Prediction of Steamflood Performance in Heavy Oil Reservoirs Using Correlations Developed by Factorial Design Method

Abstract

Abstract By using the factorial design method, statistically significant correlations have been developed which enable one to predict steamflood performance in terms of project life, oil recovery and cumulative steam-oil ratio. The effects of various reservoir rock and fluid properties and steamflood design and operating variables on steamflood performance were discussed. Introduction The ideal way of predicting reservoir performance under steamflood is through numerical simulation. However, this approach may not be always feasible due to either the lack of a reliable thermal simulator or the lack of qualified personnel to run the simulator. In some instances, such as estimation of oil reserves, screening thermal prospects or making preliminary engineering design, detailed simulation may not preliminary engineering design, detailed simulation may not be warranted. Besides, early in the development of a detailed simulation, a simple method of predicting reservoir performance will provide a means of comparison. Under all performance will provide a means of comparison. Under all these circumstances, some correlations which allow the prediction of steamflood performance without resorting to prediction of steamflood performance without resorting to numerical simulation will be useful. The purpose of this work is to develop such correlations. The end result of this study is a simple computer program, written in BASIC, which can be used to predict steamflood performance in heavy oil reservoirs (Appendix A). A large amount of simulation by use of a three-dimensional numerical model has been made in this work so that the users of the correlations can predict steamflood performance without the expense of doing simulation. The independent variables used in the correlations include reservoir rock and fluid properties such as reservoir thickness, porosity, permeability, initial oil saturation, and oil viscosity, along permeability, initial oil saturation, and oil viscosity, along with steamflood design and operating variables such as pattern size, steam injection rate, and steam quality. With these quantities known, one can use the correlations hereby developed to predict steamflood performance in regard to project life, oil recovery and cumulative steam-oil ratio project life, oil recovery and cumulative steam-oil ratio (SOR). In 1980 Gomaa developed a set of correlation charts for predicting oil recovery and cumulative oil-steam ratio, predicting oil recovery and cumulative oil-steam ratio, emphasizing the effects of steam quality, mobile oil saturation, reservoir thickness and net-gross ratio. One conspicuous absence in the independent variables included in his work is the oil viscosity which could greatly affect the steamflood performance. Gomaa's method uses graphical solutions which usually lack precision because the reading of values from a chart is subject to the user's judgment. The errors will be compounded if several charts needed to be read to obtain the answer. In his method, finding the oil recovery requires the use of no less than four charts. BASIC ASSUMPTIONS The reservoir is horizontal, with no dipping. The reservoir is homogeneous throughout the entire thickness, with no intervening shale breaks. There is neither gas cap at the top nor free gas inside the oil sand. There is no water sand underneath the oil sand. The oil is sufficiently heavy to be adequately represented by a single hydrocarbon component which is non-volatile. P. 67