Predicting the effect of nanocomposites on asphaltene removal using a comprehensive approach

Abstract

Asphaltene precipitation is one of the major problems in petroleum industry that causes significant costs to the petroleum industry. Numerous researches have been done by researchers in order to mitigate, inhibit, or postpone these precipitates by using various inhibitors. Nanoparticles and nanocomposites have been applied to overcome this problem recently. In this study, a new model based on Least Square Support Vector Machine (LSSVM) approach is proposed to predict the effect of nanocomposites on asphaltene removal as a function of influencing parameters including nanocomposite type, temperature, pH, and the ratio of nanocomposite concentration to the initial asphaltene content (D/C0). Optimization process is done by Coupled Simulated Annealing (CSA) algorithm for tuning the developed model. Model accuracy is represented by calculating the statistical parameters. The predicted values by the model and actual data were in an excellent agreement with AARE, R2, and MSE values of 2.9431%, 0.9882, and 4.1511, respectively for overall data. Furthermore, sensitivity analysis was conducted on the input parameters which demonstrated that D/C0 and pH has the highest impact on asphaltene removal by nanocomposites and should be considered during process optimization.