Application of evolutionary algorithms in identification of solidification parameters

• Autorzy: Majchrzak E. , Mendakiewicz J. , Paruch M.
• Języki publikacji: EN

Abstrakty

EN

Purpose: The casting-mould system is considered. Additionally, it is assumed that part of internal parameters determining the course of thermal processes, e.g. volumetric specific heat of mould, mould thermal conductivity, casting thermal conductivity and the like is unknown. Formulated in this way an inverse problem can be solved using different methods and in this paper the possibility of evolutionary algorithms application is presented. To solve the problem knowledge of cooling/heating curves at selected set of points from casting/mould domain is necessary. The evolutionary algorithm allows to minimize the fitness function containing the differences between the 'measured' cooling curves and the same curves found on the basis of boundary initial problem numerical solution for the assumed set of parameters. The calculated cooling/heating curves have been found using explicit scheme of finite difference method. It turned out that the algorithm proposed gives sufficiently exact results of identification and it can be successfully applied in the scope of thermal theory of foundry process. Design/methodology/approach: In this work numerical modelling of solidification process is applied. A cast iron solidifying in a sand mould is analyzed. The information concerning the courses of cooling/heating curves at the selected set of points from the domain considered is used in order to identify the unknown parameters of the process analyzed. Findings: Application of evolutionary algorithms gives sufficiently exact results of identification of solidification parameters. Research limitations/implications: Further work requires an introducing of real temperature measurements to the model presented. Practical implications: The paper shows the possibilities of solidification parameters identification on the basis of temperature measurements. Originality/value: The evolutionary algorithms presented allow to identify the parameters of solidification process e.g. volumetric specific heat of mould, mould thermal conductivity, casting thermal conductivity and the like.

Słowa kluczowe

EN

numerical techniques; inverse problem; parameters identification; evolutionary algorithm

PL

techniki numeryczne; problem odwrotny; identyfikacja parametrów; algorytm ewolucyjny

• Wydawca: International OCSCO World Press
• Czasopismo: Journal of Achievements in Materials and Manufacturing Engineering
• Rocznik: 2007
• Tom: Vol. 23, nr 2
• Strony: 67--70
• Opis fizyczny: Bibliogr. 15 poz., rys., tab.

Twórcy

autor

Majchrzak E.

autor

Mendakiewicz J.

autor

Paruch M.

• Department for Strength of Materials and Computational Mechanics, Silesian University of Technology, ul. Konarskiego 18 a, 44-100 Gliwice, Poland,

Bibliografia

• [1] B. Mochnacki, J.S. Suchy, Numerical methods in computations of foundry processes, PFTA, Cracow, 1995.
• [2] W. Kapturkiewicz, Modelling of cast iron solidification, Akapit, Cracow 2003 (in Polish).
• [3] M. Janik, H. Dyja, Modelling of three dimensional temperature field inside the mould during continuous casting of steel, Journal of Materials Processing Technology 157-158 (2004) 177-182.
• [4] E. Majchrzak, J. Mendakiewicz, A. Piasecka-Belkhayat, Algorithm of mould thermal parameters identification in the system casting -mould - environment, Journal of Materials Processing Technology 162-163 (2005) 1544-1549.
• [5] R. Szopa, Macro and macro/micro models of solidification. Numerical aspects of process simulation, Materials Science Forum 539-543 (2007) 2564-2569.
• [6] K. Kurpisz, A.J. Nowak, Inverse Thermal Problems, Computational Mechanics Publications, Southampton-Boston, 1995.
• [7] B. Mochnacki, J.S. Suchy, Identification of alloy latent heat on the basis of mould temperature (Part 1), Archives of Foundry 6/2 (2006) 324-330.
• [8] E. Majchrzak, M. Dziewoński, A. Metelski, Estimation of nuclei density in solidifying casting using the Kolmogoroff model, Proceedings of the 46th International Scientific Conference „Foundry - Solidification and Crystallization of Metals” FOUND'2005, Wisła, 2005, 89-92.
• [9] E. Majchrzak, B. Mochnacki, Identification of thermal properties of the system casting - mould, Materials Science Forum 539-543 (2007) 2491-2496.
• [10] T. Telejko, Analysis of an inverse method of simultaneous determination of thermal conductivity and heat of phase transformation in steels, Journal of Materials Processing Technology 155-156 (2004) 1317-1323.
• [11] J. Arabas, Lectures of evolutionary algorithms, WNT, Warsaw, 2001 (in Polish).
• [12] Z. Michalewicz, Genetic algorithms + Data structures = Evolution programs, Springer-Verlag, Berlin, 1996.
• [13] D. Rutkowska, M. Piliński, L. Rutkowski, Neural networks, genetic algorithms and fuzzy systems, PWN, Warsaw-Łódź, 1997 (in Polish).
• [14] N. Nariman-Zadeh, A. Darvizeh, A. Jamali, A. Moeini, Evolutionary design of generalized polynomial neural networks for modelling and prediction of explosive forming process, Journal of Materials Processing Technology 162-163 (2005) 1561-1571.
• [15] T. Burczyński, A. Osyczka (eds), IUTAM Symposium on Evolutionary Methods in Mechanics, Kluwer Academic Publishers, Dortrecht-Boston-London, 2004.