Machine Learning-Based Prediction of Mechanical Properties and Performance of Nickel–Graphene Nanocomposites Using Molecular Dynamics Simulation Data

Accurately predicting the mechanical properties of graphene-reinforced metal matrix composites is of utmost importance due to its critical role in the design and utilization of nanocomposite materials. The conventional approach of employing molecular dynamics (MD) simulations for this purpose faces a substantial increase in computational costs when considering the combined effects of multiple factors. In contrast, machine learning (ML) models offer a rapid and efficient alternative by swiftly comprehending and predicting material properties following adequate training. In this paper, we employed a long short-term memory (LSTM) model, based on MD calculation data, to accurately predict the mechanical response and key mechanical properties of nickel–graphene composite nanomaterials. Specifically, we thoroughly investigated the comprehensive impact of temperature, graphene orientation angle, and graphene volume fraction on the mechanical properties. Our verification process revealed that high graphene volume and high orientation angles led to increased dislocation absorption, consequently weakening the composite material. To assess the hardness prediction capabilities, we conducted a comparative analysis between the LSTM model and classical multilayer perceptron (MLP) neural networks, as well as the traditional nonlinear regression method, support vector machine (SVM). The obtained results demonstrated that the LSTM models exhibited a remarkable ability to accurately predict the mechanical properties of nickel–graphene composite nanomaterials, showcasing Pearson correlation coefficients exceeding 0.95 when compared to the calculation data. Moreover, the LSTM model effectively comprehends and predicts the complete indentation depth–force curve, thus providing enhanced predictions of material properties. This study proposes an innovative combination of MD simulations and ML models, which holds significant application potential in predicting and designing the performance of graphene-reinforced metal matrix composite materials.