▎ 摘　　要

The present work primarily focuses on the thermo-mechanical and physical properties of polymer composites. The effects of normal load, sliding frequency, and temperature on the reciprocating sliding performance of Carbon fiber (CF) segments are evaluated. These CF segments are coated with graphene nanoplatelets (GNPs), and carbon fiber-reinforced polymer (epoxy) composites (CFRPs). Experiments are carried out at three different normal loads (55 N, 75 N, and 95 N), sliding frequency (8 Hz, 12 Hz, and 16 Hz), and temperature (25 degrees C, 40 degrees C, and 60 degrees C). The experiments are conducted for 20 min with a constant stroke length of 1.5 mm against a 62 HRC hardened chrome steel ball as a counter surface material. Optical microscopy and Scanning electron microscope (SEM) have been used to examine the worn surfaces. In addition, an artificial neural network (ANN) is applied to model the coefficient of friction on the basis of different parameters. It has been observed that adding GNPs to carbon fiber strands improves the strand's bonding tendency with the matrix, resulting in an increase in wear resistance. With the addition of GNPs on carbon fiber strands, the specific wear rate decreased by 32.39%, respectively, while the thermal conductivity increased by 47.95%. The coefficient of friction increases with the normal load up to 75 N and then drops. This is due to fiber fragments and friction film at high loads. As the sliding frequency and temperature increase, the friction coefficient gets reduced. It has been found that the ANN model is able to predict the coefficient of friction with high accuracy.