MODELING AND OPTIMIZATION OF THE COMPOSITION OF IRON-BASED ALLOYS BY APPROXIMATION WITH NEURAL MODELS AND GENETIC OPTIMIZATION ALGORITHM

Nikolay Tontchev, Martin Ivanov

Abstract


The research in this paper is intended to recommend an approach for adequate prediction of the properties of iron-based alloys for a preset composition and mode of heat treatment. Stages of creation (design), testing, production and deployment of high strength alloy steel, include the specification of the chemical composition, the parameters of the mode of thermal treatment and the final mechanical properties. The steel for its components and features for heat treatment is a technological object and therefore it is possible to apply for it an approach for modeling the properties and optimizing the composition depending on the particular application. The procedure of a reasoned elaboration of the chemical composition by the number and the amount of alloying elements is relatively new related to the pursuit of the final mechanical properties. The practical results are applicable and they can be used for: - the design of more efficient compositions in terms of the expensive alloying elements while maintaining the basic properties above a given threshold, - evaluation of the technological cost of equally applicable technological variants of varying degrees of doping steel, - determination of a rational representative of a certain class of materials best suited to the requirements previously set (most often controlled properties) among the rest of the class.


Keywords


ferrous alloys, modeling and optimization properties, neural models, genetic optimization algorithm

Full Text:

PDF

References


Bodenhofer, U. (2000). Genetic Algorithms:Theory and Applications, Lecture Notes. (Second Edition - WS 2001/2002). Linz, Austria. Retrieved from http://www4.hcmut.edu.vn/~hthoang/dktm/GA%20Theory%20and%20Application.pdf

Borovikov, V. (2008). Nejronnye seti. Metodologija i tehnologii sovremennogo analiza dannyh (2 ed.). (V. Borovikov, Ed.) Gorjachaja linija – Telekom.

Coley, D. A. (1999). An Introduction To Genetic Algorithms For Scientistsand Engineеrs. Singapore: World Scientific Publishing Co. Pte. Ltd.

Haupt, R. L., & Haupt, S. E. (2004). Practical genetic algorithms (Second Edition ed.). John Wiley & Sons, Inc.

Saaty, T. L., & Vargas, L. G. (2012). Models, Methods, Concepts and Applications of the Analytic Hierarchy Process (Vol. 175). New York - Heidelberg - Dordrecht - London: Springer. doi:10.1007/978-1-4614-3597-6

StatSoft. (2013). Model Extremly Complex Functions, Neural Networks. Retrieved 09 05, 2013, from StatSoft: http://www.statsoft.com/textbook/

Taylor, J. G. (1996). Neural Networks and Their Applications. (J. G. Taylor, Ed.) West Sussex, England, United Kingdom: John Wiley and Sons.

Tosh, C. R., & Ruxton, G. D. (2010). Modelling Perception with Artificial Neural Networks. Cambridge, United Kingdom: Cambridge University Press 2010.

Zubchenko, A. S., Koloskov, M. M., Kashirskii, Y. V., & et al. (2003). List of Grades of Steels and Alloys [in Russian] (2nd ed.). Moscow, Russia: Mashinostroenie.


Refbacks

  • There are currently no refbacks.