Comprehensive Reservoir Characterization Through Data Integration and Numerical Single Well Simulation

Abstract

This reference is for an abstract only. A full paper was not submitted for this conference. Abstract Reservoir heterogeneity, single and multi-phase formation properties dictate well productivity, reservoir performance and management strategy. Cores, open hole logs, formation testers, pressure transient tests, and production logs are usually used to estimate the reservoir heterogeneity and quantify the single and multi-phase formation properties. The results obtained from these techniques are utilized in reservoir simulations to assess the reservoir performance and optimize the reservoir production. A common limitation of these measurement techniques is that they do not provide two-dimensional spatial information of reservoir characteristics. For example, cores and logs have excellent vertical resolutions, but very small lateral radii of investigation, and the pressure transient tests have a large lateral radius of investigation, but very poor vertical resolution. Directly using the results from these measurements in reservoir simulations is difficult because of scale mismatch between the measurements and the requirements in a typical simulation model. Constructing an appropriate simulation model requires rescaling the data, and that may introduce significant uncertainties. This paper presents a comprehensive reservoir characterization method for a Middle East carbonate reservoir through integrating open-hole logs, wireline formation tester results, a series of pressure buildup, stepwise water injection and falloff tests, Electrode Resistivity Array (ERA) measurements and single well simulations. The results obtained in this study represent much larger lateral formation volume than cores and open-hole logs and have a similar vertical scale as a reservoir simulation. The primary objectives of the intentionally established two-phase flow condition wereto allow effective measurements from a specially designed Electrode Resistivity Array;to calibrate a multilayer reservoir model, which was obtained using openhole logs and formation tester results, on a scale that was suitable for reservoir simulations using the ERA measurements; andto estimate effective permeabilities at original connate water conditions as well as the final remaining oil saturation conditions. The testing system included the ERA string and set at the top 120 ft of the production zone, a permanent downhole pressure gauge, and various sensors in a production logging tool positioned near the targeted formation and a surface pressure gauge. The ERA was installed on tubing in a barefoot well rather than permanently cemented outside the casing as in previous applications. This notable difference introduced particular issues in the ERA data acquisition and interpretation, but also provided flexibility for device installation and operation. The 28 electrodes installed in the ERA string obtained sufficient vertical resolution for heterogeneity identification while the direct-current resistivity measurements enabled deep formation investigations. Therefore, the ERA measurements allowed quantifying the formation heterogeneity and assessing multiphase formation properties on the vertical and lateral scales both comparable to reservoir simulations. A dedicated interpretation tool has been developed, and was fully verified using ERA measurements in well A in the carbonate reservoir. The software integrates three essential components for ERA data analysis: dynamic simulation of fluid and salt transport; electrical simulation of potential distribution and current movements; and parameter inversion using optimization algorithms. The results from applications of this software confirmed the fundamentals of the qualitative interpretation method for ERA data. Furthermore, quantitative estimates of the vertical heterogeneity distributions were obtained that were based on extensive sensitivity simulations and history-matching of simulated resistance and ERA measurements. The key advantage of obtaining the permeability heterogeneity is that the scale of the estimated heterogeneity from ERA measurements is the same as the scale of conventional reservoir simulator blocks. Therefore, there is no need to upscale the permeability distributions obtained from ERA measurements. This may eliminate the uncertainty and unreliability of current upscaling techniques for constructing simulation models. Detailed history matching of the ERA measurements on each electrode was conducted in the study. These results were used to improve the multilayer geological model obtained from open-hole logs and wireline formation tester. In addition to the time-lapse resistivity, the testing system used in this study simultaneously recorded transient pressures, and flow rates during the stepwise water injection and falloff tests. These measurements allowed accurate determination of the artificial transients from pump strokes, friction losses in tubing, and flow rate variations. This enabled to reliably simulate the flow dynamics from reservoir to wellhead using an integrated reservoir and wellbore model. The pressure buildup, water injection, and falloff tests were analyzed to estimate the average effective permeabilities at the connate water and remaining oil saturation conditions. These estimated effective permeabilities should be more representative of the reservoir because they are at reservoir conditions and scale. One distinct feature was that the skin factor changed substantially during the tests. The skin evolution was compounded with pressure loss variation in tubing and varying water injection rates, making reliable analysis challenging. A modified Hall plot, capable of handling a zero flow rate condition in a falloff test between two injection periods, was used to quantify the skin evolution. A numerical simulation model was constructed using a reservoir simulator by integratinga dynamic wellbore model to capture the large friction losses in tubing;the multilayer geological model to represent the detailed formation heterogeneity;the average effective permeabilities from the buildup and falloff tests; andthe substantial skin evolution during the injection periods. The numerical simulation model was matched against the pressure and flow rate histories to verify the data obtained from each individual measurement. It is shown that the technology developed in this study has tremendous potential and unique advantages for reservoir description and management. In particular, by integrating currently available techniques (wireline formation tester, pressure buildup, injection and fall-off tests etc), the ERA test yielded reliable estimates of several key properties, such as vertical permeability heterogeneity, connate water saturation, ROS, movable oil, recovery factor, relative permeability, and capillary pressure, at the scale of simulation grid blocks. The results obtained from this study provide the important properties required for reservoir development and management. These results were delivered in a systematic, integrated, and in situ fashion in this study, whereas any individual method of the existing techniques provided limited results with various constraints. The interpretation and history-matching of ERA measurements as well as detailed pressure history during the injection and fall-off tests provided an integrated environment to verify and fine-tune all the results obtained using each individual technique, which led to unified and consistent formation properties for the reservoir. The integrated methodology allowed a comprehensive characterization of the carbonate reservoir and has the potential to be applied in other fields.