Decentralized supervisory based switching control for uncertain multivariable plants with variable input-output...

Area of Science:

• Control Systems Engineering
• Automation and Robotics
• Systems Theory

Background:

• Designing robust decentralized control for uncertain multivariable plants presents significant challenges.
• Existing methods may struggle with dynamic changes in plant operation, such as input-output pairing variations.

Purpose of the Study:

• To develop a stable and robust adaptive decentralized switching control strategy for uncertain multivariable plants.
• To address challenges posed by plant uncertainties and dynamic operational changes.

Main Methods:

• The strategy divides plant uncertainty regions, employing local controllers based on Quantitative Feedback Theory (QFT).
• A supervisor module makes switching decisions by comparing plant behavior to nominal models.
• Hysteresis switching logic and multirealization techniques ensure stability and bumpless transfer.

Main Results:

• The proposed method yields a stable and robust adaptive controller capable of handling complex multivariable plants.
• The strategy effectively manages input-output pairing changes, facilitating reconfigurable decentralized control.
• Simulation results demonstrate the method's effectiveness and the achievement of bumpless transfer.

Conclusions:

• The developed decentralized switching control strategy offers a robust and adaptive solution for uncertain multivariable systems.
• This approach enhances system resilience to operational changes and supports reconfigurable control architectures.
• The findings contribute to advancing the design of advanced control systems for complex industrial applications.