Prediction of Physical and Mechanical Properties for Metallic Materials Selection Using Big Data and Artificial Neural Networks

Abstract

In this work, a computer-aided tool is developed to predict relevant physical and mechanical properties that are involved in the selection tasks of metallic materials. The system is based on the use of artificial neural networks supported by big data collection of information about the technological characteristics of thousands of materials. Thus, the volume of data exceeds 43k. The system can access an open online material library (a website where material data are recorded), download the required information, read it, filter it, organise it and move on to the step based on artificial intelligence. An artificial neural network (ANN) is built with thousands of perceptrons, whose topology and connections have been optimised to accelerate the training and predictive capacity of the ANN. After the corresponding training, the system is able to make predictions about the material density and Young's modulus with average confidences greater than 99% and 98%, respectively. INDEX TERMS Artificial intelligence, big data, material selection, multilayer feedforward networks, neural network, property prediction, software-based web browser control. LIST OF SYMBOLS AND ABBREVIATIONS ADAM Adaptive Moment Estimation AI Artificial intelligence ANN Artificial Neural Networks β n ADAM algorithm parameter ADAM stability factor ε Prediction error of a neural network η ADAM step size f Error function g Gradient of the error function HTML HyperText Markup Language m ADAM first moment estimate v ADAM second moment estimate w Weights vector

Prediction of Mechanical Properties by Artificial Neural Networks to Characterize the Plastic Behavior of Aluminum Alloys

“…The low hardness of pure aluminum makes it a non-ideal material for building structures. For this application, alloying elements must be added to make it stronger [ 10 ]. These supplementary elements do not simply improve the hardness of the metal, but also modify other properties [ 11 , 12 ].…”

“…There is a wide variety of decision support systems focused on materials engineering [ 18 ]; however, very few really take advantage of technologies based on artificial intelligence [ 9 , 10 , 19 , 20 , 21 , 22 ]. Although several studies that use machine learning to address metallotechnics and the properties of metals have been published [ 23 , 24 ], aluminum alloys have hardly been investigated considering their tempers from an industrial perspective using these tools [ 25 ].…”

“…In this study, a computer-aided methodology is developed to predict some fundamental properties of aluminum alloys whose chemical composition and treatments (thermal and mechanical) are known [ 10 ]. The system presented in this work is able to predict accurately the yield stress (YS) and the ultimate tensile strength (UTS) of aluminum alloys based on Brinell hardness data [ 9 , 26 ].…”

“…In this study, a third new method is going to be developed, similar to the first one, but with the particularity that the function that relates the tensile properties with Brinell hardness will be obtained using artificial intelligence and machine learning [ 41 ]. This new methodology takes, as input, the chemical composition, the temper, and the Brinell hardness and, then, approximates the yield strength and the ultimate tensile strength [ 9 , 10 ].…”

“…Among the most significant new uses, medicine [ 46 , 47 ], warfare [ 48 ], ecology [ 49 ], security [ 50 ], education [ 51 ], oil exploration [ 52 ], or material science [ 44 , 53 ] can be emphasized. Today, ANNs have become one of the most notable AI methods due to their incredible accomplishments and unstoppable advancement [ 10 , 54 ].…”