Experimental and Artificial Neural Network Approaches to Predict the Effect of PVA (Poly Vinyl Acetate) on the Rheological Properties of Water and Crude Oil in EOR Processes

摘要

As consequence of pressure drop in oil reservoirs, production of oil will be difficult and EOR methods are used to improve recovery. In many countries Polyacrylamide is used for polymer flooding in EOR processes, but the problem is that Polyacrylamide is expensive also Acrylamide needs to be handled using best laboratory practices (such as having safe systems of work and wearing appropriate gloves, lab coat etc.) to avoid poisonous exposure since it is a neurotoxin. In this paper, it is intended to investigate the effect of PVA (Poly Vinyl Acetate) on rheological properties (such as viscosity) of crude oil, water, and their mixture and see if it is possible to use this polymer in order to enhance oil recovery. Our objective is to determine the minimum amount of polymer to increase the viscosity of water and decrease the viscosity of crude oil simultaneously. Since even a small change of polymer amount requires extra costs, and experimental investigation of small polymer volume change could be time-consuming, an artificial neural network is developed to estimate polymer volume percentage.