Two-dimensional iterative learning fault-tolerant control for batch processes via a high-order fully actuated...

Area of Science:

• Process Control
• Fault-Tolerant Systems
• Chemical Engineering

Background:

• Increasing equipment complexity leads to unavoidable faults in batch processes.
• Conventional fault-tolerant control methods often fail to meet stringent tracking accuracy requirements.
• Existing observer-controller co-design strategies can suffer from gain coupling issues.

Purpose of the Study:

• To develop a novel two-dimensional iterative learning fault-tolerant control (2D-ILFTC) strategy for batch processes.
• To address challenges posed by disturbances and actuator faults.
• To improve tracking accuracy and reduce computational burden.

Main Methods:

• Implementation of a 2D-ILFTC strategy within a high-order fully actuated framework.
• Structural separation of observer and controller design to mitigate gain coupling.
• Incorporation of historical control inputs to simplify the closed-loop system dynamics.

Main Results:

• The developed 2D-ILFTC method significantly reduced root mean square error (0.6801) and integral absolute error (0.6207) in batch stirred reactor simulations.
• The average dynamic tracking indicator also showed substantial improvement (391.05) compared to methods without compensation.
• The strategy demonstrated superior tracking performance, particularly by the 10th cycle with compensation.

Conclusions:

• The proposed 2D-ILFTC strategy effectively enhances tracking accuracy in batch processes with disturbances and actuator faults.
• Structural separation of observer and controller design proves advantageous over co-design approaches.
• The method offers a computationally efficient and high-performance solution for complex batch process control.