Dynamic Surface Intelligent Robust Control of Nonlinear Systems With Fixed-Time Sliding-Mode Observer.

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Nonlinear Dynamics

Background:

• Nonlinear systems present significant challenges in achieving high tracking control precision and fast finite-time convergence.
• Complex nonlinearities and unknown disturbances hinder robust control performance.

Purpose of the Study:

• To propose a dynamic surface intelligent robust control strategy with a fixed-time sliding-mode observer (DSIRC-SMO) for improved finite-time tracking control.
• To enhance the approximation accuracy of nonlinear systems and the anti-interference capability.

Main Methods:

• Designed a predictor-based adaptive fuzzy neural network (P-AFNN) to imitate complex nonlinearities, using prediction error for weight adaptation.
• Integrated a fixed-time sliding-mode observer (SMO) into dynamic surface control to address disturbances and modeling errors.
• Proved the fixed-time convergence of the SMO and the finite-time convergence of the overall DSIRC-SMO strategy.

Main Results:

• The P-AFNN demonstrated improved accuracy in approximating nonlinear system dynamics.
• The integrated SMO enhanced the system's anti-interference capability by timely updating dynamic surface information.
• The DSIRC-SMO strategy was proven to be effectively implementable with guaranteed finite-time convergence.

Conclusions:

• The proposed DSIRC-SMO strategy effectively improves tracking control precision and finite-time convergence for nonlinear systems.
• The combination of P-AFNN and fixed-time SMO offers a robust solution for complex nonlinear control problems.
• Validated through numerical and wastewater treatment process simulations, demonstrating practical applicability.