Evaluation and neural network prediction of the wear behaviour of SiC microparticle-filled epoxy resins

One of the main advantageous characteristics of thermosetting resins, which enable to apply them as engineering plastics and as matrices for composite materials, is the possibility of optimising their properties in different ways. This work aims to improve the low abrasive wear resistance of an epoxy resin system by adding microscopic silicon carbide powders in different contents and varying particle sizes. Abrasive tests were carried out through a pin on disc apparatus on specimens from different samples and under different working conditions. The tests highlight that plain and reinforced resins' wear increases both with the contact pressure between the counterparts and the counterface roughness. Moreover, the filled resins' wear resistance increases with the increase of content and dimensions of the filling particles. Finally, an intelligent method based on an artificial neural network was trained, using the experimental dataset, to represent a useful tool for predicting the wear behaviour of plain and filled resins under several working conditions.