Turan, Muhittin; Kahya, Volkan; Yaylacı, Ecren Uzun; Yaylacı, Murat

This paper introduces a highly accurate and computationally efficient shear deformable finite element model for the static analysis of bi-directional functionally graded beams (BD-FGBs) with various boundary conditions grounded in the first-order shear deformation theory (FSDT). The model, featuring ten degrees of freedom across five nodes, excels in capturing both axial and shear deformations with remarkable precision while maintaining a streamlined formulation. In a novel approach, Artificial Neural Network (ANN) methods are also employed alongside the finite element analysis, offering a dual-method investigation into the static behavior of BD-FGBs. This paper aims to further advance the understanding of BD-FGM beams by exploring their static behavior under diverse loading conditions and boundary constraints, employing advanced finite element methods and artificial neural network techniques. The material properties are modeled through power-law distributions, and the governing equations are derived from Lagrange’s principle. Displacements and stresses were computed under different boundary conditions (BCs), slenderness ratios (L/h), and power-law indices (px, pz). Comparative analysis with existing literature reveals the superior suitability of the proposed finite element model for static analysis, while the ANN approach further reinforces its potential as a robust, complementary tool. The innovative combination of these methods promises to offer significant contributions to the field and provides new insights into the behavior of BD-FGBs under static loads.

AtıfScopus20·WoS24·OpenAlex3

İncele

Yaylacı, Murat; Yaylacı, Ecren Uzun; Turan, Muhittin; Özdemir, Mehmet Emin; Öztürk, Şevval; Ay, Sevil

In this study, the two-dimensional crack problem was investigated by using the finite element method (FEM)based ANSYS package program and the artificial neural network (ANN)-based multilayer perceptron (MLP) method. For this purpose, a half-infinite functionally graded (FG) layer with a crack pressed through two rigid blocks was analyzed using FEM and ANN. Mass forces and friction were neglected in the solution. To control the validity of the crack problem model exercised, the acquired results were compared with a study in the literature. In addition, FEM and ANN results were checked using Root Mean Square Error (RMSE) and coefficient of determination (R2), and a well agreement was found. Numerical solutions were made considering different geometric parameters and material properties. The stress intensity factor (SIF) was examined for these values, and the results were presented. Consequently, it is concluded that the considered non-dimensional quantities have a noteworthy influence on the SIF. Also FEM and ANN can be logical alternative methods to time-consuming analytical solutions if used correctly.

AtıfScopus27·WoS30·OpenAlex6

İncele

Yaylacı, Ecren Uzun; Öner, Erdal; Yaylacı, Murat; Özdemir, Mehmet Emin; Abushattal, Ahmad; Birinci, Ahmet

This paper investigates the artificial neural network (ANN) to predict the dimensionless parameters for contact pressures and contact lengths under the rigid punch, the initial separation loads, and the initial separation distances of a contact problem. The problem consisted of two elastic infinitely layers (EL) loaded by means of a rigid cylindrical punch and resting on a half-infinite plane (HP). Firstly, the problem was formulated and solved theoretically using the Theory of Elasticity (ET). Secondly, the contact problem was extended based on the ANN. External load, the radius of punch, layer heights, and material properties were created by giving examples of different values used at the training and test stages of ANN. Finally, the accuracy of the trained neural networks for the case was tested using 134 new data, generated via ET solutions to determine the best network model. ANN results were compared with ET results, and well agreements were achieved.

AtıfScopus62·WoS60·OpenAlex13

İncele

Yaylacı, Murat; Eyüboğlu, Ayşegül; Adıyaman, Gökhan; Yaylacı, Ecren Uzun; Öner, Erdal; Birinci, Ahmet

This paper presents a comparative study of different methods, such as the analytical method, finite element method (FEM), and multilayer perceptron (MLP) for analyzing a frictionless receding contact problem. The problem consists of two layers resting on a Winkler foundation. The top layer is functionally graded (FG) along the depth and pressed using a rigid cylindrical stamp, whereas the bottom layer is homogeneous. We assumed that the contact between the two layers, and that between the FG layer and the rigid cylindrical stamp are frictionless; additionally, compressive normal tractions can be transmitted through the interface. First, the problem was solved analytically using the theory of elasticity and integral transform techniques. Second, the finite element solution of the problem was obtained using ANSYS software. Finally, the problem was extended based on the MLP, which an artificial neural network used for different problem parameters. The results of this study showed that the variations in the contact lengths at the interface between the rigid cylindrical stamp and the FG layer, those between the homogeneous layer and the FG layer, and the maximum contact pressures at these interfaces depended on various dimensionless quantities such as the stamp radius, stiffness parameter, shear modulus ratio, and elastic spring constant ratio. We observed that the results obtained with the three different methods, namely the analytical method, FEM, and MLP, are extremely compatible with each other, thus proving the accuracy of these results.

AtıfScopus87·WoS86·OpenAlex40

İncele