Prediction of Carbon dioxide Hydrate Phase Equilibria in the presence of inhibitor Using Neruo-Fuzzy

In this communication two adaptive neuro fuzzy inference system (ANFIS) models for predicting carbon dioxide hydrate dissociation conditionsin the presence of Methanol and sodium chlorideare presented. The models were trained by some selected reliable data from literature. ANFISmodels have been examined with someexperimental data which have not been used in the training and testing. Also these models have been compared to a well-known thermodynamicmodel. In all cases the ANFIS model was more accurate. Maximum absolute average deviation in ANFIS models for trained data for Methanol and NaCl were 0.88% and 0.07% respectively