Data-driven modeling for determination of asphaltene stability condition in oil system

Asphaltene precipitation is accounted as one of the most serious problems during oil production so that it can decrease the production of crude oil and cause the blockage of reservoir rock pores, etc. An accurate prediction of phase behaviour of asphaltene is therefore important in oil production industry. Accurate prediction of phase behaviour of asphaltene precipitation i.e. stability state of asphaltene precipitation in oilfields is greatly desirable. To this end, the applicability domains of the most important variables for the determination of the stability state of asphaltene precipitation viz. aromatic + resin and asphaltene + saturates have been specified by using decision tree (DT) algorithm. Next, adaptive neuro-fuzzy inference system (ANFIS) approach was implemented in order to determine the stability state of asphaltene precipitation using the efficient variables of aromatic + resin and asphaltene + saturates. The results obtained in the current study demonstrate that the models proposed in this study provide desirable results in estmating the stability state of asphaltene precipitation in oilfields.