|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
ABSTRACT The new-generation oil-base mud (OBM) microresistivity imagers provide photorealistic high-resolution quantified formation imaging. One of the existing interpretation methods is based on composite processing providing an apparent resistivity image largely free of the standoff effect. Another one is the inversion-based workflow, which is an alternative quantitative interpretation, providing a higher quality resistivity image, button standoff, and formation permittivities at two frequencies. In this work, a workflow based on artificial neural networks (NNs) is developed for quantitative interpretation of OBM imager data as an alternative to inversion-based workflow. The machine learning approach aims to achieve at least the inversion-level quality in formation resistivity, permittivity, and standoff images an order of magnitude faster, making it suitable for implementation on automated interpretation services as well as integration with other machine learning based algorithms. The major challenge is the underdetermined problem since OBM imager provides only four measurements per button, and eight model parameters related to formation, mud properties, and standoff need to be predicted. The corresponding nonlinear regression problem was extensively studied to determine tool sensitivities and the combination of inputs required to predict each unknown parameter most accurately and robustly. This study led to the design of cascaded feed-forward neural networks, where one or more model parameters are predicted at each stage and then passed on to following steps in the workflow as inputs until all unknowns are accurately obtained. Both inverted field data sets and synthetic data from finite-element electromagnetic modeling were used in multiple training scenarios. In the first strategy, field data from few buttons and existing inversion results were used to train a single NN to reproduce standoff and resistivity images for all other buttons. Although the generated images are comparable to images coming from inversion, the method is dependent on the availability of field data for variable mud properties, which at the moment limits the generalization of the NNs to diverse mud and formation properties. In the second strategy, we utilized the synthetic responses from a finite element model (FEM) simulator for a wide range of standoffs, formation, and mud properties to develop a cascaded workflow, where each stage predicts one or more model parameters. Early stages of the workflow predict the mud properties from low formation resistivity data sections. NNs then feed the estimated mud angle and permittivities at two frequencies into next stages of the workflow to finally predict standoff, formation resistivity, and formation permittivities. Knowledge of measurement sensitivities was critical to design the efficient parameterization and robust cascaded neural networks not only due mathematically underdetermined nature of the problem but also the wide dynamic range of mud and formation properties variation and the measurements. Results for processed resistivity, standoff, and permittivity images are presented, demonstrating very good agreement and consistency with inversion-generated images. The combination of two strategies, training on both synthetic and field data, can lead to further improvement of robustness allowing customization of interpretation applications for specific formations, muds, or applications.
Chen, Yong-Hua (Schlumberger) | Omeragic, Dzevat (Schlumberger) | Habashy, Tarek (Schlumberger) | Bloemenkamp, Richard (Schlumberger) | Zhang, Tianhua (Schlumberger) | Cheung, Phillip (Schlumberger) | Laronga, Robert (Schlumberger)
The new high-definition oil-based mud (OBM) imager is a pad-based microelectrical imager operating at high frequency to establish capacitive contact with the formation in wellbores filled with nonconductive mud. From multiple modes of operation, formation resistivity-like images are generated using an efficient composite data-processing scheme that approximates formation resistivity either by filtering or applying a correction to minimize the contribution of the OBM to the measured signal. Data from the different modes are ?blended? based on estimated formation parameters to generate an optimized image. This approach requires some knowledge of mud electrical properties.
In addition to the composite processing scheme, we also developed a model-based parametric inversion for quantitative interpretation. The Gauss-Newton algorithm matches the measurements to an accurate computationally efficient approximate forward model built by multidimensional fitting of the data generated using a finite-element simulation. The workflow overcomes the underdetermined inversion problem and calibration limitations of the measurements. The inversion allows flexible model definition and parameterization, including refinement of the calibration, and can process intervals of logging data and measurements from multiple buttons simultaneously. The workflow stabilizes the inversion and improves the consistency of the processed results. To overcome the underdetermined nature of the problem and speed up the inversion, we use a sequence of inversion runs to first iteratively estimate the mud properties for a small depth section of the log; this estimate is then used to invert for the button standoff and the formation resistivity and permittivity for longer data sections.
C/O tools are being applied in an increasingly wider variety of borehole configurations. One such is a barefoot completion. This paper will discuss laboratory and computer modeling of the C/O response for Halliburton?s C/O tool (RMTE centered or eccentered in uncased wells spanning the range of 4 ½-in.. to 12 ¼-in.. Borehole fluids of fresh, 100kppm saltwater, and oil were used. Test formations included sandstone, dolostone, and limestone with freshwater, oil, and 100 kppm saltwater saturating the pore space. Generally, the sensitivity of the C/O response to formations containing oil was higher as compared with the C/O response for cased wells of the same borehole size. However, borehole size and fluid effects were also somewhat greater. A technique for translating these measurements to standard conditions (i.e., a 7-in.. cased hole with a 10-in.. borehole is also described. Translation to standard conditions has the advantage that it allows use of a standard interpretation package for those who process such data.
ABSTRACT Natural fractures maintain a significant role in many hydrocarbon plays, in both conventional and unconventional reservoirs. In exploration and development scenarios, specific fracture properties, such as orientation and density, are important. However, more critical is their internal architecture: are the fractures open to fluid flow or filled with minerals? Borehole microresistivity imaging tools are widely used to determine these fracture characteristics. In wells drilled with water-based muds, open fractures are filled with conductive borehole fluid that enables distinguishing open, water-filled fractures from resistive, mineral-filled fractures and the surrounding rock. However, many wells today are drilled with oil-based muds. In this case, mineral-filled fractures and oil-based-mud-filled fractures are equally highly resistive and cannot be directly distinguished using resistivity images only. The latest-generation wireline oil-based-mud microresistivity imagers operate in the megahertz frequency range, radiating the electrical current capacitively through the nonconductive mud column and delivering photorealistic borehole images. Both electrical conductivity and dielectric permittivity components constitute the measured signal. The quantitative interpretation uses a sequence of model-based parametric inversion runs to first estimate the mud properties of the log and subsequently invert for the standoff of the microelectrode buttons to the rock surface and the formation resistivity and dielectric permittivity within the volume of investigation. Our example case shows highly resistive, high-angle fractures from the resistivity images with their orientation and density. The standoff image determines if the mud column penetrates the fracture plane, showing an apparently high standoff compared with the surrounding rock. If the standoff appears high in the fracture plane, the fracture is classified open to fluid flow. However, are these fractures indeed fully dilated and open or are they filled with different materials—are they partially mineralized with calcite and partially open, filled with mud? To further determine the fracture fill and susceptibility to fluid flow, a new workflow employs the material dependency of the relative dielectric permittivity. The relative permittivity is estimated as function of resistivity and frequency pixel by pixel on the resistivity image. The estimate formula is empirically derived from several hundred laboratory measurements on core plugs with different fluid saturations and salinities. The resulting borehole image enables distinguishing materials in the volume of investigation. The new image shows that drilling-induced fractures have low values, which correspond to the oil in open fractures controlled by the mud. The image also shows some fractures with slightly elevated values, corresponding to rock-forming minerals (calcite) and partially low values, which are interpreted as mud, saturated with oil. As result, these types of fractures are classified as partially open with vuggy mineral fill, consistent with the core description. Higher values on the image are attributed to shales and other rocks with raised clay content. Simulation results confirm the sensitivity of such estimated relative permittivity with respect to rock parameters such as rock-matrix permittivity and water-phase tortuosity.
Chen, Yong-Hua (Schlumberger-Doll Research) | Omeragic, Dzevat (Schlumberger-Doll Research) | Habashy, Tarek (Schlumberger-Doll Research) | Bloemenkamp, Richard (Etudes et Productions Schlumberger) | Zhang, Tianhua (Etudes et Productions Schlumberger) | Cheung, Phillip (Etudes et Productions Schlumberger) | Laronga, Robert (Services Techniques Schlumberger)
The high-definition oil-based-mud (OBM) imager is a pad-based microelectrical wireline tool designed to operate in wellbores filled with nonconductive mud. To complement standard composite data processing and provide quantitative interpretation, we developed a model-based parametric inversion using the Gauss-Newton algorithm that matches the measurements to an accurate and efficient forward model built by multidimensional fitting of simulated data. The inversion-based workflow allows flexible selection of model parameters to be inverted and can process logging data from multiple depths and buttons simultaneously, stabilizing the inversion, overcoming the underdetermined problem and measurement calibration limitations.
Besides producing accurate formation resistivity, the inversion improves image quality in highly resistive and fractured formations, improves consistency among the pads, and helps eliminate “blending” artifacts. The inversion also generates a tool-standoff image detailing the borehole shape and a dielectric-permittivity image, which can be valuable for standalone formation evaluation or joint interpretation with array dielectric measurements.
The inversion algorithm is applied to field data acquired in different conditions, to illustrate its potential of inversion-based workflow to further enhance interpretation of the new OBM imager. The field data examples include various complex cases with blending artifacts, large standoff, and resistive formations. The processed resistivities compare well with standard array-induction responses, and so do the computed dielectric permittivities with those derived from an array-dielectric tool. The standoff image helps characterize fractures, faults, and other natural or drilling-induced events on the borehole surface.
The application of the microelectrical imager (Luthi, 2001) has been limited to the conductive water-based fluids (WBM), conditions favorable for use of the low-frequency galvanic measurement physics. The WBM imager is able to produce high-definition images of 0.2 in. resolution and 80% circumferential coverage in an 8-in. borehole, using 192 buttons distributed over 8 pads. Conventional interpretation of the microelectrical images covers determination of structure, identification of thin beds, classification of heterogeneities, facies classification, identification of the depositional environments, fracture analysis, and use in constraining the reservoir model (Hansen and Fett, 2000; Slatt and Davis, 2010).