A Simplified Method for Water Coning Predictions

Abstract

Members SPE-AIME Abstract Numerical simulation was used to determine the sensitivity of water coning behavior to various reservoir parameters. The simulation results were then used to develop a simplified correlation for water coning predictions. The correlation is general since the sensitivity studies cover a wide range of reservoir parameters (ratio of vertical to horizontal permeability from 0.01 to 1.0;perforated interval from 20 percent to 80 percent of oil perforated interval from 20 percent to 80 percent of oil zone thickness; production rate from 500 to 2000 RB/D gross fluid; mobility ratio from 1.0 to 10). The correlation is valid to predict watercut performance for most reservoirs with strong bottom performance for most reservoirs with strong bottom water drives except those with local barriers present, those with a high degree of stratification present, those with a high degree of stratification (Dykstra-Parsons Permeability Variation >0.8), or those with a thick oil-water transition zone. With the correlation programmed on a hand-held calculator, field engineers can conveniently predict critical rate, breakthrough time and watercut predict critical rate, breakthrough time and watercut performance. Compared to complicated numerical performance. Compared to complicated numerical models, the correlation is particularly useful when detailed reservoir data are not available, or when decision time and project cost are limited. Field engineers can use the simplified correlation to compare watercut performance for various operating strategies and make appropriate decisions for production operations. production operations Introduction A field engineer involved with coning studies is interested in knowing three things. First, he wants to know the maximum oil production rate at which a well can be produced without coning any water. This oil production rate is called the"Critical Rate." If economic necessity dictates production above this "Critical Rate," he wants to production above this "Critical Rate," he wants to know two additional things, i.e., water break-through time and watercut performance after break-through. Various investigators have attempted to provide coning models to determine (1) critical rate, (2)breakthrough time, and/or (3) watercut performance after breakthrough. The first two problems have been studied analytically and experimentally. Muskat and Wyckoff, Arthur, Chaney et al, and Chierici et al published graphical solutions for critical rate determinations, while Meyer and Garder and Schols provided equations for critical rate calculations. Sobocinski and Cornelius and Bournazel and Jeanson presented empirical correlations for breakthrough time predictions. Bournazel and Jeanson also proposed predictions. Bournazel and Jeanson also proposed empirical correlations for watercut performance predictions. The above methods usually provide a predictions. The above methods usually provide a fairly close approximation of critical rate and breakthrough time. Prediction of watercut performance usually requires the use of complicated and costly numerical models. Letkeman et al used a numerical coning model to match coning history and to investigate various completion and production techniques. Miller and Roger used a numerical simulator to study the effect of different reservoir parameters on the coning performance. Blades and Stright used the Intercomp coning model to study the water coning behavior of heavy oil reservoirs with bottom water drive. Although these numerical models offer a great deal of flexibility, they require highly detailed input data and consume large amounts of computer time and money. The purpose of this paper is to provide field engineers a simplified model programmed on a handheld calculator. With the program, they can conveniently predict critical rate, breakthrough time, and watercut performance without lengthy computations on expensive computers. This paper describes the development of the simplified model, its application and its limitations. DEVELOPMENT OF A SIMPLIFIED CORRELATION FOR WATERCUT PREDICTIONS