A novel zonotopic Kalman filter-based actuator fault detection for time delay systems

Area of Science:

• Control Systems Engineering
• Fault Diagnosis
• State Estimation

Background:

• Linear discrete time delay systems are susceptible to actuator faults, complicating state estimation.
• Accurate fault diagnosis is crucial for system reliability and safety.

Purpose of the Study:

• To propose a novel algorithm for state estimation and actuator fault diagnosis in linear discrete time delay systems.
• To develop a zonotopic Kalman filter-based actuator fault detection (Z-KF-AFD) algorithm.

Main Methods:

• Approximating noise to establish fault-state relationships.
• Iteratively deriving a zonotopic Kalman filter (ZKF) to link current and delayed data.
• Designing optimal observer estimator gain by minimizing zonotopic set size.
• Separating fault zonotopes to link current and delayed states.

Main Results:

• The Z-KF-AFD algorithm effectively estimates system states and detects actuator faults.
• Fault detection is achieved by checking if zero is within the estimated fault zonotope bounds.
• The algorithm's feasibility was validated using numerical and bidirectional DC-DC converter systems.

Conclusions:

• The proposed Z-KF-AFD algorithm offers a robust solution for state estimation and actuator fault diagnosis in systems with time delays.
• The method demonstrates practical applicability in complex engineering systems.