Dynamic mechanical properties of PTFE based short carbon fibre reinforced composites: experiment and artificial neural network prediction

Dynamic mechanical properties (storage modulus and damping) of short fibre reinforced composites were investigated in a temperature range from −150 to 150 °C. A series of polytetrafluoroethylene (PTFE) based composites blended with different contents of polyetheretherketone (PEEK) and reinforced with various amounts of short carbon fibres (CF) was considered in this paper. Dynamic mechanical thermo-analysis (DMTA) was employed using a three-point-bending configuration. The influence of different characteristics of PTFE and PEEK at various temperatures was also considered. Based on measured results an artificial neural network (ANN) approach has been introduced for further prediction purposes. The analysis shows that the number of training dataset plays a key role to the ANN predictive quality. In addition, the more complex the nonlinear relation between input and output is, the larger is the number of training dataset required. The simulation result has shown an example that the ANN is a potential mathematical tool in the structure–property analysis of polymer composites.