Double-layer model predictive control integrated with zone control.

Area of Science:

• Process Control
• Chemical Engineering
• Automation Systems

Background:

• Model predictive control (MPC) is widely used for complex industrial processes.
• Existing MPC strategies may face challenges in achieving optimal economic performance and robustness.
• Integration of zone control with MPC offers potential for improved process management.

Purpose of the Study:

• To propose a novel double-layer model predictive control (DLMPC) strategy integrated with zone control.
• To enhance steady-state target calculation (SSTC) for optimal process operation.
• To improve control smoothness, robustness, and economic performance compared to existing methods.

Main Methods:

• Developed a two-layer control architecture: steady-state target calculation (SSTC) and dynamic control.
• Pretreated controlled output variables using process principles and production data in the SSTC layer.
• Incorporated weighted and priority soft constraints relaxation for infeasible SSTC problems.
• Transmitted optimal targets and zone parameters from SSTC to the dynamic control layer for tracking.

Main Results:

• The proposed DLMPC strategy achieved smooth control.
• Demonstrated superior robustness and economic performance compared to conventional DLMPC and zone model predictive control (ZMPC).
• A simulation example validated the effectiveness of the proposed method.

Conclusions:

• The integrated DLMPC and zone control strategy provides significant advantages over traditional approaches.
• The method effectively optimizes steady-state targets and handles constraint relaxation.
• The approach offers a promising solution for enhancing industrial process control efficiency and performance.