Estimación del esfuerzo de fluencia de soldaduras de aceros microaleados


Bernardo Fabián Campillo Illanes; Edgar López Martínez; Octavio Vázquez Gómez;


Revista Ingeniería Mecánica, Tecnología y Desarrollo




An artificial neural network (ANN) was designed to estimate the yield stress as a function of the chemical composition and hardness of high strength low alloy steels (HSLA) and high strength steels (HSS). The information required for the ANN was obtained by literature search to build a database. The designed ANN was of multilayer perceptón type with back-propagation learning rule 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 predict the yield stress of a microalloyed steel according to the chemical composition and hardness. Once trained and tested the ANN, it was used to estimate the yield stress in the different zones and subzones of the weld of two experimental high strength microalloyed steels.