SciELO - Scientific Electronic Library Online

 
vol.8 número1Simulador de un LIDAR elástico atmosférico para medir los niveles de concentración de aerosoles en la tropósferaPlataformas de entrenamiento virtuales usando el sensor Kinect, Unity y técnicas de Gamificación índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

  • No hay articulos similaresSimilares en SciELO

Compartir


Acta Nova

versión On-line ISSN 1683-0789

Resumen

MOTA-HERNANDEZ, C. I.; ALVARADO-CORONA, R.; FELIX-HERNANDEZ, J. L.  y  MORALES-MATAMOROS, O.. Analysis and Application of ANN to Layers of Bored Steel. RevActaNova. [online]. 2017, vol.8, n.1, pp.94-108. ISSN 1683-0789.

Three models of artificial neural networks were analysed and applied: Multilayer Perceptron (MLP), Radial Basis Function Network (BRFN) and the combination of both models (RBFN & MLP) for the value of fracture toughness for boriding steels AISI 1045, AISI 1018 and AISI M2. The steels were exposed to a boriding treatment paste, using the model of Palmqvist type cracks, applying the model to the experimental cracking assessment toughness Balankin proposed by Campos and through the modification of the model T. Laugier fracture. The input data for each grid model taking into account the type of steel boriding, the generated layer thickness, the distance of the indentation and the indentation load. The networks were trained with data obtained experimentally under Vickers microindentation testing on iron boride layers at different loads. The data of the fracture toughness of the iron boride layer are compared with experimental data, obtaining errors of 5% for both models of neural networks.

Palabras clave : ANN's; Boriding; Fracture toughness; Indentation.

        · resumen en Español     · texto en Español     · Español ( pdf )

 

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons