Periodic Event-Triggered Model Predictive Control for Networked Nonlinear Uncertain Systems With Disturbances.

Area of Science:

• Control Systems Engineering
• Networked Systems
• Nonlinear Dynamics

Background:

• Networked nonlinear uncertain systems face challenges with time-varying disturbances.
• Traditional Model Predictive Control (MPC) can be computationally intensive and communication-heavy.
• Event-triggered control strategies aim to optimize resource usage in control systems.

Purpose of the Study:

• To develop a Periodic Event-Triggered Model Predictive Control (PETMPC) for networked nonlinear uncertain systems.
• To reduce computational load and signal transmission frequency compared to traditional MPC.
• To ensure global bounded stability of the closed-loop system.

Main Methods:

• A Generalized Proportional-Integral Observer (GPIO) is designed to estimate system states and disturbances from sampled data.
• Disturbance predictions are generated using the forward Euler method.
• An optimal control sequence is computed and stored, executed during inter-event intervals, reducing computation and transmission.

Main Results:

• The proposed PETMPC method effectively reduces the number of control sequence computations and signal transmissions.
• Stability analysis confirms that the closed-loop hybrid control system is globally bounded stable under the PETMPC law.
• Numerical simulations demonstrate the feasibility and superior performance of the PETMPC method.

Conclusions:

• The PETMPC approach offers an efficient control strategy for networked nonlinear uncertain systems.
• The event-triggering mechanism significantly optimizes resource utilization without compromising system stability.
• This method presents a viable alternative to traditional MPC for applications with communication and computation constraints.