Prediction on tribological properties of carbon fiber and TiO2 synergistic reinforced polytetrafluoroethylene composites with artificial neural networks

In this study, the artificial neural network is applied to predict tribological properties of carbon fiber and TiO2 particle synergistic reinforced polytetrafluoroethylene (PTFE) composites. Based on a measured database of PTFE composites, wear volume loss and friction coefficient are successfully calculated through a well-trained artificial neural network. Results show that the predicted data are well acceptable when comparing with the real test values under different friction conditions (slight, moderate and rigorous test conditions), and friction coefficient hold a closer correlation with the input parameters than wear volume loss. Three-dimensional plots for tribological properties as a function of test conditions and material compositions were established. Improved results can be obtained from a further optimization of the network and an increasing availability of measurement data.