Learning of Iterative Learning Control for Flexible Manufacturing of Batch Processes

Area of Science:

• Manufacturing Engineering
• Control Systems
• Artificial Intelligence

Background:

• Flexible manufacturing in smart manufacturing requires rapid controller adaptation for customized production.
• Batch processes present significant challenges for precise control and error reduction.

Purpose of the Study:

• To develop a novel control strategy for enhancing precision in flexible manufacturing.
• To address the need for fast controller adaptation in customized and batch production environments.

Main Methods:

• Implementation of a learning of iterative learning control (ILC) method utilizing neural networks.
• Recommending optimal control parameters for ILC controllers to minimize tracking errors.

Main Results:

• The proposed neural network-based ILC method demonstrated faster tracking error convergence compared to benchmark ILC.
• Achieved smaller steady-state errors for diverse set-point profiles, abstracting varied production demands.
• Outperformed a standard ILC approach across various systems and scenarios.

Conclusions:

• The learning of ILC based on neural networks offers a robust solution for adaptable and precise control in flexible manufacturing.
• This method shows significant potential for integration into the industrial Internet of Things (IIoT) for advanced manufacturing applications.