Dynamic Event-Triggered Fault Detection for Markov Jump Systems Under DoS Attacks: A Simulated Annealing...

Area of Science:

• Control Systems Engineering
• Cyber-Physical Systems Security
• Stochastic Systems

Background:

• Markov jump systems are susceptible to denial-of-service (DoS) attacks, compromising their operational integrity.
• Traditional fault detection observers (FDOs) may be communication-intensive and vulnerable to intermittent attacks.
• Limited energy of attackers characterizes nonperiodic DoS attacks, necessitating adaptive defense mechanisms.

Purpose of the Study:

• To design a robust fault detection observer (FDO) for Markov jump systems.
• To mitigate the impact of denial-of-service (DoS) attacks using a dynamic event-triggered mechanism.
• To ensure the FDO exhibits both disturbance robustness and fault sensitivity.

Main Methods:

• A dynamic event-triggered mechanism was developed to optimize communication resource usage.
• The $H_{\infty }/H_{-}$ index was integrated to balance robustness and sensitivity.
• Lyapunov functional techniques were used to derive nonlinear inequalities.
• A simulated annealing algorithm was employed for solving and parameter optimization.

Main Results:

• The proposed FDO effectively detects faults in Markov jump systems under DoS attacks.
• The dynamic event-triggered mechanism significantly reduces communication load.
• The $H_{\infty }/H_{-}$ index ensures desired observer performance.
• The simulated annealing algorithm successfully found optimal parameters.

Conclusions:

• The designed fault detection observer is effective for Markov jump systems facing DoS attacks.
• The dynamic event-triggered mechanism offers a resource-efficient approach to fault detection.
• The methodology provides a robust solution for securing critical systems.