1. Artificial neural networks prediction of in-plane and out-of-plane homogenized coefficients of hollow blocks masonry wall
- Author
-
Karam Sab, Myriam Laroussi Hellara, Ioannis Stefanou, Houda Friaa, Abdelwaheb Dogui, École Nationale d’Ingénieurs de Monastir (ENIM), Géotechnique (CERMES), Laboratoire Navier (NAVIER UMR 8205), and École des Ponts ParisTech (ENPC)-Centre National de la Recherche Scientifique (CNRS)-Université Gustave Eiffel-École des Ponts ParisTech (ENPC)-Centre National de la Recherche Scientifique (CNRS)-Université Gustave Eiffel
- Subjects
Accuracy and precision ,Hollow blocks masonry ,Computer Science::Neural and Evolutionary Computation ,02 engineering and technology ,01 natural sciences ,Homogenization (chemistry) ,Artificial neural networks (ANN) ,Out of plane ,0203 mechanical engineering ,0103 physical sciences ,[SPI.MECA.MEMA]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Mechanics of materials [physics.class-ph] ,Back-propagation ,In-plane and out-of-plane loadings ,010301 acoustics ,Influence of bond ,Mathematics ,Computer simulation ,Artificial neural network ,business.industry ,Mechanical Engineering ,Structural engineering ,Masonry ,Condensed Matter Physics ,Equivalent elastic properties ,Backpropagation ,Finite element method ,020303 mechanical engineering & transports ,Orthotropic Love-Kirchhoff plate ,Mechanics of Materials ,Periodic numerical homogenization ,business - Abstract
International audience; A masonry wall is a composite structure characterized by a large variety in geometrical and material parameters. The determination of the effective macroscopic properties, through the homogenization scheme, depends on a great number of variables. Thus, in order to replace heavy numerical simulation, in this paper, the use of artificial neural networks (ANN) is proposed to predict elastic membrane and bending constants of the equivalent Love-Kirchhoff plate of hollow concrete blocks masonry wall. To model the ANN, a numerical periodic homogenization in several parameters is used. To construct the model, five main material and geometrical input parameters are utilized. Multilayer perceptron neural networks are designed and trained (with the best selected ANN model) by the sets of input-output patterns using the backpropagation algorithm. As a result, in both training and testing phases, the developed ANN indicates high accuracy and precision in predicting the equivalent plate of a hollow masonry wall with insignificant error rates compared to FEM results.
- Published
- 2020
- Full Text
- View/download PDF