Bibliographic citations
Llanos, H., (2016). Análisis del comportamiento no lineal de una estructura de pavimento flexible mediante el empleo de redes neuronales [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/5404
Llanos, H., Análisis del comportamiento no lineal de una estructura de pavimento flexible mediante el empleo de redes neuronales [Tesis]. : Universidad Nacional de Ingeniería; 2016. http://hdl.handle.net/20.500.14076/5404
@mastersthesis{renati/707786,
title = "Análisis del comportamiento no lineal de una estructura de pavimento flexible mediante el empleo de redes neuronales",
author = "Llanos Álvarez, Henry Gastón",
publisher = "Universidad Nacional de Ingeniería",
year = "2016"
}
The use of falling weight deflectometer – based back calculation techniques to determine the cape module is a largely method used in the developed countries. Many of commercial programs they do not take into account nonlinear, consequently realistic do not produce aftermaths. The Mich - Pave is structural- analysis software of pavements with methods of finite elements incorporated her no linearity, that it was used to generate a data base synthetic, to develop the structural based model in artificial neural networks. ANN SOM it was used to classify the data base in clusters with similar characteristics, to form groups training, validation and test that is utilized in ANN backpropagation. ANN backpropagation was trained to backcalculating the mechanical properties. This thesis showed than by ANN employed with structure proper, the coefficient correlation (R), in the asphaltic concrete module Eac, base and/or sub base k1 and subgrade K1, 0.975, 0.913 and 0.979 are respectively. Consequently ANN can be utilized as a fast and tool to the back calculating of the cape modules considering nonlinear.
This item is licensed under a Creative Commons License