Evaluating Stochastic Fundamental Natural Frequencies of Porous Functionally Graded Material Plate with Even Porosity Effect: A Multi-Machine Learning Approach.

Autor: Raturi, Himanshu Prasad, Kushari, Subrata, Karsh, Pradeep Kumar, Dey, Sudip
Předmět:
Zdroj: Journal of Vibration Engineering & Technologies; Feb2024, Vol. 12 Issue 2, p1931-1942, 12p
Abstrakt: Purpose: The present paper proposes to explore the fundamental natural frequencies of a porous functionally graded material (FGM) plates. The effect of porosity in an FGM structure is studied in an uncertain quantification domain. Method: A finite element model is developed considering isoperimetric quadratic element and evaluating the first three natural frequencies for the porous FGM plate. Hereafter, a stochastic approach is explored by incorporating Monte Carlo simulation (MCS) and application of machine learning (ML) models. A detailed comparison is conducted to evaluate the predictive efficiency of five machine learning models viz., radial basis function (RBF), linear regression (LR), Gaussian progression regression (GPR), artificial neural network (ANN), and support vector machining (SVM). Results and Conclusion: The authenticity of the models is evaluated based on the mean and standard deviation error analysis. The error analysis provides adequate confidence on the authenticity of the ML models. The predicted results are depicted in the form a three-dimensional probability density function and scatter plots. The effective results and discussion portray an efficient stochastic model to quantify the uncertainty in the porous FGM structure. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index
Externí odkaz: