Abstract
Machine learning is the process of creating algorithms that extract useful facts from data automatically. The goal of this paper is to use an artificial neural network and a cubic spline model to predict various physical quantities displacement components in a thermoplastic solid, such as elastic waves, vector form, volume fraction field, thermal waves, stress components, and carrier density concentration (plasma waves). The mean absolute scaled error (MASE), the mean absolute percentage error (MAPE), and the symmetric mean absolute percentage errors (SMAPE) are used to compare the accuracy of two models. The true displacements are given their maximum expected values. These factors have also been described using various descriptive statistics and diagrams. Statistical significance was found in the examination of the correlation between the variables, and a comparison was conducted between the findings and prior results acquired by others. The findings show that voids, rotation, optical temperature, and thermal relaxation all have a significant impact on the phenomena, and they are in line with earlier physical findings. Furthermore, it is demonstrated that certain physical variables describing such systems may display this property, allowing for the development of an analytical criterion for the advent of dynamical chaos.
Highlights
Machine learning is the process of creating algorithms that extract useful facts from data automatically. e goal of this paper is to use an artificial neural network and a cubic spline model to predict various physical quantities displacement components in a thermoplastic solid, such as elastic waves, vector form, volume fraction field, thermal waves, stress components, and carrier density concentration. e mean absolute scaled error (MASE), the mean absolute percentage error (MAPE), and the symmetric mean absolute percentage errors (SMAPE) are used to compare the accuracy of two models. e true displacements are given their maximum expected values. ese factors have been described using various descriptive statistics and diagrams
Attia et al investigated the thermoelastic analysis of functionally graded material (FGM) plates sitting on variable elastic foundations using an improved four variables plate theory
E goal of this paper is to predict the displacement components of elastic waves (u), vector form (v), volume fraction field (V), temperature (T), stress components, and carrier density concentration in a thermoelastic solid using an artificial neural network as a machine learning model and the cubic spline model (N). e rotating, semiconducting, and photothermal effects discussed in [1, 5] have been generalized in this study. e symmetric mean absolute percentage errors (SMAPE), mean absolute scaled error (MASE), and mean absolute percentage error (MAPE) are generated to compare the accuracy of two models
Summary
A multilayer perceptron (MLP) is an artificial neural network (ANN) that produces a set of outputs from a set of inputs. E procedure of an artificial neural network (ANN) is to change the weights for a specific training set in order to accurately distribute the provided input patterns. A cubic spline is a piecewise function of three-degree polynomials. A cubic spline is a piecewise function defined on the interval [a, b] and divided into k intervals [xi−1, xi], with the property that a x0 < x1 < · · · < xk−1 < xk b. It is defined by the polynomial of degree 3 on the interval [a, b]. Ei is a residual value, yi is an observed value, and yi is a predicted value
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
