Tribological performance of ionic liquid modified graphene oxide/silicone rubber composite and the correlation of properties using machine learning methods

Abstract

AbstractThis study investigates and predicts the tribological properties of an imidazolium ionic liquid modified graphene oxide (ILGO) with silicone rubber (QM) composite. The pin on the disc tribometer was utilized to conduct experimental tribological property analysis, with load, sliding velocity, and temperature as changing parameters. Coefficient of friction (COF) of QMILGO1.5 was 42% lower than that of pure QM. Study found that ionic liquid serves as a self‐lubricating layer for graphene, establishing a solid graphene‐to‐ionic liquid interface bond with the rubber matrix. The experimental data were utilized for training artificial neural networks (ANNs), which were then used to predict the COF of the nanocomposites for values for which the experiment was not performed. The produced composite's predictions of friction coefficient utilizing the ANN technique were quite close to experimental results. The work's fundamental goal is to buy experimentation verifies the COF of functionalized graphene oxide (ILGO) with silicone rubber composite, use the actual experimental values to train a deep neural network using Multilayer perceptron, and then use the trained network to predict the values of COF for which obtaining the values by experimentation was difficult.