Abstract
Equations of state present computational problems predicting the behavior of near critical reservoirs. Around the critical point the properties of the liquid and the gas become the same and the equations of state often fail to converge, making the compositional simulation impossible. In 1988 Firoozabadi' documented the performance of the industry's most common equations of state. He reported that they can readily predict the behavior of complex reservoir fluid systems at conditions away from the critical regions, but could not describe the sharp changes that take place in the vicinity of the critical point. This study developed two methods for avoiding the computational problems of near-critical reservoirs. The methods are simple and can be readily applied whenever dealing with this kind of reservoir. The methods do not involve the re-tuning of the equation of state because this can be an expensive and time consuming process. The methods approach the problem by artificially modifying two adjustement parameters in the simulator. Method #I, modifies the temperature of the reservoir, and Method #2 modifies the composition (represented by the fraction of gas used in the sample recombination process). By modifying either of this parameters the computational problems around the critical point can be avoided making the compositional simulation feasible. This study will demonstrate that by using either of the two methods presented here, accurate estimates of primary recovery (within 1% of the OOIP and OGIP) can be achieved. This two methods can assist the engineer in the estimation of the primary recovery of near critical reservoirs when time and resources are a limitation but a rapid and accurate solution is required. The methods also demonstrate that for near critical fluids a reservoir can be treated as a volatile oil, or a retrograde gas without having a significant effect on the primary recovery estimates.
Ordonez, Roberto E (1995). Compositional simulation of primary depletion for near critical reservoirs using the VIP simulator. Master's thesis, Texas A&M University. Available electronically from
https : / /hdl .handle .net /1969 .1 /ETD -TAMU -1995 -THESIS -O73.