Well Testing and Analysis Techniques for Layered Reservoirs

Abstract

Summary This paper presents new testing and analysis techniques to obtain individual-layer permeabilities and skin factors for layered reservoirs. The new multilayer testing technique consists of a number of sequential flow tests with a production logging tool that simultaneously measures the wellbore pressure and flow rate at the top of each layer. Two different analysis techniques are presented to estimate layer parameters. The first technique, which is the logarithmic convolution parameters. The first technique, which is the logarithmic convolution method, estimates the approximate values of parameters. The second technique, which is the nonlinear least-squares estimation method, improves the first estimates. It is shown that layer permeabilities and skin factors can be estimated uniquely from simultaneously measured wellbore pressure and flow-rate data that are acquired from all layers sequentially. It is also shown that these individual-layer parameters cannot be estimated from the conventional drawdown- or buildup-test wellbore pressure data. Several synthetic examples are presented to illustrate pressure data. Several synthetic examples are presented to illustrate the application of multilayer testing and analysis techniques. Introduction Most oil and gas reservoirs are layered (stratified) to various degrees because of sedimentation processes over long geological times. Layered reservoirs are composed of two or more layers that may have different formation and fluid characteristics. These reservoirs are usually divided into two groups:layered reservoirs without crossflow (commingled systems), where layers communicate only through the wellbore, andlayered reservoirs with crossflow, where layers communicate at the contact planes throughout the reservoir. Accurate determination of permeability, skin factor, and pressure for each layer is permeability, skin factor, and pressure for each layer is necessary to understand the reservoir performance. For example, unbalanced depletion of layers with different parameters creates many problems, such as high GOR in parameters creates many problems, such as high GOR in high-permeability layers. Conventional buildup tests from layered reservoirs usually suffer from crossflow between layers, particularly if the layers communicate only through the wellbore and/or the permeability contrast between layers is high. The crossflow problem becomes more severe if the pressure and/or the drainage radius of each zone is different. The wellbore crossflow may continue during the entire period of the buildup test. A false straight line on a period of the buildup test. A false straight line on a semi-log plot may even be observed. In many instances, the pressure data alone may not reveal any information about pressure data alone may not reveal any information about the wellbore or formation crossflow. Even if crossflow is not a complicating factor, the major problem for layered systems is still the estimation of problem for layered systems is still the estimation of individual-layer permeabilities and skin factors from conventional well tests. The conventional drawdown and buildup tests usually reveal only the behavior of the total system. Furthermore, the behavior of a multi layer formation may not be distinguished from the behavior of a single-layer formation even though a multilayer reservoir may have a distinct behavior without wellbore storage effects. There are, however, a few special cases where the conventional tests may work. A detailed study of the behavior of two-layer reservoirs with crossflow was done by Prijambodo et al. Unlike many earlier researchers, they investigated the effect of each layer's skin on the semilog straight-line behavior of the two-layer systems without the wellbore storage effect. They also examined limitations of the semilog methods for two-layer systems with many different combinations of vertical and horizontal permeabilities and skin factor with and without crossflow. permeabilities and skin factor with and without crossflow. A different approach for estimating layer parameters by means of optimum test design was investigated by Dogru and Seinfeld. They used a numerical model, but did not include skin and wellbore storage effects. Nevertheless, they show that there are serious problems with the ability to observe and question how well-posed parameter estimation is for layered reservoirs with a parameter estimation is for layered reservoirs with a single transient test. Most work on layered reservoirs has been the derivation of solutions for boundary-value and initial-value problems, the investigation of sensitivity of solutions to layer parameters (forward problem), and the estimation of the parameters (forward problem), and the estimation of the average flow capacity and skin factor of the total formation from a single transient well test. We present a new testing technique for layered reservoirs to estimate individual layer permeabilities and skin factors uniquely. This new test will be called a "multilayer test" hereafter. The multilayer testing technique consists of a number of sequential flow tests, with a production logging tool measuring the wellbore pressure and flow rate at the top of each different layer. SPEFE P. 342