Asphaltene deposition in petroleum reservoirs: dynamic test & connectionist modeling

Asphaltene Deposition is a complex process that can lead to a decline in oil production rates due to permeability and wettability alteration. Asphaltene instability occurs due to variations in thermodynamics properties such as pressure, temperature, and mixture composition. In this study, dynamic experiments were conducted using oil samples to measure important phase behaviour properties such as bubble point pressure (BPP) and the amount deposited as asphaltene. A thermodynamic model was also developed to determine equilibrium composition of the oil samples considering asphaltenes. We investigated the potential application of using feed-forward Artificial Neural Network (ANN) optimized by Imperialist Competitive Algorithm (ICA) and Particle Swarm Optimization (PSO) to estimate the boiling point pressure and asphaltene deposition. Comparison between the smart technique predictions and the experimental data shows an acceptable match. It is found that pressure drop and temperature are the most important factors contributing to asphaltene precipitation. Employing laboratory PVT data and connectionist modeling can result in the construction of an asphaltene phase envelope through an effective and accurate manner. The outcomes of this study, in terms of thermodynamic framework and predictive tools, appear to be useful in the design stage of more efficient EOR processes.