Table of Content

Open Access iconOpen Access

PROCEEDINGS

crossmark

Strengthening Mechanical Performance with Robust and Efficient Machine Learning-Assisted Path Planning for Additive Manufacturing of Continuous Fiber Composites

Xinmeng Zha1, Huilin Ren1,*, Yi Xiong1,*

1 School of System Design and Intelligent Manufacturing, Southern University of Science and Technology, Shenzhen, 518055, China

* Corresponding Author: Yi Xiong. Email: email

The International Conference on Computational & Experimental Engineering and Sciences 2024, 30(3), 1-1. https://doi.org/10.32604/icces.2024.011371

Abstract

Additive manufacturing of continuous fiber composites is an emerging field that enables the tunable mechanical performance of composite structure by flexibly controlling the spatial layout of continuous fibers. Transverse isotropic strengthening is advantageous property of continuous fiber, which is favorable to align with the principal stress orientation. However, the accuracy and efficiency of traditional methods for calculating principal stress field are unguaranteed due to the inherent complexity and variability of geometries, material properties, and operational conditions in additive manufacturing. Therefore, a machine learning-assisted path planning method is proposed to robustly and efficiently generate the continuous fiber path that significantly strengthen the mechanical performance of composite structure. The method employs a convolutional neural network enhanced with a self-attention mechanism to accurately predict principal stress fields from complex geometries, diverse material properties, and various operational conditions, ensuring robustness and efficiency of path generation. Subsequently, a wave projection function maps the infill morphology and ratio in alignment with the principal stress field distribution, enabling the generation of continuous fiber paths with minimized cutting points by solving the traveling salesman problem. Upon evaluating the method across a range of cantilever structures under diverse operational conditions and materials, the path generation efficiency remained consistent, and the structures fabricated exhibited superior mechanical performance compared to those with conventional Zig-Zag infill pattern. The method's robustness, efficiency, and effectiveness in strengthening mechanical performance render it highly suitable for real-world industrial applications.

Keywords


Cite This Article

APA Style
Zha, X., Ren, H., Xiong, Y. (2024). Strengthening mechanical performance with robust and efficient machine learning-assisted path planning for additive manufacturing of continuous fiber composites. The International Conference on Computational & Experimental Engineering and Sciences, 30(3), 1-1. https://doi.org/10.32604/icces.2024.011371
Vancouver Style
Zha X, Ren H, Xiong Y. Strengthening mechanical performance with robust and efficient machine learning-assisted path planning for additive manufacturing of continuous fiber composites. Int Conf Comput Exp Eng Sciences . 2024;30(3):1-1 https://doi.org/10.32604/icces.2024.011371
IEEE Style
X. Zha, H. Ren, and Y. Xiong, “Strengthening Mechanical Performance with Robust and Efficient Machine Learning-Assisted Path Planning for Additive Manufacturing of Continuous Fiber Composites,” Int. Conf. Comput. Exp. Eng. Sciences , vol. 30, no. 3, pp. 1-1, 2024. https://doi.org/10.32604/icces.2024.011371



cc Copyright © 2024 The Author(s). Published by Tech Science Press.
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 18

    View

  • 12

    Download

  • 0

    Like

Share Link