A Modern Approach to Reservoir Testing (includes associated papers 22220 and 22327 )
- C.A. Ehlig-Economides (Schlumberger, Houston) | J.A. Joseph (Schlumberger, Houston) | R.W. Ambrose Jr. (Schlumberger, Houston) | Cathy Norwood (Schlumberger, Houston)
- Document ID
- Society of Petroleum Engineers
- Journal of Petroleum Technology
- Publication Date
- December 1990
- Document Type
- Journal Paper
- 1,554 - 1,563
- 1990. Society of Petroleum Engineers
- 5.5.8 History Matching, 4.1.2 Separation and Treating, 2.2.2 Perforating, 4.3.4 Scale, 4.1.5 Processing Equipment, 5.1.2 Faults and Fracture Characterisation, 3.3.1 Production Logging, 5.6.4 Drillstem/Well Testing, 3.2.3 Hydraulic Fracturing Design, Implementation and Optimisation, 4.6 Natural Gas, 5.8.6 Naturally Fractured Reservoir, 7.6.6 Artificial Intelligence, 3 Production and Well Operations, 5.3.2 Multiphase Flow
- 1 in the last 30 days
- 394 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
Summary. Reservoir testing encompasses all measurements in a wellbore in response to fluid flow, including measurements made while perforating or stimulating; formation perforating or stimulating;formation pressure-transient, and pumped-well pressure-transient, and pumped-well tests; and production-log profiles and transients. The wide range of expertise required to interpret reservoir tests can impede the successful extraction of important information derived from these measurements. The purpose of this paper is to show how purpose of this paper is to show how a unified interpretation methodology, underlying a human-engineered computer procedure, reduces the complexity of the interpretation task.
Methodology separates routine data processing from interpretation judgments. The focal point of the process is the determination of an process is the determination of an adequate model of the transient behavior, or diagnosis. Use of the log-log plot of pressure change and its derivative with respect to superposition time is a proven technique for system diagnosis of transient behavior. For tests that involve continuously measured rate in addition to the pressure, the log-log plot of pressure, the log-log plot of normalized pressure change and its derivative with respect to the sandface rate-convolved time function is used. Computer processing provides automated parameter estimation, and the success of interpretation is visually confirmed. Both diagnosis and final verification of the interpretation rely on the log-log plot.
Because high-priority objectives in well testing are to evaluate performance and to determine well and reservoir parameters, these topics have received considerable attention in the literature. In this paper, we review the well-testing literature of 1986-89 and refer the reader to Gringarten for a discussion and survey of pertinent references before 1986. It is quite clear that a cornerstone of modern reservoir-test interpretation involves the use of the log-log presentation. It is also apparent that the pressure presentation. It is also apparent that the pressure derivative, developed 6 years ago, is largely responsible for the current interest in the log-log approach, an interest not enjoyed since its introduction to the petroleum literature 20 years ago.
In the last few years, several interesting variations have been developed for the basic log-log pressure change and Bourdet et al.'s pressure-derivative presentation. Some of these variations include (1)second-order pressure derivatives, designed for simultaneous use with the standard derivative to highlight and emphasize the dominant flow regimes in a transient; (2) the convolution derivative and convolution type curves, which are used analogously to their pressure counterparts when downhole flow rates are measured; (3) the Impulse test, where the formation is subjected to a short-rate impulse and then interpreted with the pressure derivative; and (4) pressure-derivative ratios, a plotting pressure-derivative ratios, a plotting presentation that eliminates the need for presentation that eliminates the need for vertical movement of test data when they are being matched to applicable new type curves. Blasingame et al. used an integral of the pressure method as a smoothing technique to supplement standard transient data analysis.
In addition to these plotting and data-presentation techniques, new interpretation models were proposed. Yaxley explored type curves describing the pressure behavior for a partially sealing fault, and Wong et al. and Cinco-Ley and Meng presented type curves for finite-conductivity fractures induced in homogeneous and fissured rocks, respectively. All studies considered wellbore storage, and the dominant flow regimes in each model were explained in terms of the behavior of its pressure derivative. Derivatives were also recently used to highlight flow-regime definitions used in testing horizontal wells. Hence, it is now common to introduce new models by explaining their behavior and flow-regime identification patterns in terms of the pressure derivative.
It has long been realized that computerized reservoir-test interpretation must be accompanied by some form of estimation algorithms that automatically determines the parameters that provide a match between transient parameters that provide a match between transient data and a selected reservoir model. Gringarten provides a list of papers on this subject; two recent articles are Refs. and 25, but these studies assumed that the correct model was known a priori. Watson et al. proposed a procedure for selecting the most appropriate model from a given pool of candidates, and artificial intelligence pool of candidates, and artificial intelligence and rule-based ap roaches nave been used to solve the same problem.
Our paper addresses the following elements of a successful test:optimized test design, validated data acquisition, use of external data, standardized data presentation and analysis, automated history matching, and result verification. Not all of these elements are necessary for every test, but test design and data validation are paramount concerns. For routine tests in established reservoirs, test objectives may permit a cursory interpretation.
We present a number of field cases to illustrate these elements. Some examples represent textbook-quality data; others illustrate how our approach applies to the practical problems analysts must frequently resolve. problems analysts must frequently resolve. A unified methodology underlies all the interpretations presented.
Optimized Test Design
Reservoir tests provide parameters that complement formation lithology, fluid volumes, and mechanical properties obtained from openhole logs and reservoir geometry determined by geological and seismic means.
JPT, December 1990
|File Size||1 MB||Number of Pages||12|