Predicting Alkaline Surfactant Polymer flood performance at fullfield scale is an interestingly vast subject. Typically, this would comprise of detailed modelling exercises at different dimensions and resolutions to calibrate the model for observed physical and geo-chemical behavior during the fluid flow. Modelling of different physical and chemical phenomena active during the ASP flood offers different complexities. To name a few, the phenomena active during the ASP flood in a reservoir are changing salinity, changing interfacial tension, changing capillary pressure and relative permeability, viscosity modification, chemical transport and retention, residual resistance factor, shear effects and so on. Nonetheless, the degree of areal and vertical heterogeneity in the field has a significant imprint on the ASP flood as an effective recovery process and poses another hurdle in the flood optimization. Also, due to large number of variables associated with the ASP process itself, the number of uncertainty and design parameters increase tremendously in comparison to any conventional recovery technique. The large list of sensitivity variables with increased computational time to model ASP active processes, makes it further restrictive to carry-out a holistic sensitivity analysis.
This paper investigates and addresses the aspects of a field scale development of an ASP flood with the help of full-fledged ASP modelling for our example field. The paper will cover the entire spectrum of the ASP feasibility study from illustrating the approach adopted to simulate the ASP active phenomena to the demonstration of the flood optimization techniques for the field. We will illustrate how the degree of areal and vertical heterogeneity in the field can cause inefficient utilization of chemicals, which in our case was found to have the most limiting effect on ASP field performance. We will discuss some of the ways to optimize the ASP flood in such a heterogeneous reservoir system. The selection of the uncertainty and design parameters and their ranges will be shown and the outcomes of the sensitivity analysis will be presented, including creaming curve analysis. We will also highlight the fullfield 3D effects on some of the conventionally established ASP parameters, for instance slug size.