be unreliable, the effects of intermittent package dropouts phenomenon and network time-delays are taken into account simultaneously. By using time-delay system method, the network-induced phenomena, adaptive event-triggered strategy and MJSs are unified into a networked MJSs time-delay system. Then, sufficient conditions are developed such that the resulting residual system is mean-square exponentially stable with the desired [Formula: see text] performance. Finally, a numerical example of the F-404 aircraft engine system is provided to illustrate the effectiveness and potential of the proposed design method.



Sign in / Sign up

Export Citation Format

Share Document