BR112018072967B1 - METHOD TO CHARACTERIZE A GEOLOGICAL FORMATION CROSSED BY AN EXPLORATION WELL - Google Patents

Abstract

Inversion-based workflows are provided for real-time interpretation of electromagnetic (EM) observation and prediction measurements. The profile of an observation zone is determined by interpreting the EM measurements of an observation zone. The observation zone profile characterizes formation dip, as well as vertical resistivity or resistivity anisotropy of one or more formation layers of the observation zone. The profile of a prediction zone is determined by interpreting the EM measurements of the prediction zone. The forecast zone profile characterizes the dip of formation as well as the horizontal resistivity, vertical resistivity, or anisotropy of one or more formation layers of the forecast zone. Workflows may also involve interpreting observation resistivity measurements to aid in observation zone characterization.

Description

[0001] This disclosure claims priority of US Provisional Patent Application No. 62/333,232, filed May 8, 2016, incorporated herein by reference in its entirety.

FUNDAMENTALS OF THE INVENTION 1. Field

[0002] This disclosure generally refers to profiling and directional drilling of oil and gas wells. More specifically, techniques are disclosed for improving well placement using data from continuous logging (LWD) tools.

2. State of the Art

[0003] Exploration wells are drilled through land formations to extract oil or other hydrocarbon-based resources. Data on downhole conditions and drill rig movement can be collected during the drilling process. By collecting and processing this information during the drilling process, the driller can modify or correct key steps in the operation to optimize well placement. Schemes for collecting data on downhole conditions and drill rig movement during a drilling operation are commonly referred to as measurement while drilling (“MWD”). Similar techniques, focusing more on measuring formation parameters during the drilling process, are commonly referred to as continuous logging (“LWD”). However, the terms MWD and LWD are often used interchangeably, and the use of either term in this disclosure will be understood to include both the collection of formation and borehole information, as well as data on drill rig movement and placement.

[0004] Electromagnetic (EM) LWD tools have been employed in oilfield applications for many years. These types of profiling tools or instruments typically include an elongated stand equipped with antennas that can be operated as sources (transmitters) or sensors (receivers). The antennas of these tools are usually formed as meshes or coils of conductive wires and can be configured to have axial (parallel to the tool axis), transverse (perpendicular to the tool axis), or inclined (neither parallel nor perpendicular to the tool axis) dipole moments. ). In operation, one or more transmitter antennas are energized by an alternating current to emit EM energy to the adjacent formation or formations. As used herein, "formation" can refer to a single layer or can include multiple layers. The emitted energy interacts with the borehole and formation to produce signals that are detected and measured by one or more receiving antennas. These detected signals reflect the interactions with the formation. By processing the detected signal data, a record or profile of formation and/or borehole properties is determined.

[0005] The introduction of directional electromagnetic (EM) logging tools has revolutionized exploration well placement. Specifically, directional EM measurements enable proactive geo-steering and geo-braking, which allows control of the trajectory of the drilled exploration well so that it is kept within the reservoir of interest or “reservoir gap”. As a result, the trajectory of the drilled exploration well can be directed and stopped along a path defined by observed reservoir boundaries and fluid contacts rather than by preconceived geometries. Data from the LWD or MWD tools is used for real-time prediction and visualization of the layer structure of the formation around the tool or drill string. This real-time visualization allows operators to control the direction of exploration well drilling operations in order to place or “backfill” the well into a specific section and location of a reservoir. Optimizing exploration well placement results in increased production by optimizing well landing, minimizing gas or water penetration, reducing deviations and managing drilling risk through better control of exploration well placement.

[0006] LWD directional EM tools are suitable for these applications due to their relatively large lateral depth of investigation into adjacent formation and azimuthal sensitivity. Several LWD directional EM tools are commercially available, such as Schlumberger's PeriScope™ and GeoSphere™ deep imaging LWD tools, which incorporate multiple axial, tilted, and transverse antennas in the drill command in the PeriScope™ LWD tool case and multiple tilted antennas in the PeriScope™ LWD tool case. of the LWD GeoSphere™ tool. Non-axial antennas generate directional measurements that are used to determine distances to near boundaries and the azimuthal orientation of formation boundaries in various types of mud. These measurements are transmitted to the well surface during drilling, processed in real time through inversion and the results displayed in a graphical (visualized) interface to provide information on distance to boundaries, formation resistivity and orientation. These EM LWD tools also include short antenna spacings with shallow depth of investigation (DOI), used to characterize formation layers close to the tool, and longer antenna spacings with deep DOIs, used to image and characterize formation layer properties. further away from the tool.

[0007] US Patent No. 6,819,110 describes an exemplary LWD directional EM tool that employs a combination of axial and transverse magnetic dipole induction couplings for two or three antenna arrays to maximize the forward contribution of the drill based in spatial sensibilities.

[0008] US Patent 8,736,270 describes a workflow for determining a drill head formation property using an LWD directional EM tool that involves inverting an observation measurement followed by inverting a prediction measurement. The inversion of the predictive measurement determines the formation feature ahead of the bit.

SUMMARY

[0009] This summary is provided to present a selection of concepts that are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in narrowing the scope of the claimed subject matter.

[0010] The present disclosure provides methods (and corresponding systems) for drilling an exploration well traversing a geological formation using one or more LWD tools along with modeling and formation inversion based on measurements from one or more LWD tools.

[0011] In one or more embodiments, the exploration well is drilled with a propagation-type resistivity tool near a drill bit and a directional EM tool with an inclined transmitter near the drill bit. A formation observation zone can be characterized by a multi-step observation processing workflow that includes i) inversion operations (referred to as "first inversion observation processing" in this document) that interpret resistivity measurements from the formation tool propagation-type resistivity to determine a horizontal Rh resistivity profile for one or more layers forming the observation zone, and ii) inversion operations (referred to as "second inversion observation processing" in this document) that interpret the tool's EM measurements Directional EM to determine formation dip and a profile of vertical Rv resistivity or resistivity anisotropy of one or more formation layers in the observation zone. The horizontal resistivity profile Rh of one or more formation layers of the observation zone as determined from the first processing of inversion observation, the formation dip and the vertical resistivity profile Rv or resistivity anisotropy of one or more layers of observation zone formation as determined from the second inversion observation processing (and possibly data characterizing other characteristics or properties of one or more layers of observation zone formation as determined by the first and/or second inversion observation processing ) may form part of a more complete profile of the observation zone.

[0012] The resulting profile of the observation zone, determined from the multi-step observation processing workflow, can be used to update a reservoir model or part of it that corresponds to the observation zone for visualization and reservoir analysis and/or reservoir characterization. Additionally, the observation zone profile, as determined by the multi-step observation processing workflow, can be used to control the operation (eg geo-targeting) of the bit.

[0013] The multi-step observation processing workflow inversions may employ one or more formation models representing a number of substantially parallel transversely isotropic (TI) formation layers. Such formation models may include values representing horizontal resistivity Rh, vertical resistivity Rv, or resistivity anisotropy (such as an Rv/Rh ratio) and thickness for a number of formation layers traversed by the tool(s), as well as formation dip (direction and magnitude) and position and orientation of tool(s) within formation layers. The formation model may also include other variables and/or parameters (not shown) that characterize the formation or exploration well environment. For example, the formation model may include variables and/or parameters that represent formation geological structures (such as faults or other heterogeneities), formation properties (such as other rock or fluid properties), and borehole effects (such as eccentricity of tool or mud filtrate invasion).

[0014] In embodiments, the formation model used for the second inversion observation processing can be initialized according to the values of the formation model that result from the first inversion observation processing. There may be a large difference in the depth of investigation between the propagation type resistivity tool and the directional EM tool such that the directional EM tool measurements do not resolve the resistivity anisotropy in the resolution of the propagation resistivity tool measurements. In this case, the formation model resulting from the first inversion observation processing can be scaled up (coarse) to account for the lower resolution of the measurements (particularly close spacing measurements) of the Directional EM tool, and the second observation processing of inversion associated with it. The extended training model can then be used as the initial training model (or assumption) for the second inversion observation processing. Certain parts of the formation model used for the second inversion observation processing (such as the layers, layer thicknesses and horizontal resistivity of the Rh values of this formation model) can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation.

[0015] In embodiments, the second inversion observation processing may involve an inversion (called a "resistivity anisotropy inversion") that interprets EM measurements from the directional EM tool to determine a vertical resistivity profile Rv or resistivity anisotropy of one or more observation zone formation layers. The extended formation model derived from the results of the first inversion observation run can be used as an initial formation model for resistivity anisotropy inversion. In addition, initial values for the vertical resistivity Rv values or anisotropy resistivity of the formation layers of the formation model can be based on the vertical resistivity Rv values or anisotropy resistivity of the formation layers of the formation model as determined by the second run reversal to an earlier observation zone of the formation. Certain parts of the formation model (such as the layers, layer thicknesses and horizontal resistivity Rh) for the resistivity anisotropy inversion can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation for the same zone training observation.

[0016] The second processing of an inversion observation may also involve an inversion (called a "dip inversion") that interprets the EM measurements of the directional EM tool to determine the dip of the formation of the observation zone. Dip inversion can follow resistivity anisotropy inversion. The extended formation model derived from the results of the first inversion observation run can be used as an initial formation model for resistivity anisotropy inversion. In addition, the initial dip of the formation model's formation layers may be based on the dip of the formation model's formation layers, as determined by the second inversion observation processing to an earlier observation zone of the formation. Certain parts of the formation model (such as the layers, layer thicknesses and horizontal resistivity Rh) for dip inversion can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation for the same observation zone of training.

[0017] The second processing of inversion observation may also involve an inversion (called "dip inversion and resistivity anisotropy") that interprets the measurements of the directional EM tool to determine the formation dip and a vertical resistivity profile Rv or resistivity anisotropy of one or more formation layers of the observation zone. Dip inversion and resistivity anisotropy can refine the formation dip and vertical Rv resistivity profile or observation zone resistivity anisotropy as determined from the separate directional anisotropy inversion and directional dip inversion. The formation model derived from the resistivity anisotropy inversion and dip inversion results can be used as an initial formation model for dip inversion and resistivity anisotropy. Certain parts of the formation model (such as the layers, layer thicknesses and horizontal Rh resistivity) used for dip inversion and resistivity anisotropy can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation to the same formation observation zone.

[0018] The formation model resulting from the second inversion observation processing can be used in the inversion prediction processing that characterizes a formation prediction zone. Inversion prediction processing can interpret EM measurements from the directional EM tool to determine the formation dip and a profile of the horizontal resistivity Rh, the vertical resistivity Rv, or the resistivity anisotropy and bed boundaries of one or more zone formation layers training forecast. The formation dip and horizontal resistivity profile Rh, and the vertical resistivity Rv or resistivity anisotropy of one or more formation layers of the prediction zone, as determined by inversion prediction processing, can provide a total resistivity profile of the training.

[0019] The formation dip and the resistivity profile of the prediction zone as determined by the inversion prediction processing can be used to update a reservoir model or part of it that corresponds to the prediction zone for visualization and analysis of reservoir and/ or reservoir characterization. In addition, the formation dip and predictive zone resistivity profile, as determined by inversion prediction processing, can be used to control the operation (eg, geo-braking) of the bit.

[0020] The EM measurements of the directional EM tool used in inversion prediction processing can be sensitive to the thresholds when the transmitter or receiver antennas cross them. To address the issue of processing efficiency and measurement sensitivity, for a region outside the forecast inversion window (i.e., the rearward observation formation), the formation layer profile of the formation model can be coarser, keeping only the “meaningful” thresholds that contribute to the response. Thus, in the coarse region behind the far receiver, the response sensitivity is very low and the effect of formation thresholds in this region is removed from the inversion prediction processing. Inversion prediction processing can use either a pixel-based or a model-based approach, as described in this document.

[0021] In order to improve the robustness of the workflow, special care can be taken in the initialization of the formation model used for the second inversion prediction processing that characterizes the anisotropic resistivity and the formation dip of the observation zone. Specifically, the formation model can start with a 0° dip (or some predefined value if available from other data) and an isotropic horizontal resistivity profile, since anisotropy and dip information are not available initially. The inversion refinement limits can be relaxed because initial results may not be as reliable due to lack of sensitivity to resistivity anisotropy. However, inversion prediction processing may be impacted initially, but as drilling progresses and new data is acquired, the workflow can automatically correct and improve the observation zone characterization as well as the observation zone characterization. forecast.

[0022] Other workflows for predictive interpretation of EM measurements are described and claimed. Workflows may also involve interpreting observational resistivity measurements to aid in prediction zone characterization.

[0023] In other exemplary workflows, surface resistivity measurements from another tool (for example, triaxial resistivity tool) can be interpreted to determine the resistivity profile (including horizontal resistivity Rh and vertical resistivity Rv or resistivity anisotropy) and formation diving for both the observation zone and the formation prediction zone.

[0024] Other features and aspects of exemplary embodiments of the present invention are described in more detail below with reference to the accompanying figures.

BRIEF DESCRIPTION OF THE FIGURES

[0025] Figure 1 is a schematic diagram of a wellsite and a drilled well incorporating an exemplary continuous logging (LWD) system in accordance with the present disclosure.

[0026] Figure 2A shows the exemplary modules of an LWD tool that is part of the downhole assembly of the LWD system in Figure 1.

[0027] Figure 2B is a schematic illustration of exemplary inclined orientations of transmitter and receiver antennas of a directional EM tool that is part of the downhole assembly of Figure 2A.

[0028] Figure 2C is a schematic illustration of an electromagnetic mutual impedance tensor that can be derived from measurements from the directional EM tool of Figure 2B, including the relationship of the components of the tensor to the orientations of the antennas.

[0029] Figure 2D is a graph summarizing a variety of directional measurements that are generated by combining the various impedance tensor components of Figure 2C, as determined from measurements of transmitter/receiver voltages during tool rotation.

[0030] Figure 2E is a schematic illustration of the exemplary propagation-type resistance tool that forms part of the downhole assembly of Figure 2A.

[0031] Figure 2F is a schematic illustration of the downhole assembly of Figure 2B, showing the tool configuration and the identification of observation, forecast and rear observation formation zones.

[0032] Figure 3 is a schematic diagram of an exemplary training model.

[0033] Figures 4A - 4C, collectively, are a flowchart of an illustrative inversion-based workflow that uses the EM resistivity and LWD directional modules from the downhole assembly of Figure 2A along with a processing workflow multi-step observation processing that characterizes the current observation zone, as well as prediction processing that characterizes the current prediction zone.

[0034] Figure 5 is a flow chart illustrating exemplary operations that may form part of the First Inversion Observation Processing (block 407) of the workflow of Figures 4A-4C.

[0035] Figures 6A and 6B collectively are a flowchart illustrating exemplary operations that may form part of the Second Inversion Observation Processing (block 413) of the workflow of Figures 4A-4C.

[0036] Figure 7 is a graphical representation of an example workflow that uses the LWD modules from the downhole assembly of Figure 2A along with a multi-step observation processing workflow that characterizes the current observation zone and also the precision processing that characterizes the current forecast zone.

[0037] Figure 8 is a graph illustrating how formation dip and resistivity anisotropy affects the estimate of the horizontal resistivity Rh derived from the RCDR apparent resistivity resolved by the First Inversion Prediction Processing (block 407) of Figures 4A-4C and 5.

[0038] Figure 9A is a flowchart illustrating exemplary operations that may form part of the First Inversion Observation Processing (eg, block 407) and used to account for a higher relative formation dip.

[0039] Figure 9B is a flowchart illustrating sample inversion operations that may be part of the Inversion Prediction Processing (eg, block 419) of the workflows described in this document.

[0040] Figure 9C is a flowchart illustrating other exemplary inversion operations that may form part of the Inversion Prediction Processing (eg, block 419) of the workflows described in this document.

[0041] Figure 10 is a flowchart illustrating a single-step inversion workflow characterizing the current observation zone; this workflow can replace operations for the First and Second Processing of Inversion Observation (blocks 405 to 413) of Figures 4A - 4C.

[0042] Figures 11A - 11C, collectively, are a flowchart illustrating a workflow that characterizes the current observation zone and the current forecast zone using only directional EM measurements from the directional EM tool of the downhole set in Figure 2A.

[0043] Figures 12A-12B, 13A-13B, 14A-14B, 15A-15B, 16, 17A-17B, 18, 19A-19B and 20A-20B are graphs that help to visualize the results of the workflows described in this document.

[0044] Figure 21 is a schematic diagram of an exemplary computing system that may be used to implement the computer processing system of Figure 1A.

DETAILED DESCRIPTION OF THE PREFERRED MODALITIES

[0045] Figure 1 illustrates a well site during drilling where the disclosed modeling and inversion methods can be employed to process acquired measurements. The well site can be located on land or at sea. In this exemplary system, a borehole or exploration shaft 11 is formed in a subsurface formation reservoir 30 by directional rotation. A drillstring 12 is suspended within the borehole 11 and has a downhole assembly 151 that includes a bit 105 at its lower end. The surface system includes the platform and a crane assembly 10 positioned over the wellbore 11, with the assembly 10 including a rotary table 16, a kelly 17, a hook 18 and a rotatable injection head 19. The drill string 12 is rotated around the rotary table 16, powered by means not shown, which engages with the kelly 17 at the upper end of the drill string. The drill string 12 is suspended from the hook 18, attached to a movable block (also not shown) and through the kelly 17 and a rotatable injection head 19 which allows rotation of the drill string relative to the hook. As is well known, a top drive system could alternatively be used.

[0046] In this exemplary embodiment, the surface system also includes the drilling fluid or mud 26 stored in a tank 27 formed at the well site. A pump 29 delivers drilling fluid 26 into the drill string 12 through a port in the injection head 19, causing the drilling fluid to flow down through the drill string 12 as indicated by the directional arrow 8. Drilling fluid exits the drill string 12 through ports in the bit 105 and then circulates upward through the annulus region between the outside of the drill string and the borehole wall, as indicated by the directional arrows 9. In this way well known, drilling fluid lubricates bit 105 and carries cuttings from the formation to the surface while being returned to tank 27 for recirculation.

[0047] As is known in the art, sensors can be provided around the wellsite to collect data, preferably in real time, on the operation of the wellsite, as well as the conditions at the wellsite. For example, such surface sensors can be provided to measure parameters such as standpipe pressure, hook load, depth, surface torque, rpm, among others.

[0048] The downhole assembly 151 of the illustrated embodiment includes several continuous logging modules (LWD) that function as individual LWD tools or multiple modules that operate together a part of a single LWD tool, possibly one or more measurement modules during drilling (MWD) and a rotary steering system that controls the direction of the bit. An exemplary BHA 151 is shown in Figure 2A. The LWD modules of the LWD tool can be housed in a special type of drill driver as is known in the art. The LWD tool's LWD modules include capabilities to measure and store directional electromagnetic response data that is sensitive to the resistivity profile of the formation surrounding the BHA 151. The MWD module(s) can also be housed in a special type of control drilling, as is known in the art. The MWD modules include facilities for measuring, processing and storing information characterizing the position and direction of the drillstring 12 and the BHA 151 bit, as well as other drilling measurements such as weight on the bit, torque and shock and/or vibration. As used in this document, the term "module", as applied to LWD and MWD devices, is to be understood either as an individual tool or as a series of multiple tools contained in a single modular device.

[0049] The BHA 151 also includes a downhole telemetry subsystem that communicates data signals and control signals between the components of the BHA 151 (including the LWD tool modules) and a profiling and control unit 200 located on the surface through electronics subsystem 35. The underground telemetry subsystem may employ various telemetry methods, such as wired telemetry methods (for example, drill pipe incorporating telemetry cables or fiber optic cables) and wireless telemetry methods (for example, (e.g., mud pulse telemetry methods, electromagnetic telemetry methods, and acoustic telemetry methods). The downhole telemetry subsystem may also provide electrical power supply signals generated by a surface located power source to supply components of the BHA 151. The BHA 151 may also include a power supply transformer/regulator to transform the electrical power supply signals provided by the surface located power source at appropriate levels suitable for use by the components of the BHA 151. In alternative embodiments, the BHA 151 may include apparatus for generating electrical power for supplying the components of the BHA, such as as a mud turbine generator powered by the flow of drilling fluid. However, other battery and/or power systems may be employed.

[0050] The well site of Figure 1 may also include the logging and control unit located on the surface 4 that interfaces with the computer processing system 203 through data communication links (shown as bidirectional dotted lines with arrows) . A control module 204 (termed "geo-steering control") can interface with the profiling and control unit 200 and with the computer processing system 203 via data communication links (shown as bidirectional dotted lines with arrows ) for geo-steering and geo-braking applications as described in this document. Data communication links between surface located components may utilize wired and/or wireless connection via one or more communication lines. The communication topology between these components located on the surface can be point-to-point, multiple point-to-point or multipoint-to-point. The wired connection(s) can utilize a variety of cable types or wires using a variety of wired protocols (serial, wired Ethernet, fiber channel, etc.). The wireless connection(s) may utilize various wireless protocols (such as IEEE 802.11, Bluetooth, Zigbee, or any non-standard optical or RF communication technology).

[0051] The computer processing system 203 can be configured to perform the inversion methods and workflows as described in this document, which can be used to visualize and control the position and orientation of the BHA 151 during drilling operations. The control module 204 communicates with the profiling and control unit 200 to control the position and orientation of the BHA 151 as determined by the operation of the computer processing system 203.

[0052] Inversion methods (and corresponding systems) generally employ an initial estimate or formation model that includes parameters that represent the geometry and properties (including anisotropic resistivity) of the land formation around BHA 151. Initial values for training model parameters can be derived in several ways, as described in this document. Simulated measurements of the BHA 151 LWD modules can be calculated based on the initial parameters of the training model. Calculating the simulated measurements can involve direct modeling, which applies the initial parameters of the formation model to a set of theoretical equations to generate the simulated measurements. The simulated measurements are then compared with the corresponding real measurements made by the LWD modules of the BHA 151. Differences between the simulated measurements and the corresponding real measurements can be used to adjust the parameters of the training model, and the adjusted training model is used again to calculate simulated measurements from the BHA 151 LWD modules. The simulated measurements based on the adjusted training model are compared with corresponding real measurements made by the BHA 151 LWD modules, and any difference between them is used to re-adjust the parameters of the training model. training. This process is generally repeated until the differences between the simulated measurements of the BHA 151 LWD modules and the corresponding actual measurements made by the BHA 151 LWD modules drop below a preselected threshold. By way of example, commonly owned US Patent No. 6,594,584 describes some modern inversion techniques and is incorporated herein by reference in its entirety.

[0053] The inversion methods (and corresponding systems) disclosed in this document are intended to improve the ability to dynamically control the position and orientation of the BHA 151, so that the BHA 151 drill follows the planned trajectory of the exploration well, if it's practical. In addition, the position and orientation of the BHA 151 can be dynamically controlled in order to remain at an ideal distance relative to the limits and contacts of the reservoir, or to avoid faults in the vicinity or other heterogeneities that are far from the BHA 151 along the planned trajectory of the exploration well during the drilling process. In addition, the position and orientation of the BHA 151 can be dynamically controlled to adjust the exploration well trajectory (geo-steering) and terminate the exploration well (geo-braking) based on detection and characterization of reservoir boundaries and contacts. or flaws or other heterogeneities that are compensated against the BHA 151 during the drilling process. As used in this document, a heterogeneity is a difference in the composition and/or character (such as a difference in porosity and/or fluid saturation) of the rock matrix of the formation that results in an abrupt or significant change in the resistivity of the rock matrix. . For example, a heterogeneity can be a boundary, contact or fault in the rock matrix.

[0054] In addition, the inversion methods (and corresponding systems) disclosed in this document can be used as part of post-drill analysis and model refinement for better reservoir characterization in vertical, deviated, high angle and horizontal wells.

[0055] Referring to Figure 2A, an exemplary BHA 151 incorporates a number of LWD modules that include a transmitter subassembly 210 and two receiver subassemblies 208 and 209 that incorporate a directional EM tool. Transmitter subassembly 210 is placed near drill 212 and includes one or more transmitter antennas. At least one transmitter antenna of the directional EM tool is tilted. Each of the receiver subassemblies 208 and 209 includes one or more receiver antennas configured to detect the EM field in order to make close-spaced (deep DOI) and long-spaced (very deep DOI) directional EM measurements of the formation in the vicinity of the BHA 151.

[0056] In one embodiment, the transmitter subassembly 210 may have a triaxial antenna array, comprising a combination of axial, slanted, and transverse antennas. As used herein, an axial antenna is one whose dipole moment is substantially parallel to the longitudinal axis of the tool. Axial antennas are commonly wrapped around the circumference of a logging tool so that the plane of the antenna is orthogonal to the axis of the tool. Axial antennas produce a radiation pattern equivalent to a dipole along the tool axis (by convention, the z direction). A transverse antenna is one whose dipole moment is substantially perpendicular to the longitudinal axis of the tool. A transverse antenna can include a saddle coil (e.g., as disclosed in commonly owned US Patent Publications 2011/0074427 and 2011/0238312, incorporated by reference in their entirety) or multiple angled coils placed together and generate an equivalent radiation pattern to a dipole perpendicular to the tool axis (by convention, the x or y direction). An inclined antenna is one whose dipole moment is neither parallel nor perpendicular to the longitudinal axis of the tool. Tilted antennas generate a mixed-mode radiation pattern (that is, a radiation pattern in which the dipole moment is neither parallel nor perpendicular to the tool axis).

[0057] The BHA 151 also includes a rotary steerable system and drill 212. The rotary steerable system can be used to dynamically adjust the direction of drilling performed by the drill 212 under commands communicated from the geo-steering control module 204 via the logging and control unit 4 and the telemetry subsystem of the BHA 151. The method used by the rotary steerable system to dynamically adjust the drilling direction can generally fall into two categories, namely “push the bit” or “point the bit”. . Bit push systems use buffers on the outside of the tool that press against the hole, causing the bit to press on the opposite side, causing a change in direction. Drill pointing systems cause the drill to change direction relative to the rest of the tool by bending the main shank past it.

[0058] In one embodiment, the antennas of the receiver subassemblies 208, 209 of the directional EM tool are angled and spaced relative to the angled antenna of the transmitter subassembly 210 along the axis of the BHA 151, as shown in Figure 2B. This tool is commercially available under the name GEOSPHERE™ from Schlumberger Technology Corporation of Sugar Land, Texas. Each of the tilted antennas has a magnetic dipole moment that is neither aligned with the tool axis (for example, the central axis of the BHA 151) nor orthogonal to the tool axis. Transmitter sub-assembly 210 and receiver sub-assemblies 208 and 209 may contain a set of orthogonal magnetic bipolar antennas, with axial and transverse coils which may also be combined with tilted antennas.

[0059] In embodiments, the antenna of the transmitter subassembly 210 and the receiver subassemblies 208, 209 of the directional EM tool can be configurable with up to three spacings (with two shown in Figures 2A and 2B).

[0060] It will be understood, however, that the embodiments disclosed in this document are not limited to any particular electromagnetic profiling tool configuration, and that the directional EM tool depicted in Figures 2A and 2B is merely an example of a tool suitable directional EM profiling.

[0061] For example, in other embodiments, the receiver subassemblies 208 and 209 can employ a combination of axial, inclined and transverse antennas, capable of detecting all components of the EM fields by measuring the magnetic field in different directions. Note that electromagnetic measurements determined by axial transmitter electromagnetic signals received by axial receiver antennas may be called conventional or non-directional measurements, while electromagnetic measurements determined by electromagnetic signals from transverse or tilted transmitter or receiver antennas may be called directional measurements. .

[0062] In yet other embodiments, the directional EM tool of the BHA 151 may include multiple transmitter subassemblies and multiple receiver subassemblies axially spaced along the tool body, as disclosed in commonly owned US Patent Publication 2017/0075021, incorporated herein by reference in its entirety. This tool is commercially available under the name PERISCOPE™ from Schlumberger Technology Corporation of Sugar Land, Texas. The multiple transmitter subassemblies may include one or more axial antennas and one or more transverse antennas spaced along the axis of the tool. The receiver subassemblies can include multiple receivers with axial antennas that are spaced along the tool axis between the transmitter subassemblies. Such receivers can be used to obtain conventional non-directional type propagation resistivity measurements. Receiver subassemblies can also include multiple receivers with angled antennas that are spaced along the tool axis around the transmitter subassemblies. Such a directional arrangement (including inclined and/or transverse receiving antennas) produces a preferential sensitivity in an azimuth orientation of the tool that best allows the detection of bed boundaries and evaluation of distance and orientation and other characteristics of underground formations to be identified and located. Furthermore, by providing both the axial transmitter and axial receiver pairs, as well as the axial transmitter and inclined receiver pairs, the tool is capable of performing non-directional and directional electromagnetic measurements. Furthermore, the tool is capable of providing symmetric and anti-symmetric measurements (up and down measurements) with the same antenna spacings. As an example, in a given embodiment, the tool may be capable of taking measurements using transmitter-receiver pairs spaced 22, 34, 84, and 96 inches apart.

[0063] In still other embodiments, the tool configuration may have multiple transmitter sets and more than two receiver subsets. Although antennas have been described in terms of magnetic dipole antennas, antennas may also comprise electrical dipole antennas. For example, magnetic dipole antennas can use coils, while electrical dipole antennas can use electrodes and/or toroids.

[0064] The antennas of the receiver subsets spaced along the directional EM tool can provide phase voltage and amplitude measurements over a wide range of frequencies below 100 kHz. For each transmitter/receiver spacing and frequency, the phase and amplitude response measurement is a linear combination of elementary electromagnetic couplings, allowing the determination of nine components of a 3D mutual impedance tensor between transmitter and receiver during tool rotation as shown in Figure 2C. In this example, the Z axis is aligned with the tool axis, the X and Y axes are perpendicular to each other and to the Z axis. The first letter of the tensor component corresponds to the direction of the dipole moment of the receiver, and the second letter corresponds to the direction the dipole moment of the transmitter. In this configuration, measurements can be made of the amplitude and phase of the voltage at a receiving antenna, induced by a signal transmitted from the transmitting antenna. These receiver amplitude and phase measurements are proportional to the magnitudes and depend on the directions of the transmitter and receiver dipole moments. 3D mutual impedance tensor couplings depend on the frequency of the transmitted signal, transmitter/receiver spacing, tool inclination, and forming parameters, including anisotropic formation resistivity (i.e., horizontal and vertical resistivity) and formation geometry (i.e., , position and orientation of boundaries, contacts, faults and other heterogeneities). Exemplary methods for determining 3D mutual impedance tensor components from voltage measurements of the LWD modules of the BHA 151 are described in detail in US Patent No. 8,736,270, commonly assigned to the assignee of the present invention and incorporated herein by reference in its entirety.

[0065] Furthermore, individual 3D mutual impedance tensor components can be used, or various combinations of the 3D mutual impedance tensor components can be combined, to obtain a set of directional measurements that is used to infer formation and structure properties. Figure 2D summarizes a variety of directional measurements that can be generated by combining various tensor components, as determined by transmitter/receiver measurements during tool rotation, and the corresponding spatial sensitivities in the XY plane. Harmonic resistivity (UHR) measurements are sensitive to formation resistivity and resistivity anisotropy and are not azimuthally sensitive with respect to the tool axis. Symmetrized directional measurements (USD) are sensitive to conductivity thresholds and contrast if the tool does not cross thresholds and are largely insensitive to relative dip and resistivity anisotropy of the formation. Measurements are created using elemental couplings that are insensitive to the tool azimuth and a coupling that is cosθ sensitive with respect to the tool rotation azimuth θ, so these measurements are also called a first harmonic measurement. In the XY plane, the spatial sensitivity of the measurements has two symmetrical lobes of different polarity, positive or negative, depending on the presence of conductivity above or below the tool (in the X direction). Antisymmetric directional measurements (UAD) are sensitive to relative dip and formation resistivity anisotropy and are largely boundary insensitive unless there is a boundary between the transmitter and receiver sub. This class of measurements is composed using the same components as the USD measurements, including couplings that have azimuthal sensitivity to the tool axis. Harmonic anisotropy (UHA) measurements are sensitive to resistivity anisotropy in non-vertical wells and formation boundaries in HA/HZ wells. These measurements are made using transverse couplings with azimuthal sensitivity to cos2θ with respect to the tool axis for the azimuth of tool rotation θ, and are also called second harmonic measurements, with sensitivity in the XY plane having four polarized lobes distributed around the tool axis. The 3D indicator measurements (U3DF) are sensitive to lateral inhomogeneities in the formation with respect to layers (such as faults at an azimuth different from the layered azimuth) that are offset from the tool in the Y direction (assuming the boundaries are in the X direction).

[0066] Formation dip (also called "dip" in this document) characterizes the inclined orientation of the bedding (or bedding) of the formation relative to a horizontal plane. Formation dip is commonly represented by a dip angle and a dip direction. The dip angle represents the steepest angle of descent of the inclined formation layer in relation to the horizontal plane. The dip direction is the azimuth of the dip direction as projected onto the horizontal plane. Formation dip can also be characterized by an apparent azimuthal orientation of the formation layering of the formation relative to the “up” orientation of the tool datum. This angle (referred to as the Apparent Azimuth) refers to the azimuthal offset of the X' normal axis of the formation bedding relative to the reference tool orientation defined by the X axis.

[0067] In embodiments, the LWD modules of the BHA 151 may also include a propagation-type resistivity tool 211 that transmits a high-frequency electromagnetic wave and measures the differences in attenuation and phase shift between induced voltages in two receivers. Phase shift and attenuation can be turned into apparent resistivity measurements, where phase shift apparent resistivities are typically shallower than attenuation apparent resistivity measurements. In one embodiment, the propagation-type resistivity tool 211 can be realized by the dual resistivity compensated (CDR) tool, which has two transmitters arranged symmetrically around two receivers, as shown in Figure 2E. Each transmitter alternately transmits electromagnetic waves of 400 kHz and 2 MHz. A propagation measurement is made by taking the difference between the phases (phase shift) and amplitudes (attenuation) of the voltages recorded at the two receivers. The attenuation increases as the conductivity increases, while the wavelength decreases as the conductivity increases. Thus, the two measurements increase monotonically with formation conductivity and can be used to generate resistivity records. The phase shift and attenuation measurements generated by the upper transmitter between the two receivers, and the lower transmitter between the two receivers, are averaged to mirror the tool's response. This average is known as borehole compensation because it also reduces the effect of borehole roughness.

[0068] Figure 2F is a schematic view of an illustrative embodiment of the BHA 151 of Figure 2A. BHA 151 includes bit 212 configured to drill a well hole in the formation as bit 212 traverses the formation. The BHA 151 also includes an inclined antenna transmitter subassembly 210 and receiver subassemblies 208 and 209 that incorporate the directional EM tool. BHA 151 also includes propagation tool 211.

[0069] According to some examples, the BHA 151 has a modular design. In some examples, the modular design includes a sub transmitter and a sub receiver. In some instances, the BHA 151 may use the same or analogous configurations and hardware architecture as described in US Patent No. 7,755,361, which is incorporated herein by reference in its entirety. In some examples, the setup differs from US Patent No. 7,755,361 in that the measurements are closer to the drill, with the transmitter tilted immediately behind the drill (or very close) and triaxial receiver assemblies arranged behind, for example, of a rotary steerable system (RSS).

[0070] According to some examples, the transmitter subassembly 210 is close (e.g., as close as reasonably possible) to the drill 212. In some examples, the transmitter subassembly 210 is less than 10 feet from the drill 212, although others examples may have transmitter subassembly 110 at a distance of 10 feet or greater from drill 212. In some examples, transmitter subassembly 210 is 6 feet or less from drill 212.

[0071] The receiver subassemblies 208 and 209 can be provided at respective distances L1 and L2 behind the transmitter subassembly 210, as shown. In some examples, the L1 distance is greater than 20 feet, but in other examples, the L1 distance is less than 20 feet.

[0072] In some examples, the distance L1 is between 20 feet and 40 feet. In some examples, the L1 distance is between 25 feet and 35 feet, for example, 30 feet.

[0073] In some examples, the L2 distance is between 60 feet and 80 feet, although in other examples the L2 distance may be outside this range.

[0074] The exemplary methods, workflows, and systems described in this document may use arbitrary spacing and number of receiver subsets. Thus, although a number of receiver and transmitter subassemblies may be described at particular distances from each other, it should be understood that any suitable number, including one, may be provided to the respective components and at any suitable spacing.

[0075] The propagation tool 211 is an LCDR distance behind the transmitter subassembly 210, as shown. In some examples, the LCDR distance is less than 10 feet, while in other examples the LCDR distance can be 10 feet or greater. In some examples, the LCDR distance is 7 feet or less.

[0076] The methods, workflows and example systems described in this document may use arbitrary frequencies and distribution of sensors.

[0077] Regardless of the number of receivers used by the BHA 151, the portion of the geological formation that falls within the sensitivity range of the BHA 151 can be logically divided into three sections by depth as shown in Figure 2F as follows: 1. ) a prediction zone (or section) 241 of formation 30 - this is the zone (or section) of formation 30 that is in the local vicinity of the BHA 151 and ahead (with respect to the direction of displacement of the bit during drilling) of the propagation-type resistivity 211; predictive zone 241 may extend ahead of transmitter subassembly 210 of the directional EM tool and may extend ahead of drill 212; 2.) an observation zone (or section) 242 of the formation 30 - this is the zone (or section) of the formation 30 that is in the local vicinity of the BHA 151 and between the propagation-type resistivity tool 211 and the receiver subassembly 209 furthest from the directional EM tool (taking into account the length of the forecast inversion window 235); and 3.) a rear observation zone (or section) 243 of the formation 30 - this is the zone (or section) of the formation 30 that is in the local vicinity of the BHA 151 and behind the farthest receiving subassembly 209 of the directional EM tool ( with respect to the direction of travel of the bit during drilling).

[0078] In one or more embodiments, the observation zone 242 may be characterized by a multi-step observation processing workflow that includes i) inversion operations (referred to as "first inversion observation processing" in this document) which interpret the surface resistivity measurements from the spread-type resistivity tool 211 of the BHA 151 (which are sensitive to variation in the surface resistivity of the observation zone 242) to determine a horizontal resistivity profile Rh for one or more zone-forming layers observation zone 242 and ii) inversion operations (referred to as "second inversion observation processing" in this document) that interpret BHA directional EM tool measurements 151 (which are sensitive to anisotropic resistivity of the formation in observation zone 242) to determine the formation dip and a profile of the vertical resistivity Rv or resistivity anisotropy of one or more formation layers of observation zone 242. The horizontal resistivity Rh of a formation layer is the resistivity (preferably in Ohm-m or Qm) parallel to the bedding plane of the formation layer. The vertical resistivity Rv of a formation layer is the resistivity (preferably in Ohmm-m or Qm) in the direction perpendicular to the bedding plane of the formation layer. The electrical anisotropy or resistivity anisotropy of a formation layer is the ratio between the vertical resistivity Rv and the horizontal resistivity Rh of the formation layer on the resistivity measurement scale. The horizontal resistivity profile Rh of one or more formation layers of observation zone 242 as determined from the first inversion observation run, the formation dip, and the vertical resistivity profile Rv or resistivity anisotropy of one or more layers observation zone formation layer 242 as determined from the second inversion observation run (and possibly data characterizing other characteristics or properties of one or more observation zone formation layers 242 as determined by the first and/or second observation run). inversion observation) may form part of a more complete profile of observation zone 242.

[0079] The profile resulting from the observation zone 242, determined from the multi-step observation processing, can be used to update a reservoir model or part thereof that corresponds to the observation zone 242 for visualization and analysis of the reservoir and/ or reservoir characterization. Furthermore, the profile of the observation zone 242, as determined from the multi-step observation processing, can be used to control the operation (eg, geo-targeting) of the bit.

[0080] The inversions of multi-step observation processing may employ one or more formation models that represent a number of substantially planar and parallel transversely isotropic (TI) formation layers. Such formation models may include values representing horizontal resistivity Rh, vertical resistivity Rv, or anisotropy of resistivity and thickness for a number of formation layers traversed by the BHA 151, as well as formation dip (direction and magnitude) and position and orientation of the BHA 151 within the formation layers. An example of an illustrative training model is illustrated in Figure 3, which includes four intermediate layers between the upper and lower limits. The upper boundary has a horizontal resistivity Rh of 3Qm and a vertical resistivity Rv of 10Qm and is assumed to be infinitely thick. The lower boundary has a horizontal resistivity Rh of 2Qm and a vertical resistivity Rv of 5Qm and is assumed to be infinitely thick. The first intermediate layer below the upper limit has a horizontal resistivity Rh of 10Qm and a vertical resistivity Rv of 20Qm and is 2 meters thick. The second intermediate layer below the first intermediate layer has a horizontal resistivity Rh of 30Qm and a vertical resistivity Rv of 40Qm and is 3 meters thick. The third intermediate layer below the second intermediate layer has a horizontal resistivity Rh of 100Qm and a vertical resistivity Rv of 100Qm and is 15 meters thick. The fourth intermediate layer below the third intermediate layer and above the lower limit has a horizontal resistivity Rh of 10Qm and a vertical resistivity Rv of 20Qm and is 13 meters thick. The formation model can also include variables that represent the formation dip angle as shown (which assumes the formation dip is slowly changing) as well as the position/orientation of the tool within the layer structure. The formation model may also include other variables and/or parameters (not shown) that characterize the formation or exploration well environment. For example, the formation model may include variables and/or parameters that represent formation geological structures (such as faults or other heterogeneities), formation properties (such as other rock or fluid properties), and borehole effects (such as eccentricity of tool or mud filtrate invasion).

[0081] In embodiments, the formation model used for the second inversion observation processing can be initialized according to the values of the formation model that result from the first inversion observation processing. There may be a large difference in the depth of investigation between the 211 propagation-type resistivity tool and the 151 BHA directional EM tool such that the directional EM tool's close spacing measurements do not resolve the resistivity anisotropy to a resolution of the measurements of resistivity of the propagation-type resistivity tool 211. In this case, the formation model resulting from the first inversion observation processing can be the scaled-up (coarse) one to account for the lower resolution of the measurements of the directional EM tool and the second processing of inversion observation associated with it. Such an extension may involve grouping together multiple formation layers and assigning them the same value for horizontal resistivity Rh and vertical resistivity Rv or anisotropic resistivity derived from individual layer resistivities. The extended training model can then be used as the initial training model (or assumption) for the second inversion observation processing. Certain parts of the formation model used for the second inversion observation processing (such as the layers, layer thicknesses and horizontal resistivity of the Rh values of this formation model) can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation.

[0082] In modality(s), the second inversion observation processing may involve an inversion (called "resistivity anisotropy inversion") that interprets the BHA 151 directional EM tool measurements (which are sensitive to changes in resistivity directional anisotropy of the observation zone 242) to determine a profile of the vertical resistivity Rv or resistivity anisotropy of one or more formation layers of the observation zone 242. The extended formation model derived from the results of the first inversion observation processing can be used as an initial formation model for the inversion of directional anisotropy. In addition, initial values for the vertical resistivity Rv values or anisotropy resistivity of the formation layers of the formation model can be based on the vertical resistivity Rv values or anisotropy of the formation layers of the formation model as determined by the second inversion processing to an earlier observation zone of the formation. Certain parts of the formation model (such as the layers, layer thicknesses and horizontal resistivity Rh) for the resistivity anisotropy inversion can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation for the same zone training observation.

[0083] The second inversion observation processing may also involve an inversion (called "dip inversion") that interprets the measurements of the directional EM tool of the BHA 151 (which are sensitive to variation in the formation dip of the observation zone 242 ) to determine the forming dip of observation zone 242. Dip inversion may follow resistivity anisotropy inversion. The extended formation model derived from the results of the first inversion observation run can be used as an initial formation model for resistivity anisotropy inversion. In addition, the initial dip of the formation model's formation layers may be based on the formation dip of the formation model's formation layers, as determined by the second inversion observation processing to an earlier observation zone of the formation. Certain parts of the formation model (such as the layers, layer thicknesses and horizontal resistivity Rh) for dip inversion can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation for the same observation zone of training.

[0084] The second inversion observation processing may also involve an inversion (called "dip inversion and resistivity anisotropy") that interprets the measurements of the directional EM tool of the BHA 151 (which are sensitive to variations in formation dip and directional anisotropic resistivity of observation zone 242) to determine a formation dip and vertical resistivity profile Rv or resistivity anisotropy of one or more formation layers of observation zone 242. Dip inversion and resistivity anisotropy can refine the formation dip and the vertical resistivity Rv or resistivity anisotropy profile of observation zone 242 as determined from the separate resistivity anisotropy inversion and dip inversion. The results of anisotropy inversion and dip inversion can be used as an initial training model for dip inversion and anisotropy. Certain parts of the formation model (such as the layers, layer thicknesses and horizontal Rh resistivity) used for dip inversion and resistivity anisotropy can be fixed to the corresponding values of the formation model that result from the first processing of inversion observation to the same formation observation zone.

[0085] The formation model that results from the second reversal observation processing (or part(s) thereof) can be used in the prediction reversal processing that characterizes the prediction zone 241. The reversal observation processing can interpret certain BHA directional EM tool measurements 151 (which are sensitive to the horizontal resistivity and directional anisotropic resistivity and formation dip of the prediction zone 241) to determine the formation dip and a profile of the horizontal resistivity Rh, vertical resistivity Rv or anisotropic resistivity and bed boundaries of one or more layers of formation of the prediction zone 241. Directional EM tool measurements may be sensitive to boundaries as transmitter or receiver antennas cross them. To address this issue, in the region behind the last receiver outside the forecast inversion window (i.e., the rear observation zone 243), the profile of the formation model's formation layers can be thicker, keeping only the boundaries " significant” that contribute to the response. The resistivity profile is therefore segmented with a high threshold behind the last receiver position. Thus, in the region behind the last receiver position, the response sensitivity is very low and the effect of formation thresholds in this region is removed from the prediction inversion processing. Prediction inversion processing can use either a pixel-based or a model-based approach, as described in this document. The formation dip and horizontal resistivity profile Rh and vertical resistivity Rv or anisotropic resistivity of one or more formation layers of prediction zone 241 as determined by inversion prediction processing may provide a total resistivity profile of prediction zone 241 , as well as the position of the bed boundaries of prediction zone 241.

[0086] The resistivity profile of the prediction zone 241 as determined by the inversion prediction processing can be used to update a reservoir model or part thereof that corresponds to the prediction zone 241 for visualization and analysis of reservoir and/or characterization of reservoir. In addition, the resistivity profile of prediction zone 241, as determined by inversion prediction processing, can be used to control the operation (eg, geo-braking) of the bit.

[0087] In order to improve the robustness, special care can be taken in the initialization of the formation model used for the second inversion prediction processing that characterizes the anisotropic resistivity and the formation dip of the observation zone 242. Specifically, the model The formation method uses an initial dip of 0° (or some predefined value if available from other data) and an isotropic resistivity profile, since anisotropy and dip information is not available initially. The inversion refinement limits can be relaxed because initial results may not be as reliable due to lack of sensitivity to anisotropy. However, the prediction inversion processing may be affected initially, but as drilling progresses and new data is acquired, the workflow can automatically correct and improve the observation zone 242 characterization, as well as the observation zone characterization. prediction 241.

[0088] Figures 4A - 4C, collectively, are a flowchart of an illustrative formation modeling and inversion workflow that uses the BHA 151 LWD modules of Figure 2A along with the multi-step observation processing workflow that characterizes the current observation zone 242 (which is the observation zone 242 for the current current location of the BHA 151) as well as the forecast processing that characterizes the current forecast zone 241 (which is the forecast zone 241 for the current reversing station or BHA location 151) for real-time interpretation during drilling as the bit moves forward. It is assumed that a previous prediction zone (which is a prediction zone 241 for a previous inversion station or location of the BHA 151) can match (or match) the current observation zone 242 based on the movement of the BHA 151 through of formation 30 during drilling.

[0089] In block 401, surface resistivity measurements (for example, phase shift measurements) of the current observation zone 242 and the current rear observation zone 243 (which is the rear observation zone for the current inversion station of current or location of the BHA 151) are acquired and processed by operating the propagation-type resistivity tool 211 of the BHA 151 during drilling as the bit moves forward. It is assumed that a forward observation zone (which is the observation zone 242 for an inversion station or previous location of the BHA 151) can match (or match) the current aft observation zone 243 based on the movement of the BHA 151 through formation 30 during drilling.

[0090] In block 403, short-spaced (for example, deep) and long-spaced (for example, very deep) EM measurements at multiple frequencies to characterize the profile of the current observation zone 242, the current rear observation zone 243 and the current prediction zone 241 is acquired and processed by operating the BHA directional EM tool 151 during drilling as the bit moves forward. Examples of such close-spaced and long-spaced EM measurements are discussed above in relation to Figure 2D. Note that lower frequencies and long spacing EM measurements can be used to increase the depth of investigation of the directional EM tool and detect targets before the drill.

[0091] At block 405, a first formation model may be initialized for use in First Inversion Observation Processing (block 407). An example of a suitable first-training model is described above and illustrated in Figure 3.

[0092] In certain cases (such as when the results of observation and prediction processing (block 419) for a corresponding early anticipation zone are not available), the first formation model can be initialized with values of Rh resistivity and layer limits determined from applying the log squared resistivity measurements of the current observation zone 242 and the current rear observation zone 241 of block 401. In one example, the application of the log squared is based on certain phase shift measurements of the propagation-type resistivity tool 211 that correspond to a long segment of the formation that extends to the maximum backward sensitivity of the propagation-type resistivity tool 211 plus the relative Tx position of the farthest EM measurement of the directional EM tool that is used for the inversion of resistivity anisotropy. This ensures that the log squared application takes into account the formation's current observation and rear observation zones. According to some examples, applying logarithm squared involves interpolating these phase shift measurements into this long segment using braces where the inflection points can be used as threshold indicators. The maximum or minimum of the phase shift measurements on this long segment can be used to calculate the resistivity of the corresponding layer. If there is no extremum or if there are multiple extremums, the average value can be obtained. Only tipping points with a significance number above a given threshold can be accepted according to some examples. The threshold of significance can be determined based on the measurement noise level. In the end, the phase shift measurements can be transformed into apparent resistivity, which can be equated to the horizontal Rh resistivity in low deviation wells. In general, the log squared process can be a sufficiently accurate estimate of formation resistivity, but there are several exceptions. First, the threshold positions can be wrong if the resistivity contrast is high. In this case, the application of logarithm squared tends to detect multiple inflection points that do not coincide with the true position of the threshold. Second, if the formation has a progressively increasing or decreasing resistivity profile, an inflection point may not be detected for each Rh step. Third, for thin layers with high contrast, apparent resistivity does not read true horizontal Rh resistivity, and apparent resistivity can be affected by apparent dip and anisotropy, leading to overestimation of Rh. In this case, the correction can be useful.

[0093] In other circumstances (such as when inversion-based workflow results from a corresponding previous inversion station are available), the first formation model can be initialized with formation dip as well as values of horizontal resistivity Rh, resistivity vertical Rv or anisotropic resistivity and layer boundaries obtained from the prediction processing (block 419) of the corresponding previous prediction zone.

[0094] In both cases, the initialization of the first formation model can also be based on formation dip, as well as values of horizontal resistivity Rh, vertical resistivity Rv or resistivity anisotropy and layer limits obtained from the processing of observation (block 413) of a previous observation zone.

[0095] In block 407, the first inversion observation processing is performed, which involves inverting the shallow resistivity measurements of the values of the current observation zone 242 by the propagation-type resistivity tool 211 (block 401) that are sensitive to the change in resistivity of the current observation zone 242. The inversion uses the first formation model and solves for the horizontal resistivity values Rh and layer boundaries of the current observation zone 242. The formation dip values and vertical resistivity Rv or resistivity anisotropy of the first formation model remain fixed in this inversion. The resolved first formation model provides a first resistivity profile of the current observation zone 242, which can provide a sufficiently detailed profile of the horizontal resistivity Rh of the formation layer(s) of the current observation zone 242 in most circumstances where vertical and low diversion wells.

[0096] In block 409, the magnification (thickening) of the resolved first-shape model that results from the First Inversion Observation Processing (Block 407) can be performed to account for the lower resolution of the Inversion Second Look Processing (block 413). There may be a large difference in the depth of investigation between the 211 propagation-type resistivity tool and the 151 BHA directional EM tool such that the directional EM tool's close spacing measurements resolve resistivity anisotropy at a spatial resolution less than the spatial resolution of the resistivity measurements of the propagation-type resistivity tool 211. In this case, the first resolved formation model may be scaled up (thickened) so as to take into account the lower spatial resolution of the measurements of the directional EM tool and the second reversal observation processing associated therewith. Such magnification may involve grouping multiple formation layers and assigning them the equivalent value for horizontal resistivity Rh and vertical resistivity Rv or anisotropic resistivity. The extended formation model can have an equal (or less) number of anisotropy parameters than the layers in the formation model. The extended training model can then be used as the initial training model (or guess) for the second inversion observation processing (block 413).

[0097] At block 411, a second formation model may be selected or initialized for use in the Second Inversion Observation Processing (block 413). An example of a suitable second formation model is described above and illustrated in Figure 3.

[0098] In certain cases (such as, when observation and prediction processing results (block 419) for a corresponding prediction zone are not available), the first enhanced formation model produced in block 409 can be selected for use as the second formation pattern in the Second Inversion Observation Processing (block 413).

[0099] In other cases (such as when the results of observation and prediction processing (block 419) for a corresponding prediction zone are available), initialization of the second formation model may involve integrating the formation dip and the formation profile vertical resistivity Rv or resistivity anisotropy values in the extended first formation model produced in block 409 where the formation dip and vertical resistivity profile Rv or resistivity anisotropy values are obtained from the characterization of the corresponding anterior anticipation zone (block 419 ).

[0100] In block 413, the processing of the second inversion observation is carried out, which involves the inversion of the EM measurements of the current observation zone 242 by the directional EM tool, which are sensitive to the variation in the resistivity anisotropy and the dip of the formation of the current observation zone 242. The inversion uses the second formation model and resolves the dip values and vertical resistivity Rv or anisotropy of the current observation zone 242. The layer thicknesses and horizontal resistivity of the Rh values of the second model of formation remain fixed in this inversion. The resolved second formation model provides a profile of the current observation zone 242, which can provide a sufficiently detailed characterization of the formation dip and profile of the horizontal resistivity Rh, vertical resistivity Rv or resistivity anisotropy of the layer(s) of formation of the current observation zone 242 in most cases.

[0101] In block 415, a reservoir model (or parts thereof) of the current observation zone can possibly be updated using the data from the resolved second formation model (block 413) for reservoir visualization and analysis and/or reservoir characterization . Furthermore, the operation (e.g. geo-steering) of the drill bit can possibly be controlled based on data from the resolved second formation model (block 413).

[0102] In block 417, a third formation model can be selected or initialized for use in Inversion Prediction Processing (Block 419). An example of a suitable third formation model is described above and illustrated in Figure 3. The third formation model may be based on the resolved second formation model that results from the Second Inversion Observation Processing (block 413) for the current zone observation point 242. For example, the second resolved formation model produced at block 413 may be selected for use as the third formation model in Inversion Prediction Processing (block 419). In another example, EM measurements from the Directional EM tool can be sensitive to boundaries when the transmitter or receiver antennas cross them. To address this issue, when building the third initial formation model, in the region behind the last receiver outside the prediction inversion window (i.e., the rear observation zone 243), the profile of the formation layers of the second formation model produced in block 413 can be thicker, keeping only the “significant” boundaries that contribute to the response. The resistivity profile is therefore segmented with a high threshold behind the last receiver position. Thus, in the region behind the last receiver position, the response sensitivity is very low and the effect of formation thresholds in this region is removed from the inversion prediction processing.

[0103] In block 419, the Inversion Prediction Processing is performed, which involves inverting the EM measurements of the current prediction zone 241 by the directional EM tool that are sensitive to variation in horizontal resistivity Rh, resistivity anisotropy and formation dip of prediction zone 241. In some embodiments, UHR measurements at all high frequencies from the directional EM tool are used in this inversion. The inversion uses the third formation model and resolves for formation dip and a profile of the horizontal resistivity Rh, vertical resistivity Rv or resistivity anisotropy and bed boundaries of the formation layer(s) from the third formation model. The third resolved formation model can provide formation dip as well as a dip profile, horizontal resistivity Rh and vertical resistivity Rv or resistivity anisotropy of the formation layer(s) of the current prediction zone 241, as well as position of bed boundaries of current forecast zone 241.

[0104] The inversion prediction processing (block 419) can use a pixel-based or model-based approach. In the pixel-based approach, the spread tool space 211 up to, for example, twice the largest spacing is typically discretized (with gradually increasing pixel size), and inversion is performed to resolve horizontal resistivity Rh and resistivity values pixel anisotropy using Rh-anisotropy differences regularization. In the model-based approach, the number of layers ahead of the propagation type resistivity tool 211 is defined and the inversion is performed for the horizontal resistivity Rh, resistivity anisotropy and thickness for each layer using regularization. Noise effects can be reduced using the L-curve method. In both cases, several initial estimates can be generated to avoid local inversion minima and to characterize the prediction zone. In some examples, initial estimates may be based on i) the resistivity profile of a previous prediction zone (block 419), ii) a homogeneous formation ahead of the propagation-type resistivity tool 211, and iii) a conductivity or layer of resistivity near and far from the directional EM tool's transmitter. It should be understood that any suitable number of estimates can be used according to some examples. The averaged solution of all initial guesses is then displayed to reduce possible noise artifacts and assess uncertainties. Formation resistivity profiles can be weighted with the mean inversion error term.

[0105] In block 421, a reservoir model (or parts thereof) of the current observation zone 241 can possibly be updated using data from the third resolved formation model (block 419) for reservoir visualization and analysis and/or characterization of reservoir. Furthermore, the operation (eg geo-targeting) of the drill bit can possibly be controlled based on data from the resolved second formation model (block 419).

[0106] Figure 5 is a flowchart illustrating exemplary operations that may be part of the First Inversion Observation Processing of block 407. Operations begin in block 501 where the first formation model is determined by inverting shallow resistivity measurements of the current observation zone 242 obtained by operating the propagation resistivity tool 211 (block 401) which are sensitive to the variation in resistivity of the current observation zone 242. The vertical and dip resistivity values or anisotropic resistivity of the first formation model remain fixed in this inversion.

[0107] In block 503, a cost function generated by the inversion results of block 501 is evaluated to determine whether the error is above a predefined threshold. In some examples, the cost function may be based on the discrepancy or mismatch between the measured and simulated surface resistivity measurements of the current observation zone 242. Otherwise, operations continue to block 513. If yes, operations continue to block 505.

[0108] In block 505, the boundary positions and horizontal resistivity of one or more formation layers in the first formation model are refined or adjusted, and operations continue to block 507. In some examples, the first formation model can be refined by changing the boundary position of the layer and the horizontal resistivity Rh of the layer for layers where the error function (or discrepancy between the measured and simulated resistivity measurements of the current observation zone 242) is high.

[0109] In block 507, the first formation model is again built by inverting the shallow resistivity measurements of the current observation zone obtained by operating the propagation resistivity tool 211 (block 401) which are sensitive to the variation in horizontal resistivity Rh from the current observation zone 242. The vertical and dip resistivity or anisotropic formation resistivity values of the first formation model remain fixed in this inversion.

[0110] In block 509, a cost function generated by the inversion results of block 507 is evaluated to determine whether the error is above a predefined threshold. In some examples, the cost function may be based on the discrepancy or mismatch between the measured and simulated surface resistivity measurements of the current observation zone 242. Otherwise, operations continue to block 513. If yes, operations continue to block 511.

[0111] In block 511, the number of layers in the first formation model can be adjusted. In some examples, the first formation model can be refined by splitting the layers to zones where the error function (or discrepancy between the measured and simulated resistivity measurements of the current observation zone 242) is high. Operations then return to block 501 to perform another iteration of the inversion process.

[0112] In block 513, the iterative process of the First Inversion Research Investment ends and the resolved first formation model provides a first resistivity profile (including Rh horizontal resistivity values and layer limits) for the formation layers of the current observation zone 242.

[0113] Figures 6A and 6B, collectively, are a flowchart illustrating exemplary operations that may be part of the Second Inversion Observation Processing of block 413.

[0114] Operations begin in block 601 where the second formation model is built using the inversion of the EM measurements of the current observation zone 242 obtained by operating the directional EM tool (block 403). These EM measurements are sensitive to changes in the resistivity anisotropy of the current observation zone 242. The horizontal Rh and dip resistivity values of the second formation model remain fixed in this inversion. In some embodiments, measurements of UHR (harmonic resistivity) of the directional EM tool at different high frequencies, mainly sensitive to the variation in resistivity anisotropy in the window corresponding to the short spacing length of the directional EM tool, are used in this inversion. This window can start 2/3 * L1 behind the current transmitter position (where L1 is the long window of the directional EM tool) to reduce the influence of formation ahead of the transmitter (Tx) of the directional EM tool.

[0115] In block 603, a cost function generated by the inversion results of block 601 is evaluated to determine whether the error is above a predefined threshold. In some examples, the cost function may be based on the discrepancy or mismatch between the measured and simulated surface resistivity measurements of the actual observation zone 242 by the directional EM tool. If not, the operations adjust one or more values of vertical resistivity Rv or anisotropy of the second formation model and the inversion of block 601 is repeated. If so, operations continue to block 605 where the resolved second formation model provides an estimate of the resistivity anisotropy or vertical resistivity Rv values of the formation layers of the current observation zone 242 and operations continue to block 607.

[0116] In block 607, the resolved second formation model (block 605) is refined using the inversion of EM measurements of the current observation zone 242 obtained by operating the directional EM tool (block 403). These EM measurements (such as UAD1, USD1) are sensitive to the variation in the dip of the current observation zone 242. The horizontal resistivity Rh and vertical resistivity Rv or anisotropy values of the second formation model remain fixed in this inversion. In some embodiments, anti-symmetric (UAD) and symmetry (USD) measurements at different high frequencies, primarily sensitive to variation in immersion in the window corresponding to the short spacing length of the directional EM tool, are used in inversion. This window can start 2/3 * L1 behind the current transmitter position (where L1 is the short spacing of the directional EM tool) to reduce the influence of formation ahead of the transmitter of the directional EM tool.

[0117] In block 609, a cost function generated by the inversion results of block 607 is evaluated to determine whether the error is above a predefined threshold. In some examples, the cost function may be based on the discrepancy or mismatch between the measured and simulated EM resistivity measurements of the current observation zone 242. If not, operations adjust the dip of the second formation model and the inversion of block 607 is repeated. If yes, operations continue to block 611, where the resolved second formation model provides an estimate of the formation dip of the layer structure of the current observation zone 242, and operations continue to block 613.

[0118] In block 613, the resolved second formation model (block 611) is refined using the inversion of EM measurements of the current observation zone 242 obtained by operating the directional EM tool (block 403). These EM measurements (such as UHR1, UAD1, USD1) are sensitive to the dip and anisotropic resistivity variation of the current observation zone 242. The horizontal and dip resistivity values of the second formation model remain fixed at this inversion. In some embodiments, UHR measurements, UAD measurements, and USD measurements at different high frequencies sensitive to variation in dip anisotropy and resistivity over an extended window can be used in this inversion. In one example, the extended window corresponds to the length of the directional EM tool from L1/3 to 5L1/3 behind the current transmitter position (where L1 is the short spacing of the directional EM tool). As measurements closer to the transmitter of the directional EM tool are used, the resistivity profile can be augmented with the inversion results for a corresponding earlier prediction zone (block 419) to account for the effects of formation ahead. The resistivity anisotropy or vertical Rv and dip resistivity values of the second formation model are reversed. To avoid a local minimum, an additional inversion with different starting point can be added to the inversion of block 613, with inversion dip by dip and assuming a value of 1.2 for all layer anisotropies in this example.

[0119] In block 615, a cost function generated by the inversion results of block 613 is evaluated to determine whether the error is above a predefined threshold. In some examples, the cost function may be based on the discrepancy or mismatch between the measured and simulated EM resistivity measurements of the current observation zone 242. If not, operations adjust one or more values of vertical resistivity Rv or anisotropy from the second model of formation and inversion of block 613 is repeated. If so, operations continue to block block 617, where the resolved second formation model provides the formation dip, as well as the anisotropy values of vertical resistivity or resistivity of the formation layers of the current observation zone 242. Solved second formation model can be selected from multiple solutions, where the solved second formation model has a lower residual.

[0120] Note that, in some embodiments, the inversion of resistivity anisotropy (blocks 601 - 605) and the inversion of inversion (blocks 607 - 611) can be omitted. In this case, the resistivity anisotropy and dip inversion (blocks 613 - 617) provide the resistivity and dip anisotropy inversion that characterizes the anisotropic resistance and dip of the formation layers of the current observation window 242.

[0121] Figure 7 is a graphical representation of an exemplary workflow that uses the LWD modules of the BHA 151 of Figure 2A along with the multi-step observation processing workflow that characterizes the current observation zone, as well as the prediction processing that characterizes the current appearance of the prediction zone, which is similar to the operations described above with respect to Figures 4A-4C, 5 and 6A-6B. The window of processed measurement data from the propagation-type resistivity tool (CDR tool) relative to the directional EM tool's transmitter subassembly 210 is shown in a wide solid gray line, and the processed windows of the various EM measurements from the relative directional EM tool to the directional EM tool transmitter 210 subassembly is shown in three distinct solid gray lines. The formation zones being characterized using the inversions of observation processing (blocks 407 and 413) and prediction processing (block 419) are specified by black arrows. Note that EM measurements (eg UAD1, UAD2, USD1, USd2, UHR1, UHR2) are designated with the reference numbers 1 or 2 that correspond to the ID of the receiver acquiring the respective EM measurements.

[0122] Figure 8 is a graph illustrating how dip and anisotropy affect the RCDR apparent resistivity and, consequently, the estimate of the horizontal resistivity Rh by the first observation inversion processing (block 407) of Figures 4A-4C and 5 Values are shown for the phase shift response measured by the 2MHz CDR tool in a homogeneous formation with Rh = 5Qm. The error in Rh progressively increases with dip or anisotropy, so that at 20° dip with an anisotropy of 5, the error is 5%. This error may be enough to affect early interpretation if not corrected. However, at a relative dip above 30°, the difference between the measured apparent resistivity and the true Rh resistivity increases and responses are more sensitive to small changes in immersion or anisotropy.

[0123] The First Inversion Observation Processing (block 407) can be configured to take into account a greater relative dip of the formation layers. Specifically, ignoring anisotropy in the First Inversion Observation Processing (block 407) can lead to more significant overestimations of the horizontal Rh resistivity values of formation layers at high relative dip. To overcome this limitation, the First Inversion Observation Processing can be iterated using the dip and anisotropies derived from the Second Inversion using directional EM (block 411) such that it converges to the true values of the horizontal Rh formation resistivity of the formation layers , but this can be time consuming and potentially unstable. In another example, the dip and anisotropy dependence of the propagation tool phase shift measurements can be taken into account within the First Inversion Observation Processing.

[0124] Figure 9A is a flowchart illustrating exemplary operations that may be part of the First Inversion Observation Processing (for example, block 407) and used to account for a higher relative formation dip. In this example, a look-up table (similar to the table in Figure 8) or other suitable data processing construct can be provided to tabulate the dependence of the apparent resistivity measured by the propagation-type resistivity tool for dip, anisotropy, and horizontal resistivity Rh for a homogeneous formation or a 1D EM simulator can be used to model tool responses to the immersion anisotropic formation model. Operations begin at block 901 where the first forming model (which is initialized at block 405) is constructed using resistivity inversion measurements of the current observation zone 242 obtained by operation of the spread-type resistivity tool 211 that are sensitive to variation on the surface resistivity of the current observation zone 242.

[0125] In block 903, a cost function generated by the inversion results of block 901 is evaluated to determine whether the error is above a predefined threshold. In some examples, the cost function may be based on the discrepancy or mismatch between the measured and simulated surface resistivity measurements of the current observation zone 242. In this case, operations continue to block 909. Otherwise, operations continue to blocks 905 and 907.

[0126] In block 905, the formation dip and/or the vertical resistivity or the resistivity anisotropy of the formation layers of the second formation model can be updated.

[0127] In block 907, the apparent resistivity (or the equivalent phase shift measurement) measured by the propagation-type resistivity tool for each layer of formation of the current observation zone 242 is passed to the lookup table or other construct of data processing, which is used to query the horizontal Rh resistivity value that corresponds to the measured apparent resistivity (or the 2MHz or 400kHz equivalent phase shift or attenuation measurement) and the updated dip and/or anisotropic resistivity of the formation represented by the appropriate lookup table entry or data processing construct. In addition, the updated horizontal resistivity values identified by consulting the table, as well as the updated dip and/or the updated vertical resistivity or the updated resistivity anisotropy of the formation layers (block 905) are used to update the first formation model and the inversion of block 901 is repeated.

[0128] In block 909, the resolved first formation model provides the formation dip, as well as the horizontal resistivity Rh and vertical resistivity Rv or resistivity anisotropy values of the formation layers of the current observation zone 242. In this way, the The first observation inversion processing of Figure 9 employs an inversion that initially lacks anisotropy and dip information. However, the inversion is repeated (e.g., multiple times) until it converges to the correct dip and anisotropic resistance of the formation layers of the current observation zone, according to some examples.

[0129] In some examples, the First Processing of Inversion Observation of Figure 9A may be adequate as long as the initialization of the first formation model (which can use the application of logarithm squared as described above) produces a reasonable estimate of the positions of threshold and apparent resistivities are not significantly affected by the shoulder-based effect. At relative dip angles greater than about 60°, propagation-type resistivity tool measurements 211 can become more sensitive to the effect of shoulder bed and bias horns on resistivity can make the process of applying the logarithm to the inadequate square for starting the first training model. In this case, other resistivity data can be used to initialize the first formation model.

[0130] In other examples, such as at higher dive slopes, it may be appropriate to scale all window lengths and distances to reflect true vertical depth (TVD) rather than measured depth (MD), i.e. split them by cosθ, where θ is the slope. Note that MD corresponds to depth as a length measured along the hole.

[0131] Figure 9B is a flowchart illustrating exemplary inversion operations that may be part of the Inversion Prediction Processing (for example, block 419) of the workflows described in this document. In this workflow, the inversion operations of block 921 invert the current-ahead zone EM measurements to simultaneously resolve for formation dip as well as resistivity profile values (including horizontal resistivity Rh and vertical resistivity Rv (or anisotropy of resistivity) for the formation layers) of the current forecast zone. EM measurements of the current prediction zone that are sensitive to the horizontal resistivity Rh and the vertical resistivity Rv (or resistivity anisotropy) and formation dip of the current prediction zone can be used in block 921 inversion operations. and multi-step can be performed as part of block 921.

[0132] Figure 9C is a flowchart illustrating exemplary inversion operations that may be part of the Inversion Prediction Processing (for example, block 419) of the workflows described in this document. In this workflow, block 951 inversion operations invert the EM measurements of the current prediction zone to resolve resistivity profile values (including horizontal resistivity Rh and vertical resistivity Rv (or resistivity anisotropy) for the formation layers) of the current forecast zone. The number of layers, layer limits and formation dip of the current forecast zone can be fixed in the inversion operations of block 951. The formation dip of the current forecast zone can be obtained from the formation dip of an observation zone or from elsewhere. EM measurements of the current prediction zone that are sensitive to the horizontal resistivity Rh and the vertical resistivity Rv (or resistivity anisotropy) of the current prediction zone can be used in block 951 inversion operations. may be performed as part of block 951.

[0133] Figure 10 is a flowchart illustrating a single-step inversion workflow that characterizes the observation zone. This workflow can replace the operations of the First and Second Inversion View Processes (blocks 405 to 413). The workflow begins at block 1001, where a first formation model is initialized for use in Single Step Inversion View Processing (Block 1003).

[0134] In certain cases, the first formation model can be initialized with Rh values and layer limits determined from the application of the logarithm squared of the resistivity measurements of the observation zones and current rear observation using the resistivity tool propagation type. The first formation model also receives an initial dip and isotropic resistivity (where vertical resistivity corresponds to horizontal resistivity) for all layers,

[0135] In other cases, the first formation model can be initialized with dip values as well as horizontal resistivity Rh and vertical resistivity values Rv or resistivity anisotropy values and layer limits obtained from Prediction Processing (block 419 ) of a corresponding previous prediction zone. Initialization of the first formation model can also be based on horizontal resistivity values Rh and layer boundaries obtained from First Inversion Observation Processing (block 1003) of a previous observation zone.

[0136] In block 1003, a one-step observation inversion processing is performed, which involves building a first formation model using the inversion of measurements of the current observation zone 242 obtained by operating the resistivity tool of the type 211 propagation and the directional EM tool. The 211 propagation-type resistivity tool and directional EM tool measurements are sensitive to the variation in dip, horizontal Rh resistivity, and directional anisotropic resistance of the current appearance around zone 242. This inversion resolves the formation dip values as well as horizontal resistivity RH and vertical resistivity Rv or resistivity anisotropy values of the formation layers of the first formation model. The resolved first formation model provides formation dip and a resistivity profile of the current observation zone, which can provide a sufficiently detailed profile of horizontal resistivity Rh and vertical resistivity Rv or resistivity anisotropy values of the formation layers of the surrounding current. from zone 242.

[0137] The inversion of block 1003 can be performed using a pixel-based approach or a model-based approach. In the pixel-based approach, the formation is discretized into thin layers of pixels (eg smaller than 0.5 m, or any other suitable size). An inversion resolves for formation dip as well as horizontal resistivity Rh and/or vertical resistivity Rv or anisotropic resistivity for each pixel using propagation-type resistivity tool measurements and EM measurements (preferably close-spaced high-frequency deep directional measurements) of the directional EM tool. Inversion can employ regularization on Rh horizontal resistivity differences and resistivity anisotropy according to some examples. In the model-based approach, the initial training model can be based on the log-squaring process (as described above). An inversion solves for the formation dip values as well as horizontal resistivity Rh and/or vertical resistivity Rv or resistivity anisotropy for the layers of the formation model using the propagation resistivity tool measurements and the EM measurements (preferably deep directional measurements high-frequency short-spaced) of the directional EM tool. Both the pixel-based and model-based approaches to one-step observation inversion do not require adjustments for high relative dip, as dip coupling and anisotropy of propagation tool phase shift measurements are automatically taken into account. .

[0138] Figures 11A - 11C, collectively, are a flowchart illustrating a workflow that characterizes the current observation zone 242 and the current anticipation zone 241 using only EM measurements from the directional EM tool. In this case, the propagation tool 211 can be omitted from the BHA 151, and the current prediction zone 241 can extend ahead of the directional EM tool transmitter subassembly 210 and can extend ahead of the drill bit 212. The current observation zone 242 extends from behind the directional EM tool transmitter subassembly 210 to the farthest receiver subassembly 209 (taking into account the length of the prediction inversion window). And the current rear observation zone 243 extends behind the farthest receiver subset 209 (with respect to the direction of travel of the bit while drilling).

[0139] The workflow starts in block 1101, where the measurements of the directional tool EM short spacing and EM long spacing are acquired at multiple frequencies and processed to characterize the current observation zone, the current rear observation zone and the current forecast zone.

[0140] At block 1103, a first formation model may be initialized for use in First Inversion Observation Processing (block 1105).

[0141] In certain cases, the first formation model can be initialized with Rh values and layer limits for known data, an initial dip and isotropic resistivity (where vertical resistivity corresponds to horizontal resistivity) for all layers.

[0142] In other cases, the first formation model can be initialized with Rh values, layer limits, dip values, vertical resistivity or anisotropy obtained from Prediction Processing (block 1115) of a corresponding previous prediction zone. The initialization of the first formation model can also be based on the formation dip values, as well as horizontal resistivity Rh, vertical resistivity Rv or resistivity anisotropy and the layer limits obtained from the observation processing (block 1109) of a previous observation zone.

[0143] In block 1105, the first inversion observation processing is performed, which involves inverting the first formation model using EM measurements of the current observation zone 242 by the directional EM tool (block 1101) that are sensitive to changes in resistivity horizontal Rh and vertical resistivity Rv or resistivity anisotropy of the current observation zone 242. In embodiments, EM measurements that are sensitive to changes in the horizontal resistivity Rh and the vertical resistivity Rv or resistivity anisotropy of the current observation zone 242 include the measurements of UHR short spacing (for example, UHR1) of the directional EM tool. The inversion resolves to the horizontal resistivity Rh and vertical resistivity Rv values or resistivity anisotropy and layer boundaries of the current observation zone 242. The formation dip value of the first formation model remains fixed in this inversion. The resolved first formation model provides a first resistivity profile of the current observation zone 242, which can provide a sufficiently detailed profile of horizontal resistivity Rh and vertical resistivity Rv or resistivity anisotropy values of the layers of current formation observation zone 242 .

[0144] In block 1107, a second formation model can be selected or initialized for use in the Second Inversion Observation Processing (block 1109). The initialization of the second formation model can be based on the resolved first formation model (Block 1105). When the resistivity profile of a corresponding early prediction zone is available from previous prediction processing (block 1115), initialization of the second formation model may involve integrating a formation dip value into the second formation model where the dip value is obtained from the previous prediction zone profile.

[0145] In block 1109, the second inversion observation processing is performed, which involves inverting the EM measurements of the current observation zone 242 (block 1101) that are sensitive to variation in the formation dip of the current observation zone 242 In embodiments, the EM measurements that are sensitive to variation in the formation dip of the current observation zone 242 include the short spacing USD and UAD measurements (eg, USD1 and UAD1) from the Directional EM tool. This inversion resolves to the second formation model, which includes a formation dip value of the current observation zone 242. The layers, layer thicknesses, and horizontal resistivity Rh and vertical resistivity Rv values or resistivity anisotropy from the second model of formation remain fixed in this inversion. The resolved second formation model predicts a formation dip as well as a resistivity profile (including horizontal resistivity Rh, vertical resistivity Rv, or resistivity anisotropy values of the formation layers) of the current observation zone 242.

[0146] In block 1111, a reservoir model (or parts thereof) of the current observation zone 242 can possibly be updated using data from the resolved second formation model (block 1109) for visualization and/or analysis of reservoir and/or reservoir characterization. Furthermore, the operation (eg geo-targeting) of the drill bit can possibly be controlled based on data from the resolved second formation model (block 1109).

[0147] In block 1113, a third formation model can be selected or initialized for use in Inversion Prediction Processing (Block 1115). The selection or initialization of the third formation model may be based on the resolved second formation model that results from the Second Inversion Observation Processing for the current observation zone 242 (block 1109). EM measurements from the Directional EM tool can be sensitive to boundaries when the transmitter or receiver antennas cross them. To address this issue, when building the third initial formation model, in the region behind the last receiver outside the prediction inversion window (i.e., the rear observation zone 243), the profile of the formation layers can be thicker, keeping only the “significant” thresholds that contribute to the response. The resistivity profile is therefore segmented with a high threshold behind the last receiver position. Thus, in the region behind the last receiver position, the response sensitivity is very low and the effect of formation thresholds in this region is removed from the inversion prediction processing.

[0148] In block 1115, the Inversion Forecast Processing is performed, which involves inverting the third formation model using EM measurements of the current anticipation zone 241 that are sensitive to variation in horizontal resistivity Rh and vertical resistivity Rv or anisotropy of resistivity as well as formation dip of current anticipation zone 241. In some embodiments, UHR measurements at all high frequencies of the directional EM tool are used in this inversion. The inversion resolves the values of horizontal resistivity Rh and/or vertical resistivity Rv or resistivity anisotropy, as well as the formation dip of the current prediction zone 241. The third resolved formation model can provide a profile of formation dip, horizontal resistivity Rh and Rv vertical resistivity or resistivity anisotropy of the formation layer(s) of current forecast zone 241, as well as position of bed boundaries of current forecast zone 241.

[0149] The inversion prediction processing (block 1115) can use a pixel-based or model-based approach. In the pixel-based approach, the prediction space is discretized (with gradually increasing pixel size), and inversion is performed to resolve the pixel resistivity anisotropy and Rh horizontal resistivity values of the pixels using Rh difference regularization and anisotropy. In the model-based approach, the number of layers of the prediction zone is defined and the inversion is performed for horizontal resistivity Rh, resistivity anisotropy and thickness for each layer using regularization. Noise effects can be reduced using the L-curve method. In both cases, several initial estimates can be generated to avoid local inversion minima and to characterize the prediction zone. In some examples, initial estimates may be based on i) the profile of a previous prediction zone (block 1115), ii) a homogeneous formation ahead of the directional EM tool, and iii) a conductive or resistive layer near and far from the transmitter. of the directional EM tool. It should be understood that any suitable number of estimates can be used according to some examples. The averaged solution of all initial guesses is then displayed to reduce possible noise artifacts and assess uncertainties. Formation resistivity profiles can be weighted with the mean inversion error term.

[0150] In block 1117, a reservoir model (or parts thereof) of the current prediction zone 241 can possibly be updated using data from the third resolved formation model (block 1115) for reservoir visualization and analysis. Furthermore, the operation (e.g. geo-steering and/or geo-braking) of the drill bit can possibly be controlled based on data from the resolved second formation model (block 1115).

[0151] In other embodiments, the multi-step inversion observation processing of blocks 1105-1109 can be replaced by the one-step Inversion Observation Processing, which involves building the first formation model by measuring the EM measurements of the current zone observation zone 242 (block 1101) that are sensitive to variation in horizontal resistivity Rh and vertical resistivity Rv or resistivity anisotropy, as well as the formation dip of the current observation zone 242. This inversion resolves to horizontal resistivity Rh and vertical resistivity Rv or anisotropy values as well as the formation dip of the formation layers of the current observation zone 242. The resolved first formation model provides a first resistivity profile of the current observation zone 242, which can provide a sufficiently detailed profile of the values of anisotropy formation dip and horizontal resistivity Rh and vertical resistivity Rv or anisotropy of the formation layers of the current observation zone 242.

[0152] In still other embodiments, the BHA 151 can use a laterolog resistivity imaging tool that is used instead of the propagation-type resistivity tool, which is not sensitive to resistivity anisotropy in low deviation wells. In this case, resistivity measurements from the laterolog resistivity image tool can be interpreted as part of the observation and prediction workflow as described here.

[0153] In still other embodiments, the BHA 151 can use a triaxial resistivity tool that is used in place of the propagation-type resistivity tool, which is not sensitive to resistivity anisotropy in low deviation wells. In this case, the resistivity measurements from the triaxial resistivity tool can be interpreted as part of the observation and prediction workflow as described here. In this case, multi-step inversions can be used as part of the observation processing where the first inversion inverts the resistivity measurements of the triaxial resistivity tool to resolve the horizontal resistivity Rh and vertical resistivity Rv values or the resistivity anisotropy of the layers of the observation zone (with a fixed formation dive). The second inversion reverses the resistivity measurements of the triaxial resistivity tool to resolve dip formation of the observation zone layers. In still other embodiments, a one-step inversion can be used as part of the observation processing where the inversion inverts the resistivity measurements of the triaxial resistivity tool to resolve the formation dip values, as well as horizontal resistivity Rh and vertical resistivity Rv or resistivity anisotropy of layers in the observation zone. In either case, the directional EM tool's close-spaced and long-spaced EM measurements need not be used to characterize the observation zone resistivity profile, but are used to characterize the prediction zone profile.

[0154] Note that the workflows as described here can be run in real time at multiple locations (sometimes called "gauging stations") along the well during drilling to model (characterize) the observation zone and the Real-time formation preview zone as drilling progresses through the formation. This real-time modeling can be used to provide real-time geo-steering of the bit and real-time visualization of the formation as drilling progresses through the formation. Simulation results

[0155] Figures 12A-12B, 13A-13B, 14A-14B, 15A-15B, 16, 17A-17B, 18, 19A-19B and 20A-20B are provided to help visualize the results of the workflows described in this document. Figures 12A-12B, 13A-13B, 14A-14B, 15A-15B, 17A-17B, 19A-19B and 20A-20B are tracking dials, which are of the same type as those described in detail in US Patent 8,862,405 , which is hereby incorporated by reference in its entirety. Figures 12A, 13A, 14A, 15A, 17A, 19A and 20A are Rv/Rh anisotropy ratio maps. The thin column/panel to the right of the respective Rv/Rh anisotropy ratio map corresponds to the values used in the synthetic model (the closer the inversion shading of these right columns, the higher the inversion quality). Figures 12B, 13B, 14B, 15B, 17B, 19B and 20B are Rh horizontal resistivity maps. The thin column/panel on the right of the respective Rh map corresponds to the values used in the synthetic model (the closer the inversion shading of these right columns, the higher the inversion quality). In these figures, the stars correspond to the transmitter position of the tool as it moves down (from left to right - each column represents the flip results for a given tool position). All layer resistivities and anisotropy ratios below the stars are obtained by inverting prediction. The bottom graph of Figures 12B, 13B, 14B, 15B, 17B, 19B and 20B shows the anisotropy error plus dip reversal and forecast reversal. The bottom graph of Figures 12A, 13A, 14A, 15A, 17A, 19A and 20A shows the estimated dip of the final observation inversion with its uncertainty.

[0156] For each inversion, these figures display the current formation which is used to find the formation ahead. You can now follow how the profile is thickened behind the far receiver. The last layer, which was displayed with a slight 10-foot gradient to white, is extended to infinity in reverse.

[0157] The prediction inversion uncertainty is added to the Rh and R/Rh anisotropy ratio graphs (gradient to white for threshold uncertainty and alternating shading for Rh uncertainty or anisotropy).

[0158] The prediction interpretation, using the complete observation and prediction workflow, is compared with the hypothetical case of perfect measurements of non-directional propagation, as if the resistivity profile until the last propagation tool measurement station were known exactly. If the perfect non-directional propagation measurements are triaxial then this is an “Rt-perfect” case with full information about all layer boundaries, resistivities, anisotropies and immersion. If the perfect non-directional propagation measurements are coaxial in vertical or low deviation wells, then this is a case of “Rh-perfect” with information about layer boundaries and unknown horizontal resistivities and anisotropies.

[0159] Figures 12A and 12B show the results and inversion errors for the "perfect Rt case". Figures 13A and 13B show the inversion and error results for the perfect Rh case. Note that the “Rh-perfect” case inversion error (Figures 13A and 13B) is up to an order of magnitude greater than the “Rt-perfect” case error (Figures 12A and 12B) when the anisotropy of the formation changes significantly within the antennas' prediction inversion window (for example, around the 60 or 90 inversion point in the illustrated example). The quality of the forecast reversal reflects this increase in error. Furthermore, although there are only small anisotropy variations in the conductive layers above zero feet TVD, the prediction capabilities and estimation consistency of the formation ahead is reduced to “Rh-perfect” from the inversion point 0 to 30 in the illustrated example. . This shows that without a complete knowledge of the observation formation (threshold, Rh, anisotropy and dip) - that is, omitting some of this information (the anisotropy in this case) compromises the prediction inversion.

[0160] Figures 14A and 14B show the corresponding tracking graph using the complete multi-step inversion processing that characterizes the observation zone. Anisotropy and dip are well recovered. The inversion quality and predictive ability is comparable to the “Rt-perfect” case as it only shows highly uncertain thin layers than the “Rt-perfect” case. Consequently, the forecast inversion error is equal to or only minimally greater than the “perfect Rt” case forecast inversion error term in the illustrated example.

[0161] Two additional results are shown in Figures 15A-15B, 16, 17A-17B and 18. In both cases, the visual inversion is able to reconstruct the formation that the tool crossed very well (boundaries, Rh, anisotropy and dip training), leading to non-committal forecast reversal results. Figure 16 shows a sample comparison between the true profile and the profile derived from the observation inversion (before the Rx2 back thickening) for the transmitter at +20 feet true vertical depth (TVD) (zero feet relative to TVD on the plot , formation below -7 feet TVD being the previous forecast result). Figure 18 shows a sample comparison between the true resistivity profile and the observation-derived profile (before thickening behind a distant receiver) for a transmitter at a true vertical depth of +100 feet (zero feet relative to TVD at plot, formation below -7 feet TVD being the previous forecast result).

[0162] Figures 19A-19B and 20A-20B show two workflow results in high relative diving. Consequently, the workflow in Figure 9 was used for these two examples. Again, the visual inversion is able to accurately reconstruct the formation that the tool crossed (boundaries, Rh, anisotropy and formation dip).

[0163] Although conventional shallow resistivity tools (coaxial antenna) are provided as per some examples, it should be understood that the algorithms and workflows described herein can be applied to manipulate total triaxial shallow resistivity measurements.

[0164] Figure 21 shows an example computing system 2100 that can be used to implement the computer processing system 203 of Figure 1A or parts thereof. Computing system 2100 may be an individual computer system 2101A or an array of distributed computer systems. Computer system 2101A includes one or more analysis modules 2103 (a program of computer-executable instructions and associated data) that can be configured to perform various tasks, in some embodiments, such as the tasks described above. To perform these various tasks, an analysis module 1303 executes on one or more processors 2105, which is (or are) connected to one or more storage media 2107. The processor(s) 2105 is (or are) also connected to a network interface 2109 to allow the computer system 2101A to communicate over a data network 2111 with one or more computer systems and/or additional computing systems, such as 2101B, 2101C, and/or 2101D. Note that computer systems 2101B, 2101C and/or 2101D may or may not share the same architecture as computer system 2101A and may be located in different physical locations.

[0165] The processor 2105 may include a microprocessor, microcontroller, processor module or subsystem, programmable integrated circuit, programmable gate array, digital signal processor (DSP) or other computing or control device.

[0166] The storage media 2107 may be implemented as one or more non-transient computer-readable or machine-readable storage media. Note that although in the embodiment of Figure 21, storage media 2107 are depicted as in computer system 2101A, in some embodiments, storage media 2107 may be distributed within and/or across multiple internal enclosures and/or external components of the 2101A computing system and/or additional computing systems. Storage media 2107 may include one or more different forms of memory, including semiconductor memory devices such as dynamic or static random access memories (DRAMs or SRAMs), programmable and erasable read-only memories (EPROMs), electrically read-only memories programmable and erasable memory (EEPROMs) and flash memories; magnetic disks, such as fixed, floppy and removable disks; other magnetic media, including tape; optical media such as compact discs (CDs) or digital video discs (DVDs); or other types of storage devices. Note that the computer-executable instructions and associated data of the analysis module(s) 2103 may be provided on a computer-readable storage medium or machine-readable storage medium 2107, or alternatively may be provided on multiple computer-readable devices or machine-readable storage media distributed across a large system with possibly multiple nodes. Such computer-readable or machine-readable storage media or media are considered to be part of an article (or an article of manufacture). An article or an article of manufacture can refer to any single component or multiple manufactured components. The storage media or media may either be located on the machine executing the machine-readable instructions, or located at a remote location from which machine-readable instructions can be downloaded over a network for execution.

[0167] It should be appreciated that the computing system 2100 is only one example of a computing system, and that the computing system 2100 may have more or less components than indicated, may combine additional components not described in the example embodiment of Figure 21, and/or computing system 2100 may have a different configuration or arrangement of components depicted in Figure 21. The various components shown in Figure 21 may be implemented in hardware, software, or a combination of hardware and software, including one or more application-specific integrated circuits and/or signal processing.

[0168] In addition, the operations of the computer processing system 203 as described in this document can be implemented by executing one or more functional modules in information processing devices such as general purpose processors or application-specific chips, such as ASICs, FPGAs, PLDs, SOCs, or other appropriate devices. These modules, combinations of these modules, and/or their combination with general hardware are all included within the scope of disclosure protection.

[0169] In one embodiment, the operations of the computer processing system 2100, as described herein, can be implemented by running one or more functional modules on an information processing apparatus (such as a workstation) located at the well site and/or in an information processing apparatus that is part of the downhole tool's BHA.

[0170] In another embodiment, the operations of the computer processing system 203 as described herein can be implemented by running one or more functional modules on a cloud-based information processing apparatus.

[0171] The methods and processes described above, such as, for example, modeling, plotting, analysis and/or control of any recited hardware, can be performed by a processing system. The processing system may include a single processor, multiple processors or a computer system. When the processing system includes multiple processors, the multiple processors may be arranged in a single device or in different devices, at equal locations or remote from one another. The processor or processors may include one or more computer processors (e.g., a microprocessor, microcontroller, digital signal processor, or general purpose computer) for performing any of the methods and processes described above. The computer system may additionally include memory, such as a semiconductor memory device (e.g., RAM, ROM, PROM, EEPROM, or Flash-programmable RAM), a magnetic memory device (e.g., a floppy disk or fixed disk), an optical memory device (for example, a CD-ROM), a PC card (for example, a PCMCIA card), or another memory device.

[0172] Thus, the methods and processes described above can be implemented as computer program logic for use with the computer's processor. Computer program logic can be modalized in a number of ways, including a source code form or a computer executable form. Source code may include a series of computer program instructions in a variety of programming languages (for example, object code, an assembly language, or a high-level language such as C, C++, JAVA, Matlab or other language). These computer instructions may be stored on a non-transient computer-readable medium (eg, memory) and executed by the computer's processor. Computer instructions may be distributed in any form as a removable storage medium with attached printed or electronic documentation (e.g. retractable software), pre-loaded with a computer system (e.g. in system ROM or fixed disk) or distributed from a server or electronic bulletin board over a communication system (eg, the Internet or World Wide Web).

[0173] Alternatively or additionally, the processing system may include discrete electronic components, coupled to a printed circuit board, integrated circuits (for example, Application Specific Integrated Circuits (ASIC) and/or logic devices (eg, a Field Programmable Gate Arrays (FPGA) Any of the methods and processes described above can be implemented using these logic devices.

[0174] Any of the methods and processes described above can be implemented as computer program logic for use with the computer's processor. Computer program logic can be modalized in a number of ways, including a source code form or a computer executable form. Source code may include a series of computer program instructions in a variety of programming languages (for example, an object code, an assembly language, or a high-level language such as C, C++ or JAVA) . These computer instructions may be stored on a non-transient computer-readable medium (eg, memory) and executed by the computer's processor. Computer instructions may be distributed in any form as a removable storage medium with attached printed or electronic documentation (e.g. retractable software), pre-loaded with a computer system (e.g. in system ROM or fixed disk) or distributed from a server or electronic bulletin board over a communication system (eg, the Internet or World Wide Web).

[0175] To the extent used in this description and the claims, a citation in the general form of "at least one of [a] and [b]" is to be construed as disjunctive. For example, a citation of "at least one [a], [b] and [c]" would include [a] alone, [b] alone, [c] alone, or any combination of [a], [b] and [ w].

[0176] While some example embodiments have been described in detail above, those skilled in the art will readily appreciate that many modifications are possible to the example embodiments without materially departing from the embodiments disclosed herein. Accordingly, all such modifications are intended to be included within the scope of this disclosure.

Claims (33)

1. Method for characterizing a geological formation traversed by an exploration well while drilling the exploration well using a resistivity tool in combination with a directional electromagnetic (EM) tool, the method characterized in that it comprises: operating the resistivity to acquire non-directional resistivity measurements of a formation observation zone, and operating the directional electromagnetic tool to acquire directional EM measurements of a formation observation zone and a formation prediction zone, wherein the electromagnetic tool comprises a transmitter subassembly having a triaxial antenna array comprising axial, tilt and transverse antennas; determine an observation zone profile by interpreting observation zone non-directional resistivity measurements and observation zone directional EM measurements, where the observation zone profile features at least formation dip as well as horizontal resistivity and resistivity vertical or resistivity anisotropy of one or more layers forming the observation zone; and determining a prediction zone profile by interpreting directional EM measurements of the prediction zone, wherein the prediction zone profile characterizes at least formation dip, as well as horizontal resistivity and vertical resistivity or resistivity anisotropy of one or more layers of prediction zone formation, where interpretation of directional EM measurements uses a formation model that is initialized according to values from the interpretation profile of non-directional resistivity measurements. 2. Method, according to claim 1, characterized in that it further comprises: updating a reservoir model of the observation zone based on the profile of the observation zone. 3. Method according to claim 1, characterized in that it further comprises: controlling the operation of a drill bit based on the profile of the observation zone. 4. Method according to claim 1, characterized in that it further comprises: updating a prediction zone reservoir model based on the prediction zone profile. 5. Method according to claim 1, characterized in that it further comprises: controlling the operation of a drill bit based on the profile of the prediction zone. 6. Method, according to claim 1, characterized by the fact that: the interpretation of directional EM measurements of the observation zone involves three distinct inversions, including an inversion of resistivity anisotropy of directional EM measurements that are sensitive to variation in anisotropy of observation zone resistivity, a dip inversion of directional EM measurements that are sensitive to variation in the formation dip of the observation zone, and an anisotropy of resistivity and dip inversion of directional EM measurements that are sensitive to variation in both anisotropy of resistivity and in the formation dip of the observation zone. 7. Method according to claim 6, characterized by the fact that: the inversion of resistivity anisotropy uses a formation model of the observation zone with a fixed formation dip and fixed horizontal resistivity for one or more formation layers of the observation zone; dip inversion uses an observation zone formation model with a fixed horizontal resistivity and fixed vertical resistivity or resistivity anisotropy for one or more layers of observation zone formation; and resistivity anisotropy and dip inversion use an observation zone formation model with a fixed horizontal resistivity for one or more layers of observation zone formation. 8. Method, according to claim 1, characterized by the fact that: the interpretation of the directional EM measurements of the observation zone involves a single inversion of the EM measurements that are sensitive to variation in both the formation dip and the resistivity anisotropy of the observation zone to resolve the formation dip and resistivity anisotropy of the observation zone. 9. Method, according to claim 1, characterized by the fact that: determining the profile of the observation zone involves interpreting the non-directional resistivity measurements of the observation zone to estimate horizontal resistivity of one or more layers of formation of the zone observation zone, and interpret the directional EM measurements of the observation zone using the horizontal resistivity estimate of one or more observation zone formation layers to determine the observation zone profile. 10. Method according to claim 9, characterized by the fact that: the spatial resolution of the non-directional resistivity measurements of the resistivity tool has a greater resolution than the directional EM measurements of the directional electromagnetic tool. 11. Method, according to claim 1, characterized by the fact that: the directional electromagnetic tool employs at least one inclined or transverse antenna in a sub transmitter and a sub receiver. 12. Method, according to claim 9, characterized by the fact that: the interpretation of non-directional resistivity measurements of the observation zone involves the inversion of such non-directional resistivity measurements, in which the inversion uses a formation model that characterizes resistivity of one or more layers of the observation zone. 13. Method, according to claim 12, characterized by the fact that the formation model is initialized from at least one of: applying the logarithm to the square of the non-directional resistivity measurements of the observation zone; a profile of a previous observation zone; and a profile of a previous forecast zone. 14. Method, according to claim 12, characterized by the fact that: the inversion of non-directional resistivity measurements of the observation zone solves the formation model that characterizes the resistivity of one or more layers of the observation zone; the solved formation model is scaled up to account for the lower spatial resolution in the directional EM measurements of the observation zone and prediction zone; and the extended formation model is used to define an initial formation model used in the interpretation of directional EM measurements. 15. Method, according to claim 12, characterized by the fact that: the inversion of non-directional resistivity measurements of the observation zone involves i) adjusting at least one of formation dips and resistivity anisotropy of one or more layers of observation zone formation and ii) further adjust the horizontal resistivity of one or more observation zone formation layers based on adjustment to one of the formation dips and resistivity anisotropy of one or more observation zone formation layers. 16. Method, according to claim 1, characterized by the fact that: the determination of the observation zone profile involves the inversion of the non-directional resistivity measurements of the observation zone in combination with the directional EM measurements of the observation zone. 17. Method, according to claim 1, characterized by the fact that: the interpretation of directional EM measurements of the observation zone employs a formation model that characterizes the resistivity of one or more layers of the observation zone. 18. Method according to claim 17, characterized in that the formation model is initialized from at least one of: applying the logarithm to the square of the non-directional resistivity measurements of the observation zone; a profile of a previous observation zone; and a profile of a previous forecast zone. 19. Method, according to claim 17, characterized by the fact that: the formation model is initialized by thickening a region outside a forecast inversion window. 20. Method, according to claim 1, characterized by the fact that: the interpretation of the directional EM measurements of the prediction zone involves the inversion of the directional EM measurements of the prediction zone. 21. Method, according to claim 1, characterized by the fact that: the interpretation of the directional EM measurements of the prediction zone involves inversion operations that simultaneously solve the formation dip, as well as the horizontal resistivity and vertical resistivity or anisotropy of resistivity of one or more layers forming the prediction zone. 22. Method, according to claim 1, characterized by the fact that: the interpretation of directional EM measurements of the prediction zone involves inversion operations that assume a fixed formation dip of the prediction zone and solve horizontal resistivity and vertical resistivity or anisotropic resistivity of one or more layers forming the prediction zone. 23. Method, according to claim 1, characterized in that: the directional electromagnetic tool comprises at least one transmitter spaced from a plurality of receivers in both short and long spacings, wherein the at least one transmitter is configurable to emit high frequency electromagnetic radiation at a number of different predefined frequencies. 24. Method according to claim 1, characterized in that: the resistivity tool is selected from the group consisting of a propagation type resistivity tool, a laterological resistivity imaging tool and a triaxial resistivity tool. 25. Method for characterizing a geological formation traversed by an exploration well while drilling the exploration well using a resistivity tool and a directional electromagnetic (EM) tool, the method characterized in that it comprises: operating the resistivity tool to acquiring non-directional resistivity measurements of a formation observation zone and operating the electromagnetic tool to acquire directional EM measurements of a formation prediction zone; determine an observation zone profile by interpreting non-directional resistivity measurements of the observation zone, where the observation zone profile features at least formation dip as well as horizontal resistivity and vertical resistivity or resistivity anisotropy of one or more layers formation of the observation zone; and determining a prediction zone profile by interpreting directional EM measurements of the prediction zone, wherein the prediction zone profile characterizes at least formation dip as well as horizontal resistivity and vertical resistivity or resistivity anisotropy of one or more layers of prediction zone formation, in which interpretation of directional EM measurements uses a formation model that is initialized according to values from the interpretation profile of non-directional resistivity measurements. 26. Method according to claim 25, characterized in that it further comprises: updating an observation zone reservoir model based on the observation zone profile. 27. Method according to claim 25, characterized in that it further comprises: controlling the operation of a drill bit based on the profile of the observation zone. 28. Method according to claim 25, characterized in that it further comprises: updating a prediction zone reservoir model based on the prediction zone profile. 29. Method according to claim 25, characterized in that it further comprises: controlling the operation of a drill bit based on the profile of the prediction zone. 30. Method, according to claim 25, characterized by the fact that: the interpretation of non-directional resistivity measurements of the observation zone involves multiple distinct inversions of non-directional resistivity measurements of the observation zone. 31. Method, according to claim 25, characterized by the fact that: the interpretation of non-directional resistivity measurements of the observation zone involves a single inversion of non-directional resistivity measurements of the observation zone. 32. Method, according to claim 25, characterized by the fact that: the interpretation of the directional EM measurements of the prediction zone involves inversion operations that simultaneously solve the formation dip, as well as the horizontal resistivity and vertical resistivity or anisotropy of resistivity of one or more layers forming the prediction zone. 33. Method, according to claim 25, characterized by the fact that: the interpretation of directional EM measurements of the prediction zone involves inversion operations that assume a fixed formation dip of the prediction zone and solve horizontal resistivity and vertical resistivity or anisotropic resistivity of one or more layers forming the prediction zone.

Images