|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Abstract This paper demonstrates the value of collecting and interpreting real-time data. With an intensive data gathering strategy, starting at wells’ inception to the mature production phase, we show how transient pressure and rate data can be used to manage a complex carbonate gas reservoir. In particular, reservoir connectivity is discerned with pulse testing and with the leading-edge p/q graph, and continuous updates of in-place volume are made with both static and dynamic material-balance methods and corroborating the same with rate-transient analysis. Interwell connectivity information was deduced during underbalanced drilling by way of interference test between two pairs of wells. Thereafter, transient-pressure tests on individual wells characterized the layered, dual-porosity system, with production logs corroborating the notion of layering. Production maturity over three years has paved the way for estimating connected in-place gas volume associated with each well using the transient-PI, and also with a new method introduced here. This new approach entails plotting both static and dynamic material-balance data on the same graph, yielding the same solution. Errors associated with real-time rate measurements presented interpretation challenges for rate-transient analysis; however, application of a physics-based filtering algorithm resolved this issue. Flow-after-flow tests that were embedded in monthly variable-rate production allocations, in turn, allowed us to obtain average-reservoir pressure explicitly to do the static material-balance analysis.
Abstract Material-balance (MB) analysis for in-place volume estimation in gas reservoirs has been in practice for decades. Nonlinear responses from geopressure reservoirs with or without aquifer influx present special interpretation challenges. One of the main challenges of in-place volume estimates involves the estimation of average-reservoir pressure with production. To that end, modern pressure sensors installed at bottomhole and/or surface largely help establish a given well's dynamic performance by way of rate-transient analysis. This paper explores the applicability and limitations of the standard analytical tools in volumetric, geopressure, and waterdrive systems for a diverse array of fluids, from dry gas to near-critical gas/condensate. The systematic approach presented in this paper attempts to increase accuracy in results by ensuring consistency in solutions from multiple methods used to first assess the average-reservoir pressure from production performance data, followed by in-place volume estimation. In this context, we examined analytical tools, such as the pav/z vs. cumulative gas production (Gp) plot, and cumulative reservoir voidage vs. cumulative total expansion plot. Both pot aquifer and unsteady-state Carter-Tracy aquifer models were considered to account for water influx. Besides the use of Cole and drive indices plots, two diagnostic log-log plots are introduced involving total expansivity and change in average-reservoir pressure. In addition, we sought solution objectivity by introducing a diagnostic tool in the Walsh and Yildiz-McEwen MB plots. Both MB methods involve plotting of cumulative reservoir voidage (F) vs. cumulative total expansion (Et), whereas the diagnostic tool consists of plotting F/Et vs. Et on the same graph. Initially, synthetic data helped us understand the overall system behavior and instilled confidence in the solutions obtained for various combinations of drive mechanisms. Statistical design of experiments prompted us to explore independent variables, such as aquifer-to-hydrocarbon PV ratio, production rate, degree of overpressure, and the aquifer source. Those learnings were validated with published and new field data encompassing an array of reservoirs with various drive mechanisms and fluid type.
Abstract This paper presents a simple diagnostic tool to identify reservoir flow behavior from a Cartesian pressure/rate graph. Some of the benefits of the proposed tool are its simplicity without requiring any calculations, leading to understanding of reservoir compartmentalization and application of an appropriate material-balance technique. Data diagnosis entails graphing pressure with rate and discerning trends; positive slope signifies the pseudosteady-state (PSS) flow period, whereas the negative slope implies infinite-acting (IA) flow. Constant-rate production exhibits infinite slope whereas constant-pressure production produces zero slope. Mathematical justifications for these diagnostic signatures are presented. During PSS flow, wells belonging to the same container will exhibit the same slope. Differences in slope are an indication of reservoir compartmentalization, lateral or vertical. Equally important we provide mathematical proof of why different wells in a multiwell reservoir system should have the same slope. Field examples from multiple gas and gas/condensate systems show how the proposed tool works in practice. Introduction With increasing usage of permanent downhole and/or surface sensing, the need for simple diagnostics becomes imperative so that actions can be taken just in time for reservoir management. Studies have shown that motivations for real-time sensing revolve around on-time action to maximize benefits. Various analysis techniques exist to analyze production rate data for estimating in-place fluid volume and remaining reserves. These methods entail from traditional decline curve analysis, such as those offered by Arps (1945) and Fetkovich (1980) to more sophisticated techniques (Agarwal et al. 1999; Blasingame et al. 1991; Blasingame et al. 1989; Mattar and McNeil 1997) involving both flowing bottomhole pressure and rate. Most of these methods apply to single wells in volumetric reservoirs producing single-phase fluids from a fixed drainage boundary. Mattar and Anderson (2003) provide a comprehensive treatment of the pertinent methods. Analytic methods (Marhaendrajana 2005; Marhaendrajana and Blasingame 2001) have also been proposed to handle well interference in multiwell reservoirs. Gringarten (2005) showed that the reservoir-compartmentalization question can be addressed by deconvolving simultaneously measured pressure/rate data for wells across a perceived fault barrier. Multidisciplinary approach has also been reported to address the compartmentalization question (Bigno et al. 1998). Changes in well performance may often be attributed to condensate banking, reservoir subsidence, fines migration precipitating changing skin, and a host of completion and/or wellbore-lift issues, besides depletion. Our challenge is to decipher the real reason for premature production decline. In this regard, Anderson and Mattar (2004) offer a few diagnostic clues about wellbore loading and changing skin, changing well productivity, and identifying external pressure support or interference. This study offers a simple methodology to diagnose long-term well performance, especially those that are influenced by outer boundaries. In particular, whether wells belong to the same or different compartments become quite evident. Because we are solving an inverse problem, independent methods must be used to eliminate potential reservoir, wellbore, and surface flowline network issues before reaching reasonable conclusions. Mathematical proofs are presented in support of the contentions presented in this study. Theoretical Considerations When production is initiated in a well, various flow regimes are encountered as transition from IA to PSS flow, with possible intervening transitional flow, occurs. Fig. 1 schematically depicts such a scenario on a Cartesian pwf-q graph. Of course, the size of the connected-pore volume (CPV) within a well's drainage boundary and the rate of fluid withdrawal dictate the decline rate during PSS flow in a closed system. One complicating factor during the boundary-dominated flow is a well's ever-changing outer boundaries precipitated by changing rates of neighboring wells, drilling infill wells, injecting fluids, and encroaching aquifer, to name a few. For a perspective, Fig. 1 is intended as a practical diagnostic tool in a closed system for reservoirs with significant mobility producing gas or oil, and is not intended for tight-gas reservoirs.
Abstract Estimating in-place volume associated with each well, leading to estimation of total reservoir in-place volume, is the cornerstone to any reservoir-management practice. Yet conventional methods do not always lend themselves for routine applications, particularly when used in singular fashion. However, combining them on the same plot has considerable merit in that they converge to the same solution when material-balance-derived average-reservoir pressure is used. This study presents a systematic procedure for estimating the initial gas in-place volume (GIIP) when real-time surveillance data of pressure, rate, and temperature are available at the wellhead. Specifically, we show that log-log diagnosis, followed by combined static- and dynamic-material-balance analysis, and transient-productivity-index (transient-PI) analysis lead to consistent solutions. Thermodynamic behavior of fluids is also explored to ensure that the converted pressures at the bottomhole and measured rates have consistency and accuracy for reservoir-engineering calculations. Layered systems were selected for this study because they represent most situations. Two synthetic cases probed issues pertaining to average-reservoir pressure computation with the pseudosteady-state (PSS) approach, and two field examples validated the approach presented here.
Summary The rapid pace of exploitation of unconventional gas and light oil plays in North America has necessitated the development of new production-forecasting methodologies to aid in reserves assessment, capital planning, and field optimization. The generation of defendable forecasts is challenged not only by reservoir complexities but also by the use of multifractured horizontal wells (MFHWs) for development. In this work, a semianalytical method (SAM) is developed to provide a solid theoretical basis for forecasting. The technique is analytical in that it uses the methods of Agarwal (2010) to calculate contacted oil in place and contacted gas in place (COIP/CGIP) from production rates, flowing pressures, and fluid properties. The rate-normalized pressure (RNP) derivative (RNP′) is a key component of the calculation; pseudopressure is used for gas cases. The technique is also empirical in that an empirical function is fitted to the resulting COIP/CGIP curve vs. time. Although the method is flexible enough that any equation can be used to represent the COIP/CGIP curve, and hence, the sequence of flow regimes exhibited by MFHWs, the equation must be capable of being integrated to allow the extraction of RNP. The stabilized COIP/CGIP during boundary-dominated flow (BDF) must be specified for forecasting—thereafter, the method uses a material-balance simulator to model BDF. Hence, if the well is still in transient flow, a range in forecasts may be generated, depending on the assumed stabilized COIP/CGIP. The new SAM addresses some of the current limitations of empirical and fully analytical (modeling) approaches. Empirical methods, which have been adapted to account for long transient and transitional flow periods associated with ultralow-permeability reservoirs, lack a theoretical basis, and therefore input parameters may be difficult to constrain. However, empirical methods are simple to apply and require a minimum amount of data for forecasting. Analytical models, while representing the physics better, nonetheless require additional reservoir and hydraulic-fracture data that may not be available on every well in the field. The SAM proposed herein is intended to bridge the gap between empirical and modeling-based approaches—it is more rigorous than purely empirical methods, while requiring a lesser amount of data than fully analytical techniques. The new method is tested against simulated and field cases (tight oil and shale gas). Although a simple power-law function is used in the current work to represent the COIP/OGIP curve, which appears adequate for the cases studied, one should note that wells exhibiting long transitional flow periods (e.g., elliptical/radial) will likely require a different functional form.