Open AccessNondestructiveApproachforDetermination of Steel MechanicalProperties
Author(s) -
Edgar Lopez Martinez,
JazmÃn Y. Juárez-Chávez,
S. Serna,
Beatriz Campillo
Publication year - 2015
Publication title -
international journal of computer and technology
Language(s) - English
Resource type - Journals
ISSN - 2277-3061
DOI - 10.24297/ijct.v14i9.7078
Subject(s) - backpropagation , artificial neural network , sigmoid function , yield (engineering) , multilayer perceptron , welding , alloy , materials science , perceptron , ultimate tensile strength , structural engineering , metallurgy , computer science , engineering , artificial intelligence
It was proposed the design of an artificial neural network (ANN) to estimate the yield strength in the welding zone of HSLA experimental steels. The input parameters of the ANN were the chemical composition and hardness. The information needed to training and testing the ANN was obtained by searching the literature of the yield strength as a function of the input parameters. The design was of the type perceptron multilayer with a rule learning of backpropagation type and sigmoidal transfer function, varying the number of nodes in the hidden layer. It was determined that the design of the ANN with 11 nodes is able to estimate the yield strength of high strength low alloy steels and ultra-high strength steels according to their chemical composition and hardness.