Abstract

This paper discusses estimation of future recovery from a dissolved-gas reservoir where it is generally necessary to assume various reservoir data that have not been taken, that are not available or whose accuracy is questionable. In this study, past production performance is matched on an individual well performance is matched on an individual well basis through the use of a two-dimensional, two-phase, mathematical reservoir model. By obtaining a production history match, it was found that a reliable estimate of future performance may be obtained. performance may be obtained

Introduction

With the discovery in recent years of the ultra-deep, large volume non-associated gas reservoirs, there may be a tendency to minimize the contribution of casinghead or dissolved gas in satisfying the total U. S. gas requirements. However, in Texas Railroad Commission Districts 7C and 8 during 1966 approximately 785 Bcf of casinghead gas was reported to have been produced. This volume represents about 53 percent produced. This volume represents about 53 percent of the total gas produced in this area. Much of this casinghead gas is sold to gas pipeline companies for transmission to various parts of the U. S. To anticipate future needs, a gas transmission company must be able to forecast the future gas availability from all present sources. Therefore, since casinghead gas is supplying a significant percent of the total gas requirements, a need has developed for a method that will accurately forecast the future availability of gas from these sources.

Many methods for predicting the future gas availabilities from dissolved-gas drive reservoirs have been presented in the literature. However,, the amount and quality of pertinent reservoir data necessary to their solution may somewhat restrict their use. Also, these solutions are generally on a total reservoir basis, assume homogeneity, and do not forecast individual well performance. Since in many dissolved-gas reservoirs the gas production is delivered to more than one purchaser, individual well performance predictions are desirable.

The ability to solve numerically the equations of fluid flow in a petroleum reservoir has given rise to the concept of reservoir modeling. As the name implies, a reservoir model attempts as closely as possible to reproduce mathematically the physical conditions prevailing in a petroleum reservoir. The utility of the model lies in its ability to represent the reservoir by a grid mesh, allowing each block to retain its own rock properties such as permeability and porosity. In the case of the dissolved-gas drive reservoir, however, the availability of necessary reservoir data may pose a stumbling block that must be bypassed if the true potential of modeling is to be realized. These data, such as relative permeability curves, periodic bottom-hole permeability curves, periodic bottom-hole pressure surveys, capillary pressure curves, pressure surveys, capillary pressure curves, and FVT analyses generally are not available to a gas transmission company. In this paper a method of solution is presented whereby individual well performance is forecast using data which generally are available to anyone. In the state of Texas the sources of these data are the published reports of the MER Hearings, the published oil proration schedules, the commercial production reports, and commercial scouting and well log service companies. Using only these data, individual wells' prior production is matched and future production production is matched and future production forecast by solution of a two-dimensional, two-phase, unsteady state mathematical reservoir model.

This content is only available via PDF.