Genetic programming for the prediction of tensile strength of cast iron

• Autorzy: Duda J. , Stawowy A.
• Języki publikacji: EN

Abstrakty

EN

In this paper we propose genetic programming (GP) to predict tensile strength of ductile cast iron. The chemical composition and pouring temperature were used as explanatory input variables (parameters), while tensile strength as dependent output variable (response). On the basis of real data set collected in a one of the Polish foundries, two different models for output variable were developed by genetic programming. Statistical analysis of obtained results and two test cases were employed to compare the accuracy of the GP model with the neural network (NN) model and a linear multiple regression model. The comparison demonstrated that the GP outperforms regress ion techniques, while it is generally worse than NN. Nevertheless GP can be a powerful tool for predicting the mechanical properties of cast iron as it provides a mathematical model, which can be further analyzed.

Słowa kluczowe

EN

information technology; foundry industry; metal parameters prediction; genetic programming

PL

technologia informatyczna; przemysł odlewniczy; programowanie genetyczne

• Wydawca: Komisja Odlewnictwa Polskiej Akademii Nauk Oddział w Katowicach
• Czasopismo: Archives of Foundry Engineering
• Rocznik: 2011
• Tom: Vol. 11, iss. 4
• Strony: 31--34
• Opis fizyczny: Bibliogr. 12 poz., rys., tab.

Twórcy

autor

Duda J.

autor

Stawowy A.

• AGH University of Science and Technology, Faculty of Management, Gramatyka 10, 30-067 Krakow, Poland,

Bibliografia

• [1] J. R. Koza, Genetic Programming: A Paradigm for Genetically Breeding Populations of Computer Programs to Solve Problems, Stanford University (1990).
• [2] M. Perzyk, A. W. Kochański, Prediction of ductile cast iron quality by artificial neural networks, Journal of Materials Processing Technology, No. 109 (2001) 305-307.
• [3] J. Voracek, Prediction of mechanical properties of cast irons, Applied Soft Computing, vol. 1, No. 2 (2001) 119-125.
• [4] A. Stawowy, A. Macioł, R. Wrona, Neural network for prediction of tensile strength of cast iron, Archives of Foundry, vol. 4, No. 11 (2004) 222-227 (in Polish).
• [5] I. Zmak, T. Filetin, Mechanical properties of ductile cast iron determined by neural networks, Proceedings of the Third International Conference on Modeling, Simulation and Applied Optimization, Sharjah, U.A.E January 20-22 2009.
• [6] M. Kovacic, P. Uratnik, M. Brezocnik, R. Turk, Prediction of the bending capability of rolled metal sheet by genetic programming, Materials and Manufacturing Processes, vol. 22, No. 5-6 (2007) 634-641.
• [7] M. Brezocnik, M. Kovacic, M. Psenicnik, Prediction of steel machinability by genetic programming, Journal of Achievements in Materials and Manufacturing Engineering, vol. 16, No. 1-2 (2008) 107-113.
• [8] H-C. Tsai, Y-H. Lin, Predicting high-strength concrete parameters using weighted genetic programming, Engineering with Computers (2011), DOI 10.1007/s00366-011-0208-z.
• [9] M. Brezocnik, J. Balic, K Kuzman: Genetic programming approach to determining of metal materials properties. Journal of Intelligent Manufacturing, 13(1):5-17, 2002.
• [10] C. Puncreobutr, B. Lohwongwatana, P. Chongstitvatana: Genetic Programming Approach to Determining Thermal Properties of Lead-free Solder Alloys, Proc. of National Computer Science and Engineering Conference, Bangkok, Thailand, November 4-6, 2009.
• [11] M. Brezocnik, M. Kovacic, M. Ficko: Prediction of surface roughness with genetic programming, Journal of Materials Processing Technology, vol. 157-158 (2004), 28-36.
• [12] C. Ferreira, Gene Expression Programming: A New Adaptive Algorithm for Solving Problems, Complex Systems, 2001.