Wyniki wyszukiwania - Biblioteka Nauki

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

1 Journal of Achievements in Materials and Manufacturing Engineering

1 2008

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Material parameters identification by use of hybrid GA

100%

Majak J. , Toompalu S. , Pohlak M.

Journal of Achievements in Materials and Manufacturing Engineering

2008

tom Vol. 27, nr 1

63-66

Purpose: of this paper is to develop material parameters identification algorithm for yield criterion BBC2003 using global optimization techniques. Design/methodology/approach: An algorithm proposed is based on use of error minimization function, which allows considering over-constraining. Due to strong nonlinearity of the problem considered a number of solutions is available. In order to determine global extreme two stage GA (global optimization technique) is treated. Findings: Numerical material parameters identification algorithm is developed. An approach provided allows reducing significantly the dimension of the nonlinear system before its numerical solution. Convergence to global extreme can be expected due to global optimization technique employed. Research limitations/implications: An analysis is done by keeping formability analysis in mind and only material parameters involved in yield criterion in space of principal stresses are considered. Thus the results can be generalized by including terms corresponding to shear stresses. Practical implications: Advanced yield criteria like BBC2003 are still not used extensively due to the complexities accrued: increasing number of material parameters (additional tests), a complex non-linear programming problem. An algorithm proposed simplifies the material parameters identification process for considered yield criteria BBC2003. The formability analysis of the 6000 series aluminium alloy sheet AA6181-T4 is considered as a case study and used for testing the algorithm proposed. Originality/value: In the case of posed optimization problem the dimension of the design space is reduced from six to two. Over-constraining and under-constraining are considered in algorithm (situations, where number of unknown parameters is not equal with the number of given constraints, are covered).