Prediction and Optimization of Electric Resistance of Printed Graphene Nanosheet Based on Artificial Intelligence

Graphene has attracted wide attention in the field of electromagnetism and has become a front runner in fabricating electronic devices, especially flexible ones. However, currently, there’s no report on the prediction of its electric resistance, which is critical to an electric component. In this work, several machine learning (ML) and artificial neural network (ANN) models are trained to predict the electric resistance(R) of printed graphene nanosheets with only the knowledge of the concentration of graphene ink, the width, and the length of the printed nanosheet. Three different ML algorithms, i.e., Kriging, Efficient Global Optimization (EGO), and Support Vector regression (SVR) are exploited, together with ANNs with different sets of parameters. Harboring high accuracy and yielding satisfactory results, the models can be applied to predict the resistance of graphene nanosheets in real-time working conditions. It paves the way for the wide application of printable graphene nanosheets in electric devices.