The deep high pressure/high temperature (HPHT) dolomite formation in Northern Kuwait has been a challenge with varied production, attributable to reservoir heterogeneity. Due to the tight nature of these rocks, matrix acidizing may not produce desired effects, thus requiring hydraulic fracturing to produce at economic rates. However, the tectonic setting in high stress environment has resulted in subpar success and inconsistent results from stimulation treatments in matrix and hydraulic fracturing applications. This paper presents a multidisciplinary approach to address the limited success in the Northern Kuwait Dolomites. An integrated petrophysical evaluation of the current wells will be followed with multi-well Heterogeneous Rock Analysis (HRA), to evaluate the reservoir heterogeneity across the field and identify the ‘sweet spots’ for future drilling locations. Evaluation and lessons learnt from the past stimulation treatments, will be used to understand geo-mechanical challenges and to help calibrate the Mechanical Earth Model (MEM) for implementation in the future wells. Finally, using a reservoir-centric stimulation design tool, stimulation type (acid fracturing vs proppant fracturing) and stimulation design optimization for future wells will be developed. A reservoir-level petrophysical evaluation of the existing wells was performed and compared to understand the reservoir heterogeneity vis. a vis. production potential. Multiple rock classes were identified within the tight dolomite interval, with a gross thickness of ~250 ft. Starting with log based MEM, results from the image log interpretation and the field observations/measurements from fracture diagnostic tests (Decline analysis, Calibration injection) were used in calibrating the MEM and mapping the Completion Quality (CQ) heterogeneity across the field. This has led to a reservoir-level understanding, which can enable planning optimal well locations, target interval and subsequent well placement/completions methodology. Finally, using the reservoir-centric design tool, an optimum design to effectively stimulate the ultralow-permeability dolomites was determined. The optimization workflow did not only include a single-faceted approach of fracture modeling, but also encompassed a production forecast using the integrated numerical reservoir simulator. Lessons learnt from the optimization workflow were further extended to designing horizontal wells (landing point, trajectory for optimal stimulation geometry), and hence to aid in field development strategy. Using the multidisciplinary unconventional workflow, the heterogeneity in reservoir quality and completion quality was evaluated, both along the wellbore and spatially. In essence, we found that natural fractures along with high Critical Net Pay (CNP) allows you to vertically connect with good RQ and thus, is required for success in these tight reservoirs. Following which, reservoir-centric stimulation design tool enabled optimization of completion and stimulation design in a holistic approach, to maximize appraisal and production opportunities.

With the advent of high-resolution methods to predict hydraulic fracture geometry and subsequent production forecasting, characterization of productive shale volume and evaluating completion design economics through science-based forward modeling becomes possible. However, operationalizing a simulation-based workflow to optimize design to keep up with the field operation schedule remains the biggest challenge owing to the slow model-to-design turnaround cycle. The objective of this project is to apply the ensemble learning-based model concept to this issue and, for the purpose of completion design, we summarize the numerical-model-centric unconventional workflow as a process that ultimately models production from a well pad (of multiple horizontal laterals) as a function of completion design parameters. After the development and validation and analysis of the surrogate model is completed, the model can be used in the predictive mode to respond to the "what if" questions that are raised by the reservoir/completion management team.

We present a novel methodology for interpretation of high-angle and horizontal (HA/HZ) well data enabling comprehensive consideration of HA/HZ logs in large-scale reservoir characterization studies. The first step in the workflow is physics-based modeling and inversion of well logs, which yields reservoir structure (boundaries and faults, dips, cross bedding, etc.) and properties near the wellbore with high resolution. Subsequently, 3D geomodels are automatically updated with geometry and property information obtained in step one from log interpretation. We employ this in a giant carbonate field study to interpret hundreds of HA/HZ wells with the eventual objective to increase recovery by 25%, using our methodology integrated as a Web service into a geomodeling workflow. From the initial 3D geological model constructed using seismic and vertical well data, HA/HZ logs were inverted, and the results propagated back to the reservoir model. We perform refinements in well "curtain" cross-sections to match multiple log data by changing properties, dips, layer thicknesses, boundary and fault positions, and then automatically update the geometry and properties of the 3D model. Such automated 3D model update has never yet been attempted; in the case study, the logs were modeled at the rate of three wells per day, as compare to three days per well previously. Through this innovative approach, we (1) attain geomodels that honor high-resolution data through physics-based log modeling and automatic incorporation of inversion results into reservoir models; and (2) make feasible for reservoir engineers to directly refine geomodels while working on such tasks as formation evaluation (FE) and reservoir characterization. Introduction The structure of reservoir models is primarily based on low-resolution seismic data. High-resolution information about the formation structure and properties contained in the well logs—such as position and shape of bed boundaries, dip and azimuth of sub-seismic faults, or cross bedding— rarely becomes part of the geomodeling, and thus critical knowledge is not reflected in the full-field geological models. With the wealth of data from the new deep directional Electromagnetic (EM) tools on the market today [1], this is a considerable missed opportunity. Well logs are used as a data source for properties population, but usually environmental effects (such as invasion and borehole presence) are ignored. This assumption is invalid even in low-deviation, and especially in HA/HZ wells. Reservoir characterization applications are highly susceptible to model inaccuracies and greatly benefit from improved structure and properties distribution that come from formation evaluation; accurate geomodeling is vital in improving recovery and, ultimately, maximizing the profitability of an oilfield [2]. Building a structural and property earth model that would take well-log data into account requires that the results of log interpretation—updated geometry and physical properties—be integrated with the prior knowledge from seismic data in such a way that the resulting model agrees with both. This is a challenging task, for two main reasons. 1) Reliable interpretation of the modern logging tool responses in HA/HZ wells is only possible with proper, physics-based simulation and inversion; however, such log modeling capabilities are rarely available to geologists, reservoir engineers, and formation evaluation engineers: the simulation codes are very specialized and inherently computationally intensive, requiring high performance computing (HPC) resources. 2) Log interpretation yields high-resolution data, and integrating it with the low-resolution structural and property models is a daunting task.

No abstract

The Lower Tipam sandstone reservoir of Miocene age in the Jaipur oilfield lies within a highly folded and faulted Assam-Shelf basin, in the north eastern part of India. The analysis of this mature field carries a lot of interest not only because the sands within the formation are hydrocarbon bearing but also because of the complexities associated with its evaluation. The complexities in general relate to a heterogeneous reservoir with complex mineralogy, varying water salinities across the field which makes the visualization of a conceptual geological model in the presence of a complex structure a real challenge. The objective of the study was thus to characterize the reservoir at wellbore level and conduct realistic inter-well and reservoir-scale geo-modeling for improved oil-field development by means of a comprehensive, interdisciplinary approach.The Jaipur area is mainly characterized by a Tertiary terrigenous sedimentary sequence comprising of fluvial to deltaic deposits overlying the Precambrian metamorphic Basement. The reservoir is oil bearing without a gas cap. The clay typing and the salinities of water bearing zone have an important bearing on the hydrocarbon saturation computation.Based on the available core data, production data and the volumetric computation at wellbore level using ELANPlus TM , the field portrayed oil occurrence in a unique pattern when visualizing the 'oil down to' i.e. OWC, oil-shale contact (OSC) and wells with residual oil saturation. The pattern appeared unusual in a homogeneous clastic reservoir.The OWC depth in the reservoir showed variations across the field, suggesting the presence of a fault network which also controlled the hydrocarbon entrapment. These faults were not easily identifiable from seismic interpretation; the discrete fault network of the area was analyzed by a process of automated fault identification. Though the difference in the OWC encountered in the wells could be explained based on the sealing faults barriers picked up in the area, but the pattern of the depths of OWC, OSC and distribution of wells with residual oil saturation areally suggests the presence of a stratigraphic trap within the formation. The 3D seismic volume was analyzed to understand the extent of the sand body forming the stratigraphic trap and to define its morphology. Iterative analyses of the seismic attributes suggested the possible geometry of the sand body. Using the log interpretation, an isochore was prepared as per the evolved morphology of the sand and was used for modeling the sub zones.The rock types were modeled geo-statistically to define the spatial heterogeneity and to visualize the distinct compartments within the reservoir. Using single well predictive modeling, the flow units and the petrophysical rock types were defined and the water saturation in the 3 dimensional spaces was modeled through saturation height function for each of the reservoir rock types which were later used for HCIIP calculations.As a result of this study the geo-cellular model of the Lower Tipam r...