Dual Event-Triggered Spatial Model Predictive Control for Distributed Thermal Processes.

Area of Science:

• Control Engineering
• Data-Driven Systems
• Thermal Process Management

Background:

• Distributed thermal processes often suffer from performance degradation due to excessive model updates and controller activations.
• Over-computation in these systems leads to inefficiencies and suboptimal control outcomes.

Purpose of the Study:

• To develop a novel dual event-triggered spatial model predictive control (DET-SMPC) strategy for distributed thermal processes.
• To mitigate performance issues caused by frequent model updates and controller activations.
• To enhance overall system performance through a data-driven framework.

Main Methods:

• Constructed a spatiotemporal model using time/space theorems to adapt to time-varying dynamics.
• Proposed an adaptive model update (MU) approach using an error-triggered generator to determine optimal switching modes.
• Introduced an event-triggered model predictive control (ET-MPC) with a Lyapunov function-derived activation threshold.

Main Results:

• The proposed DET-SMPC effectively reduces over-computation by triggering controller activations only when a predefined threshold is met.
• Adaptive MU approach identified preferable switching modes, optimizing controller performance.
• Simulations and experimental oven studies validated the efficacy of the DET-SMPC.

Conclusions:

• DET-SMPC offers a robust solution for improving performance in distributed thermal processes.
• The event-triggered mechanism significantly reduces computational load while maintaining effective control.
• This data-driven approach demonstrates practical applicability in real-world thermal management systems.