Session: 16-01-01: Government Agency Student Poster Competition

Paper Number: 150237

150237 - Advanced Registration Control in Roll-to-Roll Printing: Spatial-Terminal Iterative Learning Strategies 

Roll-to-Roll (R2R) systems are a type of industrial equipment used in the continuous manufacturing of flexible substrates. These systems typically consist of a sequence of motorized or stationary rollers that enable the efficient and continuous processing of materials such as paper, plastic films, metal foils, and textiles. R2R systems are widely used in industries such as printing, electronics, packaging, and energy storage for the production of items like flexible electronics, batteries, solar cells, and flexible packaging materials. A significant challenge in R2R printing processes is the difficulty of achieving precise alignment tolerances in multi-layer printing structures, as required by the device resolution for multi-layer printed flexible electronics, such as flexible thin-film solar cells, flexible OLED and LCD displays, wearable sensors, and smart textiles. Registration refers to the precise alignment of printed patterns across multiple layers. In practical manufacturing scenarios, the inherent deformability of flexible substrates, coupled with the complex dynamics of rollers, creates substantial challenges for achieving precise registration control. The registration error (RE) is essentially the positional or dimensional deviation that arises from unintended fluctuations in substrate tensions and substrate speeds. Although real-time feedback controllers like Proportional-Integral-Derivative (PID) or Model Predictive Control (MPC) are used for R2R tension control, they are not sufficiently responsive to transient disturbances and do not continually enhance performance when faced with angle-periodic disturbances. These disturbances are common in R2R systems due to roller rotation behaviors. This study presents a novel approach called Spatial-Terminal Iterative Learning Control (STILC) that utilizes an adaptive basis function to eliminate the registration error in an R2R gravure printing process. The adaptive basis function, together with the registration error, is employed to update the STILC compensation profile using a P-type Iterative Learning Control law. Our numerical experiments demonstrate that the proposed STILC method with the adaptive basis function successfully eliminates the RE caused by roller-motor axis mismatches. In addition, we demonstrate through simulations that our method with the adaptive basis function exhibits superior convergence performance compared to the STILC with an invariant basis function. The novel STILC approach is a promising method with significant potential for various industrial applications that encounter spatially periodic disturbances, such as . It offers particular value in scenarios where the system dynamics are not well understood, and there are no predefined state trajectories to guide the control process. This type of uncertainty is frequently encountered in engineering practice, making STILC a valuable tool for addressing real-world challenges in industrial settings. The ability of STILC to adapt and learn from spatially periodic disturbances without relying on pre-established trajectories sets it apart as a versatile and effective solution for a wide range of industrial control applications.

Presenting Author: Zifeng Wang Northeastern University

Presenting Author Biography: Zifeng Wang received his B.S degree in mechanical engineering from Shanghai Jiao Tong University in China in 2018. He was a mechanical engineer at Beko China R&D Center from 2018 to 2020. He is currently a Ph.D. candidate in Industrial Engineering at Northeastern University, Boston, MA, USA. His research interests include online learning control and data-driven methods in advanced manufacturing processes, aimed at leveraging the repetitiveness and abundant data in manufacturing scenarios to improve the automation and intelligence of industrial production systems.

Authors:

Zifeng Wang Northeastern University
Xiaoning Jin Northeastern University