Application of artificial neural networks in modelling of normalised structural steels mechanical properties

• Autorzy: Dobrzański L. A. , Honysz R.
• Języki publikacji: EN

Abstrakty

EN

Purpose: This paper presents the application of artificial neural networks for mechanical properties prediction of structural steels after heat treatment. Design/methodology/approach: On the basis of such input parameteres, which are the chemical composition, the kind of mechanical and heat treatment and dimensions of elements, mechanical properties, such as strength, impact resistance or hardness are predicted. Findings: Results obtained in the given ranges of input parameters show very good ability of constructed neural networks to predict described mechanical properties for steels after heat treatment. The uniform distribution of descriptive vectors in all, training, validation and testing sets, indicate about the good ability of the networks to results generalisation. Practical implications: Created tool makes possible the easy modelling of described properties and allows the better selection of both chemical composition and the processing parameters of investigated materials. At the same time the obtainment of steels, which are qualitatively better, cheaper and more optimised under customers needs is made possible. Originality/value: The prediction possibility of the material mechanical properties is valuable for manufacturers and constructors. It allows preserving the customers quality requirements and brings also measurable financial advantages.

Słowa kluczowe

EN

artificial intelligence network; computational material science; computational mechanics; artificial neural networks; mechanical properties

PL

sieci sztucznej inteligencji; materiałoznawstwo numeryczne; mechanika numeryczna; sztuczne sieci neuronowe; właściwości mechaniczne

• Wydawca: International OCSCO World Press
• Czasopismo: Journal of Achievements in Materials and Manufacturing Engineering
• Rocznik: 2009
• Tom: Vol. 32, nr 1
• Strony: 37--45
• Opis fizyczny: Bibliogr. 28 poz., rys., tabl.

Twórcy

autor

Dobrzański L. A.

autor

Honysz R.

• Division of Materials Processing Technology, Management and Computer Techniques in Materials Science, Institute of Engineering Materials and Biomaterials, Silesian University of Technology, ul. Konarskiego 18a, 44-100 Gliwice, Poland,

Bibliografia

• [1] M. W. Blicharski, Introduction to materials engineering. WNT, Warsaw, 1998, (in Polish).
• [2] L. A. Dobrzański, Metal engineers materials, WNT, Warsaw, 2004, (in Polish).
• [3] L. A. Dobrzański, Engineering Materials and materials design. Fundamentals of materials science and physical metallurgy, WNT, Warsaw-Gliwice, 2006, (in Polish).
• [4] L. A. Dobrzański, R. Honysz, Materials science virtual laboratory - innovatory didactic tool in the teaching of material engineering performed by traditional and e-learning methods, Proceedings of the 4th International Conference Mechatronic Systems and Materials, Białystok, 2008.
• [5] L. A. Dobrzański, A. Jagiełło, R. Honysz, Virtual tensile test machine as an example of material science virtual laboratory post, Journal of Achievements in Materials and Manufacturing Engineering 27/2 (2008) 207-210.
• [6] L. A. Dobrzański, J. Trzaska, Application of artificial neural networks in mechanical properties modelling of constructional steels, Proceedings of the Materials, Mechanical and Manufacturing Engineering Scientific Conference, Gliwice, 2000.
• [7] W. Jasinski, Materials science: lectures, Kielce University of Technology publishing house, Kielce, 2003, (in Polish).
• [8] P. Korczak, H. Dyja, E. Łabuda, Using neural network models for predicting mechanical properties after hot plate rolling processes, Journal of Materials Processing Technology 80-81 (1998) 481-486.
• [9] U. Markowska-Kaczmar (ed.), Neural networks in applications, Wrocław University of Technology publishing office, Wrocław, 1996, (in Polish).
• [10] T. Masters, Neural networks in practice, PWN, Warsaw (1996) (in Polish).
• [11] W. S. McCulloch, A logical calculus of the ideas immanent in nervous activity, Bulletin of Mathematical BioPhysics 5 (1943) 115-133.
• [12] D. Rutkowska, M. Pilinski, L. Rutkowski, Neural networks, genetic algorithms and fuzzy systems, PWN, Warsaw, 1996, (in Polish).
• [13] W. Sitek, L. A. Dobrzański, J. Zacłona, The modelling of high-speed steels’ properties using neural networks, Journal of Materials Processing Technology 157-158 (2004) 245-249.
• [14] Z. Sterjovski, D. Nolan, K. R. Carpenter, D. P. Dunne, J. Norrish, Artificial neural networks for modelling the mechanical properties of steels in various applications, Journal of Materials Processing Technology 170 (2005) 536-544.
• [15] R. Tadeusiewicz, Elementary introduction for neural networks techniques with sample applications, Academic Publishing House PLJ, Warsaw, 1998, (in Polish).
• [16] J. Trzaska, L. A. Dobrzanski, Application of neural networks for designing the chemical composition of steel with the assumed hardness after cooling from the austenitising temperature. Journal of Materials Processing Technology 164-165 (2005) 1637-1643.
• [17] L. C. Yin, j. Zhang, J. Zhong, Application of neural networks to predict the elevated temperature flow behavior of a low alloy steel, Journal of Computational Materials Science (2009) (in print).
• [18] W. You, Y. X. Liu, B. Z. Bai, H. S. Fang, RBF-type artificial neural network model applied in alloy design of steels. International Journal Of Iron And Steel Research 15/2 (2008) 87-90.
• [19] J. Żurada, M. Barski, W. Jedruch, Artificial neural networks. PWN, Warsaw, 1996, (in Polish).
• [20] http://www.hutabatory.com.pl/
• [21] http://www.statsoft.pl/
• [22] PN-EN 10025:2007
• [23] PN-EN 10083-2:2008
• [24] PN-EN 10028-3:2008
• [25] PN-EN 10083-3:2008
• [26] PN-EN 10089:2005
• [27] PN-EN 10085:2003
• [28] PN-EN 10084:2008