Machine Learning Model Selection for Performance Prediction in 3D Printing

被引：1
作者：
Nair A. [1 ]
J J. [1 ]
Raj K. [1 ]
机构：
[1] Department of Mechanical Engineering, Kalasalingam Academy of Research and Education, Virudhunagar, Krishnankoil
来源：
Journal of The Institution of Engineers (India): Series C | 2022年 / 103卷 / 04期
关键词：
3D printing; Elongation; Machine learning; Prediction; Regression; Tension strength;
D O I：
10.1007/s40032-022-00835-7
中图分类号：
学科分类号：
摘要：
Prediction of elongation, roughness, tensile strength yield of a 3D printed product is important to boost the quality of the printing process. This paper inspects and compares most machine learning regression strategies to build a robust representation that best models the 3D printing process. 3D printing process is affected by nine primary input factors such as layer height, wall thickness, infill density, infill pattern, nozzle temperature, bed temperature, print speed, material and fan speed. These factors influence the quality of the 3D printed product. Multiple machine learning models are run based on the experiments and the best model is selected based on the prediction performance. Among the various machine learning strategies, the best is identified which best predicts the output factors. It was inferred that “radial basis function regressor” algorithm was able to best predict the results for elongation with a mean absolute error of 0.3421 and root mean square error of 0.3421. “Additive regression” was able to best predict the results for elongation with a mean absolute error of and 28.2969 and root mean square error of 39.2895. “Random committee regression” was able to best predict the results for tensile strength with a mean absolute error of 4.112 and root mean square error of 5.2789. © 2022, The Institution of Engineers (India).
引用
收藏
页码：847 / 855
页数：8
相关论文
共 13 条
  • [1] Shahrubudin N., Lee T.C., Ramlan R., An overview on 3D printing technology: technological, materials, and applications, Procedia Manuf., pp. 1286-1296, (2019)
  • [2] Tiwari S.K., Pande S., Agrawal S., Bobade S.M., Selection of selective laser sintering materials for different applications, Rapid Prototyping J., 21, 6, pp. 630-648, (2015)
  • [3] Ferretti P., Leon-Cardenas C., Santi G.M., Sali M., Ciotti E., Frizziero L., Relationship between FDM 3D printing parameters study: parameter optimization for lower defects, Polymers, 13, 13, (2021)
  • [4] Essinger S.D., Rosen G.L., An introduction to machine learning for students in secondary education, Digital Signal Processing Workshop and IEEE Signal Processing Education Workshop, pp. 243-248, (2011)
  • [5] Villanueva Zacarias A.G., Reimann P., Mitschang B., A framework to guide the selection and configuration of machine-learning-based data analytics solutions in manufacturing, Procedia CIRP., 1, 72, pp. 153-158, (2018)
  • [6] Delli U., Chang S., Automated process monitoring in 3D printing using supervised machine learning, Procedia Manuf., 26, pp. 865-870, (2018)
  • [7] Bajic B., Bajic B., Cosic I., Lazarevic M., Sremcev N., Rikalovic A., Machine Learning Techniques for Smart Manufacturing: Applications and Challenges in Industry 4.0. in Conference: 9Th International Scientific and Expert Conference, pp. 29-38, (2018)
  • [8] Lieber D., Stolpe M., Konrad B., Deuse J., Morik K., Quality prediction in interlinked manufacturing processes based on supervised & unsupervised machine learning, Procedia CIRP., 7, pp. 193-198, (2013)
  • [9] Ademujimi T.T., Brundage M.P., Prabhu V.V., A review of current machine learning techniques used in manufacturing diagnosis, IFIP Adv. Inf. Commun. Technol., 513, pp. 407-415, (2017)
  • [10] Dey A., Machine learning algorithms: a review, Int. J. Comput. Sci. Inf. Technol., 7, 3, pp. 1174-1179, (2016)
12