Application of ANFIS-GA algorithm for forecasting oil flocculated asphaltene weight percentage in different operation conditions

Abstract

ABSTRACTAsphaltene which cover range of 1% to over 10% of oil by weight, is well-known as most problematic part of oil that can deposit during production in reservoir, well tubing, and surface production lines, and consequently impose a serious restriction on production which in turn increases total cost of entire operation. Through decades an extensive research has been performed in order to identify asphaltene molecular structure, its behavior at different condition, and its separation mechanism from oil. One of most critical parameter associated with asphaltene precipitation modeling is flocculated asphaltene weight percentage in oil at given operation condition. In this study, to eliminate cost and time associated with experimental procedure that concern with determining this critical parameter, a novel ANFIS network with the help of Genetic algorithm has been developed, which trained and tested by over 400 experimental data. The constructed network show good performance regarding this critical-parameter forecasting, and therefore can be used as a general tool in order to provide input for any asphaltene-concern modeling, with confidence.