Fixed-Time Control for a Flexible Smart Structure With Actuator Failure: A Broad Learning System Approach.

Area of Science:

• Structural Engineering
• Control Systems Engineering
• Artificial Intelligence in Engineering

Background:

• Tall building structures are susceptible to vibrations from external forces and internal dynamics.
• Uncertainties in structural models and actuator failures pose significant challenges to effective vibration control.
• Existing control methods may not guarantee performance within a fixed-time or handle unknown actuator degradation.

Purpose of the Study:

• To develop an adaptive fault-tolerant control (AFTC) strategy for robust vibration suppression in uncertain stand-alone tall building-like structures (STABLS).
• To ensure fixed-time stability and performance guarantees against model uncertainties and actuator effectiveness failures.
• To estimate the lower bound of actuator health, even when it is initially unknown.

Main Methods:

• Implementation of an adaptive fixed-time sliding mode control approach.
• Integration of adaptive improved radial basis function neural networks (RBFNNs) within a broad learning system (BLS) for uncertainty estimation.
• Development of a fault-tolerant mechanism to address actuator effectiveness failures.

Main Results:

• Theoretical and practical demonstration of guaranteed fixed-time performance for the STABLS under uncertainty and actuator failures.
• Successful estimation of the lower bound of actuator health.
• Validation of the proposed vibration suppression method through simulations and experimental results.

Conclusions:

• The proposed AFTC method effectively suppresses vibrations in uncertain STABLS, ensuring fixed-time stability.
• The approach robustly handles actuator effectiveness failures and estimates unknown actuator health.
• The findings confirm the practical applicability and efficacy of the developed control strategy for structural vibration mitigation.