PSI - Issue 68

Dj. Ivković et al. / Procedia Structural Integrity 68 (2025) 839 – 844 Dj. Ivković et al. / Structural Integrity Procedia 00 (2025) 000–000

843

5

a

b

Error

Fracture toughness

112

100 120

0 1 2 3 4 5 6 7 8

104

7

0 20 40 60 80

65

64

Error, %

1,5

X5CrNiMo 17-12-2 Fracture toughness, MPa·m1/2

X6Cr17

CES EDU PACK 2010 ANN

X5CrNiMo 17-12-2

X6Cr17

Fig. 3 Comparison of fracture toughness values founded in CES EDU PACK 2010 and values predicted by ANN (a) calculated error (b). . Conclusion Materials engineering represents a key scientific field, as it provides wide specter of information about various materials and their properties. These properties are of great importance for constructors and it allows quality, reliability and longevity of products to be raised. Material testing requires a fair amount of resources as well as time. When fatigue limit and fracture mechanic tests are conducted, test time for a single sample can take hours. To be able to save some of the resources, artificial neural networks are used to predict values of material properties, based on selected input data. In this case, two ANN were created for fatigue limit and fracture toughness prediction. Property prediction was based on chemical composition of materials. After training of ANN, chemical composition of X5CrNiMo17-12 2 and X6Cr17 was inserted, and fatigue limit and fracture toughness values were predicted. Obtained values were compared with values from CES EDU PACK 2010. The difference between predicted values and real values is less than 10%, so it could be concluded that following network model with described architecture and parameters can be used successfully for predicting mentioned properties of stainless-steel grades. The accuracy of network could be further improved through increasing number of input parameters and increasing number of data sets used for training. Acknowledgements Research presented in this paper was partially financially supported by the project TR35024 of the Ministry of Education, Science and Technological Development of Republic of Serbia. References Basheer, I., Hajmeer, M., 2000. Artificial neural networks: fundamentals, computing, design, and application, Journal of Microbiological Methods, 43, 3-31. https://doi.org/10.1016/S0167-7012(00)00201-3 EN 10088-1:2005. Stainless steels – Part 1: List of stainless steels https://standards.iteh.ai/catalog/standards/cen/952db42f-8160-4518-8932 c51bc76f8715/en-10088-1-2005 EN 10088-2:2005. Stainless steels – Part 2: Technical delivery conditions for sheet/plate and strip of corrosion resisting steels for general purposes https://standards.iteh.ai/catalog/standards/cen/5da77ead-c665-4c16-a063-1b086a1543c2/en-10088-2-2005 EN 10088-3:2005. Stainless steels – Part 3: Technical delivery conditions for semi-finished products, bars, rods, wire, sections and bright products of corrosion resisting steels for general purposes. https://standards.iteh.ai/catalog/standards/cen/4e6c80d2-c72d-42b3-aeae-2564ae23eb38/en 10088-3-2005 EN 10088-4:2009. Stainless steels – Part 4: Technical delivery conditions for sheet/plate and strip of corrosion resisting steels for construction purposes https://standards.iteh.ai/catalog/standards/cen/a9506eec-011c-47b3-9313-6ec0bb544a7b/en-10088-4-2009