AISI 1040 Çeliğinin Mikroyapı Resimlerinden Mekanik Özelliklerinin Derin Öğrenme ile Tahmini

Year 2024, Volume: 12 Issue: 2, 707 - 718, 29.06.2024

Rıdvan Sert , Ömer Şahin , Volkan Kılıçlı , Fecir Duran

https://doi.org/10.29109/gujsc.1472209

Abstract

Malzeme biliminde işlem-mikroyapı ve mekanik özellikler arasındaki çok iyi bir ilişki bulunmaktadır. Çeliklerin oda sıcaklığındaki mekanik özellikleri doğrudan mikroyapıda bulunan ferrit, sementit ve perlit hacim oranlarına ve tane boyutlarına bağlıdır. Bu çalışmada, AISI 1040 çeliğinin mikroyapı görüntülerinden yapay zekâ ile oda sıcaklığındaki çekme özelliklerinin tahmini gerçekleştirilmiştir. AISI 1040 çeliğinden ASTM-E8/E8M standardına uygun olarak hazırlanan çekme numuneleri oda sıcaklığında çekme testine tabii tutulmuştur. Sonraki adımda aynı çekme numunelerinin deforme olmamış bölgelerinden metalografik numune hazırlanıp mikroyapı resimleri elde edilmiş, ferrit ve perlit hacim oranları görüntü analizi yazılımıyla hesaplanmıştır. Bu veriler ile özgün bir veri seti oluşturulmuştur. Evrişimsel Sinir Ağı kullanılarak, mikroyapı resimlerinden akma, çekme ve kopma gerilimi değerleri tahmin edilmiştir. Gerçekleştirilen deneyler sonucunda mikroyapı resimlerinden AISI 1040 çeliğinin mekanik özelliklerinin başarılı bir şekilde tahmininin gerçekleştirilebileceği ortaya konulmuştur (MSE=4,36, RMSE=2,08, MAE=1,66, R2=0,99).

Keywords

AISI 1040 çeliği, mikroyapı, mekanik özellikler, derin öğrenme, evrişimsel sinir ağı

Prediction of Mechanical Properties of AISI 1040 Steel from Microstructure Images with Deep Learning

Abstract

In materials science, there is a well-established relationship between processing, microstructure, and mechanical properties. The mechanical properties of steels at room temperature are directly dependent on the volume fractions and grain sizes of ferrite, cementite, and pearlite present in the microstructure. This study has implemented the prediction of tensile properties at room temperature for AISI 1040 steel using artificial intelligence based on microstructural images. Tensile specimens prepared according to the ASTM-E8/E8M standard from AISI 1040 steel were subjected to tensile testing at room temperature. Subsequently, metallographic samples were prepared from the undeformed regions of the same tensile specimens and microstructural images were obtained, with ferrite and pearlite volume fractions calculated using image analysis software. These data have contributed to a unique dataset. Using a Convolutional Neural Network, yield strength, tensile strength, and elongation at fracture values were predicted from the microstructural images. The experiments demonstrated that the mechanical properties of AISI 1040 steel can be successfully predicted from microstructural images (MSE=4.36, RMSE=2.08, MAE=1.66, R2=0.99).

Keywords

AISI 1040 steel, microstructure, mechanical properties, deep learning, convolutional neural network

References

• [1] S. Wang, J. Li, X. Zuo, N. Chen, and Y. Rong, “An optimized machine-learning model for mechanical properties prediction and domain knowledge clarification in quenched and tempered steels,” J. Mater. Res. Technol., vol. 24, pp. 3352–3362, 2023.
• [2] G. Xu, J. He, Z. Lü, M. Li, and J. Xu, “Prediction of mechanical properties for deep drawing steel by deep learning,” Int. J. Miner. Metall. Mater., vol. 30, no. 1, pp. 156–165, 2023.
• [3] M. A. Shaheen, R. Presswood, and S. Afshan, “Application of Machine Learning to predict the mechanical properties of high strength steel at elevated temperatures based on the chemical composition,” Structures, vol. 52, pp. 17–29, 2023.
• [4] Y. Diao, L. Yan, and K. Gao, “A strategy assisted machine learning to process multi-objective optimization for improving mechanical properties of carbon steels,” J. Mater. Sci. Technol., vol. 109, pp. 86–93, 2022.
• [5] A. Choudhury, “Prediction and analysis of mechanical properties of low carbon steels using machine learning,” J. Inst. Eng. (India) Ser. D, vol. 103, no. 1, pp. 303–310, 2022.
• [6] J. Xiong, T. Zhang, and S. Shi, “Machine learning of mechanical properties of steels,” Sci. China Technol. Sci., vol. 63, no. 7, pp. 1247–1255, 2020.
• [7] S. M. Azimi, D. Britz, M. Engstler, M. Fritz, and F. Mücklich, “Advanced steel microstructural classification by Deep Learning methods,” Sci. Rep., vol. 8, no. 1, pp. 1–14, 2018.
• [8] J. Muñoz-Rodenas, F. García-Sevilla, J. Coello-Sobrino, A. Martínez-Martínez, and V. Miguel-Eguía, “Effectiveness of machine-learning and deep-learning strategies for the classification of heat treatments applied to low-carbon steels based on microstructural analysis,” Appl. Sci. (Basel), vol. 13, no. 6, p. 3479, 2023.
• [9] A. C. Cheloee Darabi, S. Rastgordani, M. Khoshbin, V. Guski, and S. Schmauder, “Hybrid data-driven deep learning framework for material mechanical properties prediction with the focus on dual-phase steel microstructures,” Materials (Basel), vol. 16, no. 1, p. 447, 2023.
• [10] F. Kibrete, T. Trzepieciński, H. S. Gebremedhen, and D. E. Woldemichael, “Artificial intelligence in predicting mechanical properties of composite materials,” J. Compos. Sci., vol. 7, no. 9, p. 364, 2023.
• [11] D. Pandya and D. Shah, “Experimentation and its prediction of process parameters effects on elongation in tensile test of AISI 1008 steel using ANN model,” Procedia Technol., vol. 14, pp. 282–289, 2014.
• [12] Z. Li, F. Liu, W. Yang, S. Peng, and J. Zhou, “A survey of convolutional neural networks: Analysis, applications, and prospects,” IEEE Trans. Neural Netw. Learn. Syst., vol. 33, no. 12, pp. 6999–7019, 2022.
• [13] S. Albawi, T. A. Mohammed, and S. Al-Zawi, “Understanding of a convolutional neural network,” in 2017 International Conference on Engineering and Technology (ICET), 2017, pp. 1–6.
• [14] A. Gülcü and Z. Kuş, “Konvolüsyonel Sinir Ağlarında Hiper-Parametre Optimizasyonu Yöntemlerinin İncelenmesi,” Gazi Üniv. Fen Bilim. Derg. C Tasar. ve Teknol., vol. 7, no. 2, pp. 503–522, 2019.
• [15] N. Aloysius and M. Geetha, “A review on deep convolutional neural networks,” in 2017 International Conference on Communication and Signal Processing (ICCSP), 2017, pp. 0588–0592.
• [16] A. Dhillon and G. K. Verma, “Convolutional neural network: a review of models, methodologies and applications to object detection,” Prog. Artif. Intell., vol. 9, no. 2, pp. 85–112, 2020.
• [17] R. H. Abiyev and A. Ismail, “COVID-19 and pneumonia diagnosis in X-ray images using Convolutional Neural Networks,” Math. Probl. Eng., vol. 2021, pp. 1–14, 2021.
• [18] “Google colaboratory,” Google.com. [Online]. Available: https://colab.research.google.com/notebooks/welcome.ipynb. [Accessed: 02-Apr-2024].
• [19] M. Ishtiaq, A. Inam, S. Tiwari, and J. B. Seol, “Microstructural, mechanical, and electrochemical analysis of carbon doped AISI carbon steels,” Appl. Microsc., vol. 52, no. 1, 2022.
• [20] X. Tu et al., “Enhancing strain capacity by the introduction of pearlite in bainite and polygonal ferrite dual-phase pipeline steel,” Materials (Basel), vol. 14, no. 18, p. 5358, 2021.