Modeling of precipitated asphaltene using the ANFIS approach

Asphaltene precipitation is introduced as a seriously problematic issue in the petroleum reservoirs that causes filling porosity of rocks and reduction of oil production. Hence, estimating the amount of precipitated asphaltene has huge importance in preventing the deposition of asphaltene. The present study was done to estimate the precipitated asphaltene as a function of temperature, dilution ratio, and molecular weight of different n-alkanes using adaptive neuro-fuzzy inference system (ANFIS). Moreover, another new scaling model was also developed to compare with the ANFIS model. In addition, these two developed models have been compared with previously developed correlations. The obtained values of R² for the ANFIS and scaling models were 0.9912 and 0.9862, respectively. These tools are simple to use and can be used as an accurate approximation of the precipitated asphaltene as a function of temperature, dilution ratio, and molecular weight of different n-alkanes.