Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Predicting defects of EBM-based additive manufacturing through XGBoost

XGBoost를 활용한 EBM 3D 프린터의 결함 예측

  • Jeong, Jahoon (Department of Mechanical and Systems Engineering, Korea Military Academy)
  • Received : 2022.02.18
  • Accepted : 2022.03.24
  • Published : 2022.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

This paper is a study to find out the factors affecting the defects that occur during the use of Electron Beam Melting (EBM), one of the 3D printer output methods, through data analysis. By referring to factors identified as major causes of defects in previous studies, log files occurring between processes were analyzed and related variables were extracted. In addition, focusing on the fact that the data is time series data, the concept of a window was introduced to compose variables including data from all three layers. The dependent variable is a binary classification problem with the presence or absence of defects, and due to the problem that the proportion of defect layers is low (about 4%), balanced training data were created through the SMOTE technique. For the analysis, I use XGBoost using Gridsearch CV, and evaluate the classification performance based on the confusion matrix. I conclude results of the stuy by analyzing the importance of variables through SHAP values.

본 논문은 3D 프린터 출력 방식 중 하나인, 전자빔용해법(EBM)의 공정 간에 발생하는 결함에 영향을 미치는 요인들을 데이터 분석을 통해 규명하는 연구이다. 선행 연구들을 기반으로 결함발생에 주요한 원인으로 지목되는 요소들을 참고하였으며, 공정 간 발생하는 로그파일 분석을 통해 결함 발생과 연관된 변수들을 추출하였다. 또한, 해당 데이터가 시계열 데이터라는 점에 착안하여 window의 개념을 도입하여, 현재 공정 층으로부터 총 3개 전 층까지의 데이터를 포함하여 분석에 사용 될 변수들을 구성하였다. 해당 연구의 종속변수는 결함발생유무이기에 이진분류를 통한 분석을 하였으며, 이때 결함 층의 비율이 낮다는(약 4%) 문제로 인해 SMOTE 기법을 적용하여 균형잡힌 훈련용 데이터를 만들었다. 분석을 위해 Gridsearch CV를 활용한 XGBoost를 사용하였고, 분류 성능은 혼동행렬을 기반으로 평가하였다. 마지막으로, SHAP값을 통한 변수 중요도 분석을 통해 연구의 결론을 내렸다.

Keywords

Acknowledgement

This work was supported by 2022 research fund of Korea Military Academy(Hwarangdae Research Institute).

References

  1. J. Patel, Data-Driven Modeling for Additive Manufacturing of Metals: Proceedings of a Workshop, Washington, DC: The National Academies Press, ch. 1, pp. 1-2, 2019.
  2. S. K. Everton, M. Hirsch, P. Stravroulakis, R. K. Leach, and A. T. Clare, "Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing," Metal and Design, vol. 95, pp. 431-445, Apr. 2016.
  3. M. Montazeri, "Smart additive manufacturing: In-process sensing and data analytics for online defect detection in metal additive manufacturing processes," Ph. D. dissertation, University of Nebraska-Lincoln, Lincoln, MA, 2006.
  4. Y. J. Liu, S. J. Li, H. L. Wang, W. T. Hou, Y. L. Hao, R. Yang, T. B. Secrombe, and L. C. Zhang, "Microstructure, defects and mechanical behavior of beta-type titanum porous structures manufactured by electron beam melting and selective laser melting," Acta Materialia, vol. 113, pp. 56-67, Jul. 2016. https://doi.org/10.1016/j.actamat.2016.04.029
  5. O. Kwon, H. G. Kim, M. J. Ham, W. Kim, G. -H. Kim, J. -H. Cho, N. I. Kim, and K. Kim, "A deep neural network for classification of melt-pool images in metal additive manufacturing," Journal of Intelligent Manufacturing, vol. 31, pp. 375-386, Feb. 2020. https://doi.org/10.1007/s10845-018-1451-6
  6. M. Khanzadeh, W. Tian, A. Yadollahi, H. R. Doude, M. A. Tschopp, and L. Bian, "Dual Process Monitoring of Metal-Based Additive Manufacturing Using Tensor Decomposition of Thermal Image Streams," Additive Manufacturing, vol. 23, pp. 443-456, Oct. 2018. https://doi.org/10.1016/j.addma.2018.08.014
  7. T. Chen and C. Guestrin, "XGBoost: A Scalable Tree Boosting System," in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, New York: NY, pp. 785-794, 2016.
  8. Y. Lee and J. Sun, "Predicting Highway Concrete Pavement Damage using XGBoost," Korean Jorunal of construction Engineering and Management, vol. 21, no. 6, pp. 46-55, Nov. 2020.
  9. N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, "SMOTE: Synthetic Minority Over-sampling Technique," Journal of Artificial Intelligence Research, vol. 16, no. 1, pp. 321-357, Feb. 2002. https://doi.org/10.1613/jair.953