Makine Öğrenmesi Tabanlı Regresyon Metotları ile Üç Boyutlu (3B) Baskı Parça Kalitesinin Tahmini

Öz

Bu çalışmada farklı alanlarda kullanılacak ürünlerin üç boyutlu yazıcılarda imal edilirken baskı parametreleri dikkate alınarak kullanım amacına göre parçanın pürüzlülük oranı, yüke dayanımı ve çekme kuvvetine göre uzama gerilmesi verileri değerlendirilmiş ve ürün kalitesi makine öğrenmesi regresyon metotları ile optimize edilmiştir. Ürün kalitesinin tahmini için Kaggle platformundan elde edilen “3D Printer Material Requirement” açık kaynak veri seti kullanılmıştır. Bu veri setinde sisteme girdi olarak verilen; katman yüksekliği, duvar kalınlığı, dolgu yoğunluğu, dolgu deseni, nozul sıcaklığı, tabla (yatak) sıcaklığı, baskı hızı, baskı malzemesi (PLA ve ABS) ve fan hızı parametrelerine göre baskı sonucu ürünün pürüzlülüğü, yüke dayanım gücü ve çekme kuvvetlerinin etkisiyle ürünün uzama gerilmesi değerleri incelenmiştir. Bu değerler doğrultusunda da ürünün kullanım amacına göre kalitesi tahmin edilmeye çalışılmıştır. Katman yüksekliği, duvar kalınlığı, dolgu yoğunluğu, dolgu deseni, nozul sıcaklığı, yatak sıcaklığı, baskı hızı, baskı malzemesi ve fan hızı gibi parametreler, çıktı performansını etkileyen temel faktörler olarak kullanılmıştır. Bu çerçevede, Linear Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Gaussian Process Regression (GPR), Multi-Layer Perceptron (MLP) tahmin modelleri geliştirilmiş ve model performansları, doğruluk (R²), hata oranları (RMSE, MSE, MAE) ve işlem süresi gibi metrikler açısından değerlendirilmiştir. Bu yöntemler içerisinde GPR ile uzanım, gerilim mukavemeti ve pürüzlülük açısından en başarılı tahmin oranları sırasıyla 0,98, 0,9 ve 1 olarak elde edilmiştir. Elde edilen bulgular, 3B yazıcıların üretim süreçlerinde kalite tahmini ve optimizasyonu için makine öğrenmesi uygulamalarının etkili bir araç olduğunu göstermektedir. Ayrıca bu çalışma, 3B baskı süreçlerinde kalite kontrolü ve tasarım optimizasyonuna yeni bir perspektif sunmaktadır.

Anahtar Kelimeler

• 3D baskı
• Makine öğrenmesi
• Regresyon
• Baskı parametreleri

Predicting Three-Dimensional (3D) Printing Product Quality with Machine Learning-Based Regression Methods

Öz

This study examines how printing parameters affect the roughness, tensile strength, and elongation of 3D-printed parts used in various applications. Machine learning-based regression models were employed to optimize product quality. The open-source "3D Printer Material Requirement" dataset obtained from the Kaggle platform was utilized to predict product quality. This dataset includes input parameters such as layer height, wall thickness, infill density, infill pattern, nozzle temperature, bed temperature, print speed, printing material (PLA and ABS), and fan speed. These parameters were analyzed for their impact on the product's roughness, load resistance, and elongation under tensile force. Based on these evaluations, product quality was estimated according to its intended use. Parameters such as layer height, wall thickness, infill density, infill pattern, nozzle temperature, bed temperature, print speed, printing material, and fan speed were identified as key factors influencing output performance. Within this framework, prediction models including Linear Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Gaussian Process Regression (GPR), and Multi-Layer Perceptron (MLP) were developed, and their performances were assessed using metrics such as accuracy (R²), error rates (RMSE, MSE, MAE), and computational time. Among these methods, GPR demonstrated the highest prediction accuracy for elongation, tensile strength, and roughness, with respective values of 0.98, 0.9, and 1. The findings indicate that machine learning applications are effective tools for quality prediction and optimization in the production processes of 3D printers. Furthermore, this study provides a novel perspective on quality control and design optimization in 3D printing processes.

Anahtar Kelimeler

• 3D printing
• Machine learning
• Regression
• Printing parameters

Kaynakça

1. J. Park, M. Chang, I. Jung, H. Lee, K. Cho, "3D Printing in the design and fabrication of anthromorphic hands: A review.", Adv. Intell. Syst., vol. 6, pp. 1-13, 2024.
2. M. Sovetova, J. K. Calautit, "Influence of printing parameters on the thermal properties of 3D-printed construction structure", Energy, vol. 305, no. 132265, pp. 1-12, 2024.
3. E. S. Chen, A. Ahmadianshalchi, S. S. Sparks, C. Chen, A. Deshwal, J. R. Doppa, K. Qiu, "Machine learning enabled design and optimization for 3D-Printing of high fidelity presurgical organ models", Adv. Mater. Technol., vol. 2400037, pp. 1-11, 2024.
4. 3D Printer Material Requirement, URL: https://www.kaggle.com/datasets/shubhamgupta012/3d-printer-material-requirement/data.
5. B. Taşar, A. Gülten, "EMG-Controlled Prosthetic hand with Fuzzy Logic Classification algorithm", vol. 321, pp. 321-341, 2017.
6. O. Yaman, T. Tuncer, B. Taşar, "DES-Pat: A novel DES pattern-based propeller recognition method using underwater acoustical sounds", Appl. Acoust., vol. 175, no. 107859, pp. 1-13, 2021.
7. A. K. Tanyıldızı, "Prototype design and manufacturing of a four-legged exploration robot with a three-dimensional (3D) printer", Int. J. 3D Print. Technol. Digit. Ind., vol. 7, no. 2, pp. 233-242, 2023.
8. [8] H. Ma, Y. Kou, H. Hu, Y. Wu, Z. Tang, "An investigate study on the oral health condition of individuals undergoing 3D-Printed customized dental implantation", J. Funct. Biomater., vol. 15, no. 156, pp. 1-12, 2024.

9. [9] F. Gorski, R. Wichniarek, W. Kuczko, M. Zukowska, "Study on properties of automatically designed 3D-printed customized prosthetic sockets", Mater., vol. 14, no. 18, 2021.
10. M. van der Stelt, M. P. Grobusch, A. R. Koroma, M. Papenburg, I. Kebbie, C. H. Slump, T. J. J. Maal, L. Brouwers, "Pioneering low-cost 3D-printed transtibial prosthetics to serve a rural population in Sierra Leone – an observational cohort study", EClinicalMedicine, vol. 35, pp. 1-9, 2021.
11. M. Mao, Z. Meng, X. Huang, H. Zhu, L. Wang, X. Tian, J. He, D. Li, B. Lu, "3D printing in space: from mechanical structures to living tissues", Int. J. Extreme Manuf., vol. 6, no. 023001, pp. 1-10, 2024.
12. A. Zaman, J. Seo, "Design and control of autonomous flying excavator", Machines, vol. 12, no. 23, pp. 1-17, 2024.
13. O. Doğan, M. S. Kamer, "Optimum spur gear design and production with additive manufacturing method", Bitlis Eren Univ. J. Sci., vol. 10, no. 3, pp. 1093-1103, 2021.
14. M. Yang, C. Li, H. Liu, L. Huo, X. Yao, B. Wang, W. Yao, Z. Zhang, J. Ding, Y. Zhang, X. Ding, "Exploring the potential for carrying capacity and reusability of 3D printed concrete bridges: Construction, dismantlement and reconstruction of a box bridge", Case Stud. Constr. Mater., vol. 20, no. e02938, 2024.
15. F. T. Omigdobun, N. O. Uwagboe, A. G. Udu, B. I. Oladapo, "Leveraging machine learning for optimized mechanical properties and 3D printing of PLA/cHAP for bone implant", Biomimetics, vol. 9, no. 587, pp. 1-23, 2024.
16. M. Kasim, B. Owed, "The influence of infill density and speed of printing on the tensile properties of the three-dimension printing polylactic acid parts", J. Eng. Sustain. Dev., vol. 27, no. 1, pp. 95-103, 2023.
17. Y. Zhang, K. Mao, S. Leigh, A. Shah, Z. Chao, G. Ma, "A parametric study of 3D printed polymer gears", Int. J. Adv. Manuf. Technol., vol. 107, pp. 4481-4492, 2020.
18. R. J. R. Pereira, F. A. de Almeida, G. F. Gomes, "A multiobjective optimization parameters applied to additive manufacturing: DOE-based approach to 3D printing", Structures, vol. 55, pp. 1710-1731, 2023.
19. A. R. Sani, A. Zolfagharian, A. Z. Kouzani, "Artificial Intelligence-augmented additive manufacturing: insights on closed-loop 3D printing", Adv. Intell. Syst., vol. 6, pp. 1-19, 2024.
20. O. Sevli, "Prediction of the Material to be Used in 3D Printing with Machine Learning Techniques", Int. J. 3D Print. Technol. Digit. Ind., vol. 5, no. 3, pp. 596-605, 2021.
21. S. R. Dabbagh, O. Ozcan, S. Tasoglu, "Machine learning-enabled optimization of extrusion-based 3D printing", Methods, vol. 206, pp. 27-40, 2022.
22. V. S. Jatti, M. S. Sapre, A. V. Jatti, N. K. Khedkar, V. S. Jatti, "Mechanical properties of 3D-Printed components using fused deposition modeling: optimization using the desirability approach and machine learning regressor", Appl. Syst. Innov., vol. 5, no. 112, pp. 1-15, 2022.
23. N. Çelik, S. Kapan, B. Taşar, "Effects of various parameters on entropy generation and exergy destruction in a coil wire inserted heat exchanger by using deep learning neural network method", Int. Commun. Heat Mass Transf., vol. 161, no. 108481, 2025.
24. N. Çelik, B. Taşar, S. Kapan, "Performance optimization of a heat exchanger with coiled-wire turbulator insert by using various machine learning methods", Int. J. Therm. Sci., vol. 192, no. 108439, 2023.
25. M. Li, S. Yin, Z. Liu, H. Zhang, "Machine learning enables electrical resistivity modeling of printed lines in aerosol jet 3D printing", Sci. Rep., vol. 14, no. 1, 2024.
26. M. R. Ebers, K. M. Steele, J. N. Kutz, "Discrepancy Modeling Framework: Learning missing physics, modeling systematic residuals, and disambiguating between deterministic and random effects", arXiv:2203.05164v2 [stat.ML], 2023.
27. A. Nair, J. Jebakumar, K. Raj, "Machine learning model selection for performance prediction in 3D printing", J. Inst. Eng. India Ser. C, vol. 103, no. 4, pp. 847-855, 2022.