Asynchronous fault detection filtering for nonhomogeneous Markov jump systems with dynamic quantization subject to a novel hybrid cyber attacks

• Published in: ISA transactions Vol. 154; pp. 73 - 81
• Main Authors: Hua, Mingang, Sun, Ni, Deng, Feiqi, Fei, Juntao, Chen, Hua
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.11.2024
• Subjects: Asynchronous fault detection filter; Dynamic quantization; Hidden Markov model; Hybrid cyber attacks; Markov jump systems; Hidden Markov model; Hybrid cyber attacks; Markov jump systems; Asynchronous fault detection filter; Dynamic quantization
• ISSN: 0019-0578; 1879-2022; 1879-2022
• DOI: 10.1016/j.isatra.2024.09.004

The problem of asynchronous fault detection filtering for nonhomogeneous Markov jumping systems with dynamic quantization and hybrid cyber attacks is addressed in this paper. The introduction of polytopic-structure-based transition probabilities is employed to describe the nonhomogeneous Markov process. An asynchronous fault detection filter is proposed, which utilizes the hidden Markov model to achieve comprehensive access to the plant mode information. Prior to transmission to the filter, the measurement output of the system undergoes quantization using a dynamic quantizer. The novel hybrid cyber attacks model being discussed involves four types of attacks: deception attacks, denial-of-service attacks, no attack, and hybrid attacks with both deception and denial-of-service attacks. By constructing Lyapunov functional, sufficient conditions are presented for achieving the stochastic stability with H∞ performance. Under the complex network environment, the industrial application of the presented asynchronous fault detection filtering model is demonstrated on a non-isothermal continuous stirred tank reactor. The simulation results confirm the practicality of the proposed design method. •The dynamic quantizer is adopted to quantify measurement output signals to reduce information exchange frequency and communication burden.•A convex polyhedron is used to describe the time-varying transition probability matrix.•Due to the mode information between the filter and the plant cannot be fully obtained in the networked control system, a HMM-based asynchronous mechanism is proposed.•In order to improve the system performance, a novel hybrid network attack model is proposed, which includes four attack modes.