Scientific paper ID 1475 : 2017/3
ABOUT THE INSTRUMENTS OF MULTIPARAMETRIC APROXIMATION AND MULTI-CRITERIAL OPTIMIZATION
Nikolay Tontchev, Martin Ivanov
A multi-criteria approach is applied for an expert assessment of the impact of the elements in the alloy composition on a priori chosen quality indicators to improve the mechanical properties of the products. In the paper is present approach is recommended for adequate prediction of the properties of magnesium alloys in the set composition and mode. To identify the optimal composition and experimentation beyond using a proxy database to determine the most accurate approximation fulfilling certain conditions. The approximation is done by means of neural models. This approach allows more precise automatic prediction of the properties of the alloys according to the composition alloy.
Моделиране на свойства Парето фронт Многокритериална оптимизация Магнезиеви сплави.Materials Science Modelling of properties multi-criteria optimization magnesium alloys.Nikolay Tontchev Martin Ivanov
 Tontchev N. Materials Science, Effective solutions and Technological variants, 2014/3/3, LAMBERT Academic Publishing.
 Dobrzański L.A, R. Honysz, Informative technologies in the material products designing, Journal of Archives of Materials Science and Engineering, Vol. 55, (2012), pp.37-44.
 Dobrzański L.A, R. Honysz, Application of artificial neural networks in modelling of quenchedand tempeRp0,2d structural steels mechanical properties, Journal of Achievements in Materials andManufacturing Engineering, 40/1, (2010)50-57.
 Edited by J.G. Taylor, Neural Networks and Their Applications, King”s College London, John Wiley & Sons Ltd, 1996.
 Tosh Colin R., Graeme D. Ruxton, Modelling Perception With Artificial Neural Networks, Cambridge University PRp0,2 ss 2010.
 StatSoft, Electronic Statistics Textbook: http://www.statsoft.com/textbook/.