prediction of the carbon nanotube quality using adaptive neuro–fuzzy inference system

Multi-walled carbon nanotubes (cnts) are synthesized with the assistance of water vapor in a horizontal reactor using methane over co-mo/mgo catalyst through chemical vapor deposition method. the application of adaptive neuro-fuzzy inference system (anfis) technique for modeling the effect of important parameters (i.e. temperature, reaction time and amount of h2o vapor) on the quality of the cnt process is investigated. using experimental data, qualities of cnts are determined for training, testing and validation of developed anfis model. from the analysis carried out by the anfis-based model, the mean square deviation and a regression coefficient are found to be 4.4% and 99%, respectively. the validation results confirm that the ability of the proposed anfis model for predicting the quality of the cnt process over a wide range of operational conditions. in addition, sensitivity analysis indicates that the temperature has the significant effect (i.e. 94%) on the quality of the cnt process.