Residence Time Distribution and Modeling of Mechanical Properties of Extruded Nanocomposite Foams Using Adaptive Neuro‐Fuzzy Inference System

Abstract

Effects of two screw configurations (compression and mixing screws) and two barrel temperatures (150 and 1607C) on residence time distribution (RTD) and modeling of mechanical properties of nanocomposite foams were studied. Multiple inputs single output (MISO) models were developed using Adaptive Neuro-Fuzzy Inference System Modeling (ANFIS) to predict radial expansion ratio (RER), unit density (UD), bulk spring index (BSI) and bulk compressibility (BC) of the nanocomposite foams. An individual ANFIS model was developed for each mechanical property using torque (Nm), pressure (psi) and melt temperature (7C) as input parameters. Models were evaluated by the R 2 of prediction and root mean squared error (RMSE) for training and testing. Wide-angle X-ray diffraction (WAXD) results showed that the diffraction peaks for the mixing screws at 1507C were narrow and had higher intensity indicating a strong intercalation in the nanocomposites. Screw configuration and temperature had significant effects (P,0.05) on the mean residence time (MTR) which affected the mechanical properties of the nanocomposites.