Iterative learning control of direct write additive manufacturing using online process monitoring

Urbanski, Christopher John

Permalink

https://hdl.handle.net/2142/120082 Copy

Description

Title

Iterative learning control of direct write additive manufacturing using online process monitoring

Author(s)

Urbanski, Christopher John

Issue Date

2023-05-02

Director of Research (if dissertation) or Advisor (if thesis)

Alleyne, Andrew

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• additive manufacturing
• 3D printing
• direct write printing
• material extrusion
• iterative learning control
• process monitoring
• process control
• 3D scanning

Abstract

The development of in situ process monitoring techniques for additive manufacturing (AM) has increased in recent years. While extrusion-based AM methods enable the fabrication of complex structures, ensuring the geometric accuracy of these structures requires direct measurements of the deposited material. Moreover, part fidelity can be improved by implementing control strategies to correct the geometric errors detected through process monitoring. Despite current research focusing on in situ process monitoring, few efforts investigate the relationships between process inputs and resulting print geometry for extrusion-based AM. Consequently, process control is often based on causal relationships or neglected altogether. This work presents a process monitoring and control strategy for reducing the geometric errors in parts fabricated via extrusion-based direct write printing. A laser scanner integrated into the AM system directly measures the deposited material in situ during the print but not in real time. These measurements are processed online with a custom algorithm to determine the material’s spatial placement and bead width errors. An iterative learning control (ILC) algorithm is applied to the deposition process to compensate for the geometric errors. We experimentally validate the process monitoring and control strategy on a direct write printing system by fabricating 3D periodic lattice structures, specifically functionally graded scaffolds. Here, the ILC algorithm uses the online measurements to learn the errors in the structure’s repetitive elements as they are printed, then corrects the errors in subsequently fabricated elements. The results show improved material bead widths in scaffolds fabricated using the proposed process monitoring and control strategy.

Graduation Semester

2023-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2023 Christopher Urbanski

Owning Collections