Practical Predefined-Time Sliding-Mode Adaptive Resilient Control for PMSM Cyber-Physical Systems

Area of Science:

• Control Systems Engineering
• Cyber-Physical Systems Security
• Robotics and Automation

Background:

• Permanent magnet synchronous motors (PMSMs) are crucial in cyber-physical systems (CPSs).
• Networked control systems face vulnerabilities to malicious cyber-attacks impacting PMSM performance.
• Existing control algorithms may have limitations in convergence time and resilience to uncertainties.

Purpose of the Study:

• To develop a novel resilient control algorithm for uncertain PMSMs under cyber threats.
• To ensure predefined-time convergence of trajectory tracking errors (TTEors) despite malicious attacks.
• To enhance the robustness and security of PMSM-based CPSs.

Main Methods:

• Introduction of a new Lyapunov stability criterion with adjustable gain reaching law and predefined-time convergence (PreTC).
• Design of a sliding-mode (SMS) controller incorporating PreTC.
• Utilizing an extreme learning machine (ELM) for real-time identification of physical layer models and cyber-attacks.
• Development of a sliding-mode adaptive resilient controller without explicit physical model information.

Main Results:

• The proposed algorithm guarantees predefined-time (PDT) stability for PMSMs in CPSs.
• Demonstrated superior resilience against cyber-attacks, parameter perturbations, and external disturbances.
• Achieved a motor speed tracking error accuracy of 0.008 in comparative simulations.
• Validated effectiveness through comparisons with four distinct resilient control algorithms.

Conclusions:

• The developed resilient control algorithm effectively ensures PMSM speed control within a specified timeframe under cyber-attacks.
• The algorithm provides enhanced robustness and security for PMSM-driven CPSs against various uncertainties.
• This work offers a significant advancement in securing networked power drive systems.