Fast actuator fault-tolerant control for a class of nonlinear sampled-data systems via deterministic learning

Area of Science:

• Control Systems Engineering
• Nonlinear Dynamics
• Fault Diagnosis and Tolerant Control

Background:

• Nonlinear sampled-data systems are susceptible to actuator faults, impacting performance and safety.
• Existing fault-tolerant control (FTC) methods may struggle with the rapid evolution of faults from incipient to larger magnitudes.

Purpose of the Study:

• To develop a fast fault-tolerant control strategy for nonlinear sampled-data systems experiencing staged actuator faults.
• To enhance fault detection and isolation (FDI) sensitivity and response speed for incipient and significant actuator failures.

Main Methods:

• A deterministic learning approach (DLA) was employed to construct learning controllers and identifiers.
• Exponential stability of linear time-varying (LTV) discrete-time systems was utilized to acquire control and fault diagnosis knowledge.
• Pattern-based FTC schemes and diagnosis estimators were implemented for fault response.

Main Results:

• The proposed method achieved simultaneous exponential convergence of tracking and parameter estimation errors.
• Enhanced sensitivity to incipient actuator faults and fast FTC response to larger faults were demonstrated.
• Simulation results validated the effectiveness of the developed FDI and FTC schemes.

Conclusions:

• The DLA-based FTC strategy effectively handles staged actuator faults in nonlinear sampled-data systems.
• The approach provides rapid fault detection, isolation, and control adaptation, improving system resilience.
• The method offers a robust solution for enhancing control performance and safety under fault conditions.