SYSTEMS AND METHODS FOR CLASSIFYING MECHANICAL QUALITY OF A SUBTERRANEAN FORMATION USING MEASUREMENTS OBTAINED DURING DRILLING

Abstract

Systems and methods for evaluating mechanical rock quality of subterranean formations include processing of mechanical rock property data to generate one or more corresponding mechanical quality metrics. The mechanical quality metrics may then be used, for example, to plan and execute subsequent drilling or completions (e.g., hydraulic fracturing) operations. In certain implementations, the mechanical rock property data is computed from drill bit vibration data collected during drilling of a wellbore through the subterranean formation.

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application is related to and claims priority under 35 U.S.C. § 119 from U.S. Provisional Application No. 62/804,629 entitled “Systems and Methods for Classifying Mechanical Quality of a Subterranean Formation Using Measurements Obtained During Drilling,” filed on Feb. 12, 2019, the entire contents of which are fully incorporated by reference herein for all purposes.

TECHNICAL FIELD

The present disclosure involves, among other things, generating and evaluating mechanical quality of subterranean rock formations, including evaluation of mechanical quality using mechanical rock property data obtained during drilling of a wellbore using measurement while drilling (MWD) techniques.

BACKGROUND AND INTRODUCTION

Drilling and completion of hydrocarbon-producing wellbores is dependent on the structure and properties of the subterranean formation through which the wellbore extends. For example and among other things, the strength and brittleness of the rock within the subterranean formation each contribute to how readily the subterranean formation can be drilled and/or hydraulically fractured. By understanding the composition and properties of the subterranean formation, drilling and completion operators can readily identify optimal wellbore paths, hydraulic fracturing locations, and the like to improve the efficiency with which a wellbore is formed and completed and to maximize subsequent production from the well.

The advantages associated with greater understanding of a subterranean formation's structure and properties often conflict with the time and costs associated with obtaining and processing the necessary data to gain such insights. Moreover, data related to a formation's structure and properties is often complex and therefore subject to misinterpretation.

It is with these problems and issues in mind, amongst others, that various aspects of the present disclosure were developed.

SUMMARY

In one aspect of the present disclosure, a method of classifying mechanical quality of material in subterranean formations is provided. The method includes, with a processor, processing signal data to compute displacements of a drill bit, the displacements resulting from interaction of the drill bit with a material of a subterranean formation during drilling of a wellbore through the subterranean formation. The method further includes processing the displacements to compute a property of the material along the wellbore and generating a mechanical quality metric for a location within the material based on the computed property.

In certain implementations, the signal data includes acoustical signal data, the signal data is stored on a non-transitory computer readable medium, and the signal data is measured using a sensor of a bottom hole assembly in operable communication with the non-transitory computer readable medium to store the signal data. In such implementations, processing the signal data comprises processing the acoustical signal data to obtain a property value representative of the property. In such implementations, the signal data may include signal data corresponding to at least one of an axial acceleration of the drill bit, a lateral acceleration of the drill bit, a torsional acceleration of the drill bit, a centripetal acceleration of the drill bit, or a rotational acceleration of the drill bit. Processing the acoustical signal data to obtain the property may also include computing at least one of Poisson's ratio or Young's modulus of elasticity using the acoustical signal data. In at least certain implementations, processing the signal data to compute the property value may include computing shear modulus using the signal data.

Generating the mechanical quality metric may include calculating, among other things, unconfined compressive strength (UCS), a brittleness metric, or an anisotropy metric at the one or more locations using the acoustical signal data.

In at least certain implementations, the method may further include performing at least one well operation based on the mechanical quality metric at the location, wherein the at least one well operation includes at least one of a drilling operation or a fracturing operation.

In another aspect of the present disclosure, a method of evaluating mechanical quality of subterranean formations is provided. The method includes, using a computing processor operatively coupled to a non-transitory computer readable memory, processing sensor signal data captured during drilling of a wellbore through a subterranean formation using a drill bit, by computing property values for a mechanical material property at a sequence of locations in the subterranean formation along the wellbore. The method further includes processing the property values to generate a mechanical quality metric for each location of the sequence of locations and storing the mechanical quality metrics in the non-transitory computer readable memory.

In at least certain implementations, the method further includes processing location data from a measurement while drilling (MWD) tool captured during drilling of the wellbore to correlate the property values to the sequence of locations. In such implementations, storing the mechanical quality data for the sequence of locations may further include aggregating the mechanical metrics for a consecutive subset of locations of the sequence of locations and storing an aggregated mechanical metric for a range corresponding to the consecutive subset of locations.

In other implementations, the sensor signal data is acoustical signal data.

In still other implementations, the sensor signal data is provided by one or more sensors of a bottom hole assembly.

In other implementations the mechanical material property includes Young's modulus of elasticity (YME). In such implementations, processing the property values to obtain the mechanical quality metric for each location of the sequence of locations may include calculating unconfined compressive strength (UCS) or a brittleness metric using the YME value of the location.

In another aspect of the present disclosure, a method of classifying mechanical quality of material in subterranean formations is provided. The method includes, using a computing processor operatively coupled to a non-transitory computer readable memory, computing property values for a mechanical material property at a sequence of locations in the subterranean formation along the wellbore. The method further includes processing the property values to generate a mechanical quality metric for each location of the sequence of locations and storing the mechanical quality metrics in the non-transitory computer readable memory.

In certain implementations, generating the mechanical quality metric includes calculating at least one of unconfined compressive strength (UCS), a brittleness metric, or an anisotropy metric at each location of the sequence of locations.

In other implementations, the method further includes performing at least one well operation based on the mechanical quality metrics.

In yet another aspect of the present disclosure, a tangible, non-transitory, computer-readable media having instructions encoded thereon is provided. The instructions, when executed by a processor, being operable to compute property values for a mechanical material property at a sequence of locations in the subterranean formation along the wellbore, process the property values to generate a mechanical quality metric for each location of the sequence of locations, and store the mechanical quality metrics in a non-transitory computer readable memory.

In certain implementations, the property values for the mechanical material property are computed from vibration of a drill bit measured during interaction of the drill bit with the subterranean formation while drilling the wellbore.

BRIEF DESCRIPTION OF THE FIGURES

FIGS. 1A-1B Illustrate reservoir-to-well connectivity where brittle rocks are generally associated with larger fracture creation and better proppant support that is more permeable than ductile rock that produces smaller, less productive fractures which are prone to rapid compaction and closure and are less permeable.

FIGS. 2A 2B are diagrams of a drill bit assembly including sensors for measuring bit accelerations and forces on the bit, and which includes at least one processing unit and tangible storage media in which to store acceleration and/or force data, and which may also store processed acceleration and/or force data of the drill bits interaction with a formation while drilling.

FIGS. 3A-3C illustrate the cutting action of a drill bit, stick slip fracturing of a rock formation, torque on bit and weigh on bit forces, and related torque and displacement curves.

FIG. 4 illustrates root mean square computations for axial acceleration of the bit as used in computing rock properties.

FIG. 5A is a rotary displacement spectra obtained from measurement while drilling data.

FIG. 5B is a diagram illustrating lateral and rotary acceleration of the bit, while cutting, and useful in computing mechanical rock properties among other advantages.

FIG. 6 is a diagram illustrating drill bit behavior when encountering a mechanical discontinuity or geological boundary, the cutting face of the drill bit may change its orientation, as detectable from acceleration data, in response to the orientation and stresses acting on the heterogeneity.

FIG. 7 is a graph of rock strength curves computed from torque and penetration per revolution information based on measurement while drilling data.

FIG. 8A illustrates stress strain relationships based on an orientation of a well relative to a transverse isotropic axis of material symmetry and mechanisms whereby measurement while drilling techniques may be used to compute elastic coefficients, from which temporal and/or spatial variations in one or a combination of more of the measurements obtained from the geophysical signal processing techniques are used to identify the nature and occurrence of fractures, fracture swarms and other mechanical discontinuities (boundaries) such as bedding planes and/or faults that offset or otherwise separate rock formations with different mechanical rock properties.

FIG. 8B is a stress strain curve reflecting the relationships of FIG. 8A.

FIG. 9A illustrates a constitutive stress strain relationship for an axis of material symmetry parallel a well axis, and computations using torque on bit, weigh on bit and axial displacement to obtain elastic coefficients from measurement of bit vibration and/or forces acting on the bit.

FIG. 9B illustrates a constitutive stress strain relationship for an axis of material symmetry perpendicular a well axis, and computations using torque on bit, weigh on bit and rotary displacement to obtain elastic coefficients from measurement of bit vibration and/or forces acting on the bit.

FIGS. 10A-10C illustrate a constitutive stress strain relationships for an axis of material parallel and perpendicular a well axis, using acceleration data.

FIG. 11 illustrates constitutive stress strain relationships for an axis of material symmetry parallel a well axis and axis of symmetry perpendicular a well axis, with Poisson's ratio and Young's modulus of elasticity computed from measurement while drilling data of torque on bit, weigh on bit, and acceleration data to obtain lateral and axial displacement.

FIG. 12 illustrates stress strain curves obtained by the methods discussed herein.

FIG. 13 illustrates two relative curves of Poisson's ratio for a vertical well, with cross over points identifying likely preexisting fractures in the fracture detection log.

FIG. 14A illustrates two relative curves of Young's modulus of elasticity for a horizontal well.

FIG. 14B illustrates two relative curves of Poisson's ratio for the same horizontal well, with cross over points identifying mechanical rock properties indicative of preexisting fractures.

FIG. 15 is a diagram illustrating the forces acting on a formation in connection with a drill bit and drilling fluid system when conducting drilling operations that are sufficient to overcome the failure criteria of a pre-existing fault, and reactivation of a fault of pre-existing fracture, which can be evidenced by extracting a signal from the drilling vibrations that is related to a microseismic event with attendant primary, compressional (P) and secondary, or shear (S) arrivals.

FIG. 16 is a flow diagram of a method of obtaining mechanical rock properties of a formation proximate a well bore from measurements of bit behavior taken while drilling.

FIG. 17 is a diagram illustrating stress and strain relationships developed from acoustical signals captured by a three axis accelerometer positioned on a measurement while drilling assembly of a drill.

FIG. 18 is a schematic illustration of a bottom hole assembly including sensors for measuring bit accelerations and forces on a bit.

FIG. 19 is a flow diagram depicting a method of obtaining mechanical rock quality information for a subterranean formation and performing one or more well-related operations based on such information.

FIG. 20 is an example visualization of mechanical rock quality information for a wellbore.

FIG. 21 is a special purpose computer programmed with instructions to execute methods discussed herein.

DETAILED DESCRIPTION

The present disclosure involves an inventive way of evaluating mechanical rock quality in a subterranean formation. In at least some implementations, the mechanical rock quality may correspond to a susceptibility of the rock to drilling, fracturing, or other operations related to formation and completion of a hydrocarbon producing well. In certain implementations, mechanical quality is determined from mechanical rock property data obtained from drilling vibrations generated by the deformation of the rock of the subterranean formation in response to forces acting on the rock formation related to a drill bit and drilling fluid system.

As will be appreciated from the devices, systems and methods provided and disclosed herein, aspects of the present disclosure may also involve the determination of absolute values of mechanical rock properties, which becomes possible from the application and use of (i) the forces or accelerations of the drill bit and (ii) the displacements or motions of the drill bit such as those obtained from recording the near-bit mechanical drilling vibrations in relation to the drill bit interacting with the rock formation. In one specific example, drilling vibrations experienced by the drill bit from its breaking of rock while drilling, propagate as acoustical signals that are translated into data by accelerometers or other sensors positioned proximate the drill bit. The acoustical signals are translated into mechanical rock properties according to the techniques discussed herein. Further, by first drilling through some known media, such as cement in the well, the system may capture data and generate scalars that can be used to transform derived mechanical rock properties, from data capture in the same manner, for an unknown media, such as a formation being drilled through, into absolute values of mechanical rock properties for that formation.

Aspects of the techniques and measurements discussed herein, in whole or in part, may be achieved using downhole tools that are simple and rugged, allowing for a magnitude in order reduction in logging cost to characterize near-wellbore rock mechanical properties, which may further include identifying and characterizing intersected existing fracture locations. The low cost to log a well, which may typically be less than 0.5% of the total well cost, allows for widespread use of the technique. Detailed knowledge of rock property variability along a wellbore allows for grouping like-for-like rock types in variable length stages, avoiding losing reserves due to a lack of fracture initiation relative to mixed rock strength stages. Scaling mechanical rock property computations to further obtain absolute values of such mechanical rock property computations may further enhance the methods and systems.

Further elaboration of the techniques discussed herein describe how mechanical rock properties, and in particular elastic coefficients of a rock formation, can be determined through the application and use of innovative, new stress-strain relationships that systematically relate measurements of the forces acting on rock formation in connection with the drill bit and drilling fluid system (stress) to the variations in the drilling vibrations generated by the deformation of the rock formation in response to the cutting action of the bit (strain).

The elastic coefficients can be used, in general to describe the deformation of a rock formation in response to the forces acting on the rock formation, and in particular, to predict the deformation of a hydrocarbon bearing formation in response to the forces acting on the formation where the forces are fluid pressures generated during the emplacement of hydraulic fractures in connection with a hydraulic fracture stimulation treatment along a horizontal well.

The processing of drilling vibrations when recorded using sensors deployed in a borehole in connection with a bottom hole assembly (BHA) according to aspects of the present disclosure, can provide measurements of mechanical rock properties including the nature and occurrence of mechanical discontinuities, such as pre-existing fractures, which can be used to target sections of the well where the rock properties are conducive to economical hydraulic stimulation and to avoid sections that are viewed as sub-commercial, where the rock properties are not conducive to economical hydraulic stimulation.

The present disclosure generally relates to the production of commercially significant hydrocarbons from oilfield drilling operations and completion operations. Recent unconventional resource development has identified a need for economical determination of rock properties and natural fracture swarm locations along a horizontal well in order to optimize the location and intensity of hydraulic stimulation treatments.

The techniques described in this disclosure will provide new information for selecting hydrocarbon bearing zones by differentiating between brittle rocks generally associated with larger fracture creation and better proppant support that is more permeable than ductile rock that produces smaller, less productive fractures which are prone to rapid compaction and closure and are less permeable. Natural fracture identification also refines the process of hydraulic stimulation optimization by providing direct measurement of zones that offer higher permeability and higher hydrocarbon productivity.

Physical Basis for Obtaining Mechanical Rock Properties from Drilling Vibrations

Aspects of the present disclosure may involve methodologies that use broad band measurements (e.g., continuous, high resolution) of drilling vibrations and drilling dynamics data taken proximate the drill while conducting drilling operations bit to log the mechanical properties of a rock formation. Such mechanical properties may subsequently be used to compute or otherwise generate mechanical quality metrics for the formation, which may indicate the susceptibility of the formation (or parts of the formation) to various well-related operations, such as drilling or hydraulic fracturing.

Drilling vibrations generated by the deformation and failure of a rock formation are generally related to the mechanical properties of the rock being drilled. It is generally understood that the depth of cut or the tooth penetration into the rock is inversely related to the strength of the rock. Higher amplitude drilling vibrations occur in rocks that undergo a greater depth of cut and deeper tooth penetration in response to the forces acting on the formation in connection with the drill bit and drilling fluid system, whereas lower amplitude drilling vibrations occur in rocks that undergo relatively lower depth of cut and lesser tooth penetration. Increased depth of cut indicate the bit is moving into an area of lesser relative mechanical rock strength, and decreased relative drilling vibrations indicate the bit is moving into an area of greater relative rock strength all other things being equal.

Generally speaking, rock formations that take a relatively long time to drill through or where the rate of penetration is slow are generally referred to strong or hard formations and have a lesser depth of cut in relation to rock formations that are relatively weaker and less rigid. These basic principles have enabled the application and use of techniques that take measurements of the hardness of a rock formation by forcing a tool into a rock to make an indentation where the depth of the indentation relative to the force applied is used to obtain a hardness characteristic that is essentially a mechanical property of a rock formation.

The presence of mechanical discontinuities, such as pre-existing fractures, and geological boundaries, such as faults, in a rock formation generally act to weaken the rock formation. Fractured rock formations, are generally weaker and less rigid than intact, un-fractured or stiff or otherwise competent rock formations. As the drill bit encounters fractures in a rock formation the tooth penetration or depth of cut and subsequently the drilling vibrations will increase, because the rock formation is less rigid because it has been weakened by the presence of fractures. Stated differently, as the drill moves into and through existing fractures, the measured mechanical rock strength will decrease relative to the same rock formation without a fracture or with lesser fractures, for example.

Additional details and applications relating to the calculation of mechanical rock property data from drill bit vibrations are further discussed in U.S. Pat. Nos. 10,132,162; 9,644,039; 10,519,769; and 10,544,673 and U.S. patent application Ser. No. 15/263,278, the entirety of which are incorporated herein by reference.

General Description of Fracture Identification

Signal processing techniques are used to process the drilling data to identify locations where the changes in the drilling vibration indicate that the drill bit has encountered a mechanical discontinuity or geological boundary. If the changes in the drilling vibration as expressed through the results of the geophysical signal processing techniques are rapid and discrete in both space and time, and then return back to a long-term trend or the levels that were recorded prior to the change in the drilling vibrations, then it would indicate that that the drill bit has encountered and crossed a discrete mechanical discontinuity because mechanical rock properties that are discrete in both space and time are uniquely separated from the mechanical properties of a rock formation such as would be in the case of a drill bit penetrating a fracture face. If the changes in drilling vibration are rapid and discrete and continue over a short interval, then that would indicate multiple fractures or a swarm of fractures has been encountered.

If the signal processing techniques indicate that the changes in the drilling vibration are rapid, but then do not revert back to the level prior to the change and instead carry on at a new, significantly different level, then that indicates a mechanical boundary where the mechanical boundary that separates or offsets two different rock formations such as a bedding plane and or fault has been encountered and crossed. Whether or not the boundary is related to a fault or a bedding plane depends on the inclination of the bit with respect to the orientation of the stratigraphy of the rock formation being drilled. Other information may also be used to determine whether or not the mechanical boundary was a bedding-plane fault or bedding plane that acted as a zone of weakness that had experienced measurable displacement in the past.

The description provides an approach to evidence the presence of fractures, fracture swarms and other mechanical discontinuities such as faults and bedding planes that offset or otherwise separate rock formations with different rock properties. The approach uses geophysical signal processing techniques that are sensitive to changes in the drilling vibrations where the changes are relative to some baseline, such as a normalized preceding set of drilling vibration data, and whether or not the changes are discrete and then return back to the level prior to the change or are maintained at a new level that is different than the level observed prior to the change.

The following technique disclosed below further elaborates on the outlined principles to provide a general, independent approach to specify the mechanical properties of a rock formation by processing the drilling vibrations in relation to the forces acting on the formation in connection with a drill bit and, in some instances, a drilling fluid system, which includes the mud motor and the drilling fluid, including mud, that turns the motor. This approach specifies the mechanical properties of a rock formation through the application and use of innovative, new stress-strain relationships, among other advances.

Because the cutting depth or the penetration of the bit tooth into the formation is a measurement of displacement, the drilling vibrations describe the strain experienced by a rock formation in response to the cutting action of the bit where greater cutting depths and a greater penetration of the bit tooth relative to the same volume of rock result in a higher strain. Strain is understood to describe a change in volume of a rock under some force. In some aspects, strain is reflected by axial displacement of the bit per turn of the bit. In one example, axial displacement is computed as a double integral of the accelerometer data (axial) for one revolution of the bit to yield a distance measurement for one revolution (or some other known number of turns). If this strain is calculated with respect to time, then the drilling vibrations can be used to describe the strain rate. The converse with regards to the strain and the rate of strain is also held to be evident.

Through the techniques described herein, the drilling vibration characteristics, which may be supplemented with drilling dynamics data including forces on the bit, such as torque on bit and weight on bit, are translated into mechanical properties of a rock formation. The depth of cut, as obtained based on vibration assessment, may be normalized against direct weight on bit and/or torque on bit measurements, or weight on bit or torque on bit measurements extrapolated from vibration information.

General Stress-Strain Relationships

Stress-strain relationships are established by systematically relating forces acting on the formation in connection with the drill bit and drilling fluid system to the geophysical signal processing of drilling vibrations generated by the deformation of the rock in response to the cutting action of the bit. This approach allows elastic coefficients (K) to be derived in accordance with the following equation where (e) is the general deformation (strain) of a rock formation in response to the forces acting on a rock formation (S) (stress).

S=K e

In accordance with some implementations set out herein, strain (the motion or displacement of the bit is obtained by signal processing the drilling vibrations as they are transmitted along the drilling assembly by sensors deployed in a borehole in connection with a BHA. Stress, in accordance with some implementations set out herein, is obtained from either (i) downhole or (ii) surface measurements of torque and/or weight on bit, or (iii) in another example, the accelerations of the bit are related to forces on bit where it understood that the acceleration is a representation of force per unit mass. It should be appreciated that forces can be converted to stresses with knowledge of the effective contact area of the bit and formation, and the effective rock volume the bit is acting on. Conversely, forces can be substituted for stresses with the understanding that a geometric correction in relation to the effective contact area is required to obtain absolute values for the mechanical rock properties. One example of such a contact area is the area of the bit.

The equations of linear elasticity are useful for describing the relationship between the changes in shape and position of a material in relation to the forces acting on the material. Such stress-strain relationships are known in general as Hooke's law where the coupling of the stress-strain relationship behavior is described through a matrix of coefficients whose values depend on the conditions used to load the material in relation to the structural symmetry of the material being loaded. These coefficients are colloquially known as the cij's and can be arranged in well-known and convenient forms to represent Young's modulus of elasticity (YME) and Poisson's ratio (PR). In one example of the technique presented here, the YME and PR values are systematically determined by the loading conditions of the bit in relation to the axis of material symmetry used to describe the rock formation.

In one specific implementation, the constitutive equations of linear elasticity are uniquely expressed through the application and use of MWD data to (i) populate the variables of the constitutive equations of linear elasticity and (ii) undertake an analysis of the constitutive equations to obtain measurements of near-wellbore mechanical rock properties of (a) Young's modulus of elasticity and (b) Poisson's ratio. Further, variations in the mechanical rock properties (e.g., YME and PR) are used to identify the nature and occurrence of mechanical boundaries or discontinuities in the subsurface such as fractures.

More specifically, a technique to determine near-wellbore mechanical rock properties, YME and PR, from MWD data may involve processing measurements of the weight on bit (WOB), torque on bit (TOB), Annular fluid pressure (Ap), angular bit speed (RPM) and components of motion describing the acceleration of the bit, including axial, and the rotary or tangential accelerations to (i) obtain sets of MWD data corresponding to known temporal and spatial positions along the borehole, (ii) calculate the forces acting on the rock formation in connection with the drilling apparatus and drilling fluids, (iii) calculate the displacements of the bit as it is accommodated by the deformation of the rock formation, (iv) inform the terms and loading conditions (variables) of a linear, elastic stress-strain relationship that describes the constitutive behavior of the rock formation in relation to the orientation of the well and (v) using the constitutive linear elastic equations as determined through the application and use of the MWD data to calculate the aforementioned mechanical rock properties, (YME and PR) and (vi) analyzing the YME and PR with respect to the axis of material symmetry in relation to the orientation of the well to identify the nature and occurrence of mechanical boundaries and discontinuities such as fractures and bedding planes among other things.

Certain implementations of the present disclosure involve techniques to specify, in general, the mechanical properties of a rock formation from an analysis of drilling vibrations generated by the cutting action of the bit and the deformation of the formation in response to the forces acting on the rock formation in connection with the drill bit and drilling fluid system while conducting drilling operations. Deformation may include elastic deformation, plastic deformation, and failure of the rock, which may be considered fracturing. Stated differently, aspects of the present disclosure involve obtaining information associated with the drilling of a borehole, while drilling, to identify mechanical rock properties of the formation being drilled. Such mechanical rock properties may be used, in some examples, to identify the presence of natural fractures or rock properties more or less susceptible to stimulation techniques. For example, knowing mechanical rock properties along a borehole or the presence of natural fractures along a borehole may be used to optimize hydraulic fracturing operations by focusing such fracturing on areas where it will be most effective, among other advantages. Mechanical rock properties may include elastic coefficients (e.g., the cij's), strength measurements such as initial yield strength, peak compressive strength, tensile strength YME, PR, shear modulus, bulk modulus, strain-hardening exponents, Thomson coefficients and other mechanical rock properties.

The Nature of Elastic Coefficients as They Pertain to Fractures

The mechanics of drilling a well provide a natural, in-situ, means to measure the deformation of a rock formation and gather data suitable for determining mechanical rock properties, because the penetration of the drill bit is in and of itself accommodated by repeatedly fracturing the rock formation by using the bit to generate forces on the rock formation that are sufficient to overcome the failure strength of the rock, measurements of such in relation to the techniques described here may be used in predictable ways to determine the presence of natural (in situ) fractures, fracture swarms (cluster of fractures), bedding planes, fault boundaries, and other information. In some instances, variations in mechanical rock properties are used to identify fractures, bedding planes and the like.

Elastic coefficients that describe a relatively large deformation in response to the forces acting on a rock formation indicate the rock formation is weaker and less rigid. Therefore mapping the spatial variations of the elastic coefficients provides information where there are zones of weakness in the rock formation. If the nature and occurrence of the zones of weakness in the rock formation as evidenced by changes in the elastic coefficients are localized in space or otherwise discrete relative to the surrounding elastic coefficients that would indicate the presence of a fracture or other mechanical discontinuity. Systematic changes in the spatial distribution of the elastic coefficients as are derived by the techniques herein are used to identify mechanical discontinuities and geological boundaries of rock formations such as bedding planes or faults that act to separate or offset rock formation with different rock properties, where the differences in rock properties are evidenced by the nature and distribution of the elastic coefficients.

As will be understood from the present disclosure, mechanical strength and deformation of the reservoir rock influences fracture creation, propagation and ability to maintain fracture permeability. FIG. 1 is a simplified diagram illustrating the difference between fractures induced in a relatively brittle rock formation, and which may be include naturally occurring fractures, versus fractures induced in a relatively ductile rock formation, which may include fewer or no natural fractures. As illustrated in FIG. 1A, a horizontal section 10 of a borehole has been drilled through relatively brittle rock 12 and hydraulically fractured. In contrast, FIG. 1B illustrates a horizontal section 14 of a borehole drilled through relatively ductile rock 16 and hydraulically fractured. The fractures 18 created in the relatively brittle rock tend to penetrate deeper into the reservoir than the fracture 20 in ductile rock. Moreover, reservoir rock trending to the brittle end of the normal range tends to have higher initial production rates and lower decline rates. The techniques described in this disclosure will provide new information for selecting hydrocarbon bearing zones by differentiating between brittle rocks generally associated with larger fracture creation and better proppant support that is more permeable than ductile rock that produces smaller, less productive fractures which are prone to rapid compaction and closure and are less permeable. Similarly, techniques discussed herein may also identify areas where natural fractures may exist provide similar advantages as brittle rock. Generally speaking, as will be understood from the disclosure, various mechanical rock properties discussed herein will provide mechanisms whereby rock formations may be characterized, along the well bore, as to the relative brittleness or ductileness, or the relative susceptibility to stimulation techniques along the formation, which may include the identification of existing fractures or at least rock properties indicative of existing fractures.

In at least some implementations of the present disclosure, the mechanical rock properties of the rock formation and characteristics of the rock formation based on such properties may be used to assign a classification to the rock formation, such as a quality metric or quality classification to portions of the rock formation. For example, in at least one example implementation, characteristics of the rock formation may be used to classify certain portions of the rock formation as having “good” or “high” quality (e.g., high brittleness and low strength or other characteristics indicating high susceptibility to subsequent fracturing operations); “bad” or “poor ” quality (e.g., low brittleness and high strength indicating relatively low susceptibility to fracturing operations), or some combination thereof (e.g., in cases where only one of high brittleness and low strength is observed).

Data Acquisition Techniques

FIG. 2B is a diagram of a bottom hole assembly portion 20 of a drill string where the bottom hole assembly includes a drill bit 22, a mud motor 24, a bit sub 26 including various measurement components positioned between the drill bit and the mud motor, and sections of pipe 28 within a horizontal section 30 of a borehole, also referred to herein as a well bore. The vibration data used in the described methodologies may be recorded as close to the source (drill bit) as practical to avoid attenuation through the bottom hole assembly. In one example, vibrations are translated into data from acoustical signals interacting with accelerometers. Similar vibrations data may be collected when drilling through concrete or some other known media, in the vertical borehole section, as discussed in more detail below. One possible location for recording is directly behind the drill bit and ahead of the mud motor using the bit sub, although multiple bit subs may be used along the drill string for geophysical processing of the desired signal. Drilling a wellbore involves using a portion of the weight of the drill string, known as weight on bit (WOB), to push the drill bit into a formation 30. The rotating force on the drill bit, known as torque on bit (TOB), can come from the surface or from a mud motor close to the drill bit. When using a mud motor, drilling mud is pumped down the drill string until it encounters the power drive section of the mud motor where a portion of the mud pressure and flow is converted into a rotational force, which is mechanically coupled to the bit to thereby place rotational torque on the bit 22 to turn the bit. The rotational force on the bit can also be augmented by or come exclusively from mechanisms at the surface on the drilling rig.

The objective of the drilling process is to break the rock down into fragments that are small enough that they can be lifted and evacuated from the wellbore with drilling fluids in order to continue to accommodate the forwards motion of the bit. It should be noted that the action of the drill bit on a rock formation causes the fracturing of the rock formation, which fracturing is experienced as vibrations of the drill, along the borehole to drill the hole, and in the formation immediately adjacent the borehole. Moreover, the drill may encounter existing fractures 34 while drilling. Hydraulic fracturing, in contrast, is a process that occurs during the completion phase by injecting fluid into the borehole, typically with perforation clusters 22 in the casing, to initiate fractures 18/20 into the formation surrounding the bore hole, as illustrated in FIG. 1.

In the illustrated diagram, the bit sub 26 is shown between the bit and the mud motor. The bit sub is a cylindrical component that is operably coupled between the mud motor 24 and the drill bit 22 in a way that allows the mud motor to turn the bit. The bit sub includes a housing, typically in a cylindrical shape, or another mechanism to support various possible measurement components 36 including strain gauges, one or more accelerometers, pressure sensors, which may measure the pressure of the mud flow, temperature sensors which may measure the circulating temperature of the mud or other temperatures and which may be used to provide correction or offset of measurements or calculations that vary with temperature, gyroscopes which may be used to measure inclination and/or directional changes of the bit and string, and/or other components to measure or derive the information discussed herein.

In one example, as shown in FIGS. 2A and 2B, the strain gauges are mounted on the bit sub to determine torque on the bit and the weight on the bit (the force turning the bit and the force pushing the bit into the rock formation). Various possible ways of mounting the strain gauges, or combinations of strain gauges, are possible. Additionally, as shown in FIG. 2A, which is a representative front view of the bit 22, accelerometers are placed to measure axial, rotary, torsional (also referred to herein as centripetal), and/or lateral acceleration of the bit. Note, the bit axis is in the center of the circle, whereas axial acceleration may be measured somewhat offset from the axis depending on the placement of the accelerometer. Acceleration measurement may be accomplished by using one or more multi-axis accelerometers. The bit sub, or other such component, may also include a processor and memory to store computer executable instructions to implement various possible methodologies, and possibly preprocess data, as well as a power source which may be one or more batteries. Data storage, such as the memory or other data storage, is also provided to store the collected data. The measurement components, alone or in various possible combinations, may be provided in other locations of the drill string in the general proximity of the drill bit.

Geomechanics in Relation to the Cutting Action of the Bit

FIGS. 3A-3C are a sequence of diagrams illustrating a close up view of a cutter 32 portion of a drill bit in a borehole, slipping, sticking on a portion of rock, and then slipping loose when the forces on the bit are sufficient to overcome the rock causing the rock to fracture and the bit to rotate—collectively referred to as stick slip behavior. The rock deformation mode for a PDC bit is shearing as opposed to a roller cone bit which is punching. Models that describe drilling behavior are in a large part informed on the mechanics of drilling with a roller cone bit and while these models have been extended for the application and use of PDC bits, they suffer uniquely from their inherent inability to reconcile the fundamentally different nature of rock deformation. As will be appreciated from this disclosure, the innovative, new techniques disclosed here seek to advance the application and use of PDC bits, as well as other bits, to characterize mechanical rock properties and in particular for the identification of the nature and occurrences of fractures. The figures describe the depth of cut in relationship to the area and displacement of the fractures created in response to the forces acting on the bit. The cutting action of a particular type of bit but should not be construed to limit implementations of the present disclosure in the use of other types of drill bits that generate acoustic emissions from rock failure in response to the forces acting on the geometry and configuration of the bit.

More specifically, as the bit turns, the interaction of the bit with the rock formation at any instant in time, produces a complex distribution of forces acting on the formation in connection with the bit and drilling fluid system (e.g., the mud motor) where the orientation and magnitudes of the forces acting on the rock formation are related to the configuration and geometry of the cutters on the bit. Generally speaking, drilling is not a smooth and consistent process. Instead, depending many things including the axial force on the bit, rotational torque on the bit, rock properties, and presence or absence or existing fractures, the bit cuts, gouges, spins, snags, and otherwise drills the borehole in a very complicated and varying fashion. In some instances, the complex distribution of forces acting on the formation is insufficient to initially overcome the strength of the rock formation in relation to the cutting action of the bit and the bit will stop rotating or stick.

As illustrated in FIGS. 3A and 3B, as the cutter begins to stick, the torque applied to the bit increases from a relatively steady value. As the forces, such as the illustrated torque on bit, applied to the bit change either through manual or automated interaction with the surface drilling apparatus or through the non-linear feedback of elastic energy stored within the drilling string or some combination of both, a new weight on the bit (WOB) and torque are delivered to the bit. These forces on one or more particular cutters will continue to load the rock elastically until such point (i) the rock begins to deform plastically and the deformation is concentrated along fracture planes and (ii) when those forces provides a sufficient distribution of forces to overcome the failure strength of the rock formation, the bit will turn and rock, often snapping loose, and drilling will continue. As shown in FIG. 3C, when the forces overcome the rock, the torque will dramatically drop to the relatively steady value, until the bit sticks again. Such a stick slip action may occur at varying frequencies and displacements and make happen one or more times per revolution of the bit per cutter on the bit, thereby resulting in many such cutting behaviors each revolution of the bit.

Regardless of whether and the extent of stick slip behavior is experienced, deformation and failure of the rock cause the bit to vibrate. During rock deformation and in particular when the bit overcomes the rock strength at failure, stored elastic strain energy is released in the form of acoustic emissions. In some instances, the bit is fracturing rock and may intersect fractures and existing mechanical discontinuities. In some instances, the bit may reactivate existing fractures, which may itself generate a distinct acoustical signal in the form of an induced bit vibration.

On the Processing of Drilling Vibrations

As introduced above, drill bits 22 typically include many cutters 32 arranged with a geometry and configuration designed to generate sufficient forces to overcome the failure strength of the rock formation 30 based on the nature of the rock formation expected to be encountered when drilling a well. During a single rotation of the bit, at least one, but typically many of the cutters will overcome the failure strength of the rock formation and produce a plethora of acoustical emissions related to the scraping, cutting, fracturing, and other interactions between the bit and the rock formation. The tool may include various possible mechanisms, including a reed switch or gyroscopes, which measure revolutions per minute and provide information of each rotation of the bit. There are as many as 30 cutter heads, so each rotation may cause hundreds of acoustic pulses.

Because of the stochastic nature of acoustic emissions or in relation to the nearly simultaneous initiation and propagation of multiple fractures at the cutting face, the implementations discussed herein may use statistical methods and signal analysis tools. In one possible methodology, the RMS measurement technique determines the energy of the signal despite the shape of its waveform. This is important because the many simultaneous events will create complex waveforms with constructive and destructive interferences.

Given the stochastic nature of the acoustic emissions generated by the cutting action of the bit, it is expected that the more times the bit turns per unit time, the higher the rate of acoustic emissions. RMS levels obtained for a time window that underwent two revolutions of the bit would be expected to have higher RMS levels than those obtained during one revolution of the bit in the same time window, all other things being equal.

Measurements while drilling show that the bit speed can vary wildly and erratically during drilling operations. In some instances the bit can completely stick and then slip again as more force is gradually applied. If the stick-slip behavior of the drill bit is not accounted for in the geophysical signal processing, then the variations in the fracture measurements may be confused by with variations in the bit speed and not variations in the fracturing behavior of the rock. When the time window used to measure the RMS level of the signals that have been extracted from the drilling vibrations is normalized with respect to the bit speed, where the bit speed is recording using a gyro to sample the changes in position of the bit with respect to time or a magnet which is used to inform the position of the bit with respect to time, this normalized measurement is understood to provide a level of the acoustical emission activity generated by the fracturing of a rock formation in relation to the cutting action of the bit. The time window where data is gathered may be tied directly to bit rotation by having a time window set based on counts of the bit revolution. In other instances, the time window may be set, and the energy (from vibrations) may be normalized to account for some set turns of the bit, such as one turn of the bit.

The implementation of the technique as described uses signal processing techniques, such as Fourier transforms, bandpass filtering or other filtering, or combinations thereof, to calculate the motion of the bit and the forces on the bit from the amplitudes and frequencies of the acoustical signals recorded by the MWD apparatus (e.g. bit sub 22) that are generated in response to the cutting action of the bit (e.g., the drilling vibrations propagate up the drill string as acoustic waves sometimes referred to as collar waves or tool mode, where they are recorded as acoustical signals by the MWD apparatus). Hence, accelerometers capturing acceleration data can be used to obtain information of the forces on the bit. Forces on the bit may also be measured using strain gauges, in one possible implementation. The motion of the bit, which may be recognized as displacements, and the forces on the bit are used to populate a stress/strain relationship that allows the computation of mechanical rock properties. Mechanical rock properties may be analyzed relative to a baseline (such as an average over some wellbore distance) to identify locations where the mechanical properties of a rock formation change as the bit encounters a mechanical discontinuity or other geological discontinuity all other things being equal. In some instances, rock properties may be used to identify such locations through computations based on assumptions of the rock formation, and comparisons thereof, or otherwise.

Referring to FIG. 4, to account for the effects of the drilling efficiencies and the possibility of stick-slip behavior while conducting drilling operations, in one specific implementation, implementations of the present disclosure may use measurements of the drilling efficiencies, such as revolutions per minute (RPM) to normalize the geophysical signal processing of the drilling vibrations in order to compare the results of the signal processing along the length of the borehole in accordance with the implementations provided. It is understood through the normalization that the variations in the signal levels, such as the RMS levels, are now corrected in account of the changes in drilling efficiencies along the trajectory of the borehole. .

As such, the RMS levels, obtained as shown above, for example, are related to changes in the mechanical rock properties while the drill intersects areas of differing properties. Stated differently, it may be helpful to compensate or normalize for changes in the rotational speed of the bit (RPM) by using a time window commensurate with the rotation of the bit. Such normalization may also be useful to correct the geophysical signal processing in accordance with drilling operations where the drill string is rotated from surface in conjunction with the mud motor turning the bit, as opposed to situations where the mud motor is operating but the string is sliding not being rotated from the surface.

In a further elaboration of implementations of the present disclosure, the signal processing techniques systematically relate measurements of the forces acting on a rock formation in connection with the drill bit and drilling fluid system (stress) to the variations in the fracturing of a rock formation in response to the cutting action of the bit (strain) to obtain innovative, new stress-strain relationships where the application and use of the stress-strain relationships allow for the derivation of elastic coefficients for the stress-strain relationships. Relative variations in either one or a combination of the elastic coefficients may be used to identify the nature and occurrence of fractures, fracture swarms and other mechanical discontinuities and geological boundaries such as bedding planes and/or faults that offset or otherwise separate rock formations with different mechanical rock properties.

Accounting for the drilling efficiencies is also a consideration in the implementation of the approach, because the wear on the drill bit as the rock formation is drilled will change the configuration and geometry of the cutters on the drill bit. The mechanical wear of the drill bit will affect the distribution of the forces acting on the formation in connection with the drill bit and as the well is progressed, the manual or automated application of forces on the drilling apparatus change to account for the wear and tear of the bit.

The application and use of the stress-strain technique according the implementations of the present disclosure employed is understood to normalize the effects of reduced drilling efficiencies caused by bit wear on the derivation of the mechanical rock properties because the forces used in the stress-strain relationship are obtained from the forces acting on the bit, where the forces needed to overcome the strength of the rock are increased in relation to the penetration or depth of cut of the worn bit tooth into the formation.

The acoustical emissions associated with the deformation and failure of the rock formation while drilling are generally too minute and/or too attenuated by the intervening rock to be detectable at the surface (which may be hundreds or thousands of feet above the borehole). Because of the amount of energy released is generally expected to be slight and of relatively high frequency, the radiated waves are best viewed when transmitted from the cutting face through the bit and bottom hole assembly where they propagate along the drill string through acoustically conductive steel as a direct tool arrival and contribute to the vibration of the drill string. The drilling vibrations can be recorded on instrumentation that is sensitive to their nature and presence. Stated differently, one aspect of the present disclosure involves a drilling tool assembly including sensors and processing electronics (e.g., the accelerometers and/or strain gauges in the bit sub 26 proximate the bit 22) that are positioned to detect and record the radiated waves from the drilling induced fracturing, which may further involve identification and/or characterization of existing mechanical discontinuities, such as fractures or geological boundaries, such as faults or bedding planes.

Data Logging

In one specific implementation, a form of measurement while drilling (MWD) system or tool is employed. The MWD system uses sensors designed to measures vibration. The MWD system may also measure forces on the bit, and other parameters such as bit speed, which may be expressed as revolutions per minute, the fluid pressures, and temperature of the drilling mud or environment proximate the bit sub. The system may also include gyroscopes to obtain the orientation of the cutting face of the drill bit, in some implementations. In one specific embodiment, the MWD tool includes at least one receiver, which may include accelerometers mounted on or proximate the bottom hole assembly to record the drilling vibrations and associated acoustic emissions. In some implementations, the MWD may further include electrical, mechanical, and/or other filtering mechanisms to processes the data to remove unwanted noise or to record the data without unwanted noise. In certain instances, stages of filtering may be applied both prior to recording, and after recording but prior to processing, to remove unwanted data, or as much as necessary or possible. In an alternative enablement, the signals may be transmitted to the surface for storage and processing. In some applications it may be desirable to process the acoustical signals, such as through the processor, on board the logging tool for transmission of the significantly data-reduced processed signal to surface in real time.

Once the drilling dynamics data is collected and processed, the results are correlated back to the measured depth of the well using precise measurements of the length of drill string components as they are lowered into the well. Gamma ray LWD and casing collar measurements can further be used to correlate the absolute location of data processed from the BHA collection point or points to determine a more reliable location of the bit in relation to the subsurface.

Noise Attenuation

Because of the amount of energy released is generally expected to be slight and of relatively high frequency, the radiated waves are best viewed when transmitted from the cutting face (with cutters 32) to the bit and bottom hole assembly, where they may further propagate and are known as the direct tool arrival or collar wave and contribute to the vibration of the drill string. The acoustic emissions may be measured by accelerometers, transducers, or other devices sensitive to particle motion.

Drilling induced vibrations that are generated by the interaction of the bit with the rock formation will have harmonic frequencies that are related to the rotational speed of the bit. Most of the harmonic vibrations are expected to be low frequency. The amplitudes of the acoustical emissions in the frequency ranges of the harmonic frequencies are usually much less than the amplitude of the harmonic drilling vibrations. In one implementation, the harmonic drilling vibrations are removed by a filter, such as a high pass filter or bandpass filter, which may be implemented in the bit sub processor, or may be applied to the stored data for later processing after download of the data from the bit sub memory, that passes signal frequencies that are higher than frequencies related to the harmonic drilling vibrations, and which may also eliminate frequencies above those possibly related to fracture characteristics. Other filter types and frequency characteristics will be possible depending on various factors including, but not limited to, the rotational speed of the bit, the type of bit, the rock characteristics, the positioning of sensors, mud motor characteristics, and other attributes.

Another consideration with respect to the variations in the amplitude and frequency of the acoustic emissions is the interference of the direct tool arrival by the generation and transmission of other wave modes that are also excited by the drilling operations including the wave modes excited by the release of energy from the fracturing of the rock formation.

In addition to the acoustical emissions created by the fracturing of a rock formation in response to the cutting action of the bit, the drilling vibrations that are being recorded will also generate wave modes that propagate though the formation surrounding the wellbore and the fluids within the wellbore that can interfere with the acoustical emissions that are related to the fracturing of a rock formation at the cutting face and bias the measurements of the mechanical rock properties. For some frequencies, the other wave modes will have higher amplitudes than the amplitudes of the acoustical emissions at that particular frequency. These other wave modes will result from the propagation of various guided waves in the drilling fluid between the bottom hole assembly and the wellbore such as Stoneley waves, tube waves and direct fluid waves such as the fluid compressional waves. Other wave modes that can interfere with the direct tool arrival are surface waves that propagate and refract energy along the interface between the well bore and the fluid such as compressional head waves, surface compressional waves and shear body waves. These waves all have the potential to interfere with the propagation of energy from the direct tool arrival and if their presence is included in the signal processing could bias the calculations related to the amplitudes and frequencies used to describe the nature and occurrence of the fracturing.

To decrease the possibility of interference with the other wave modes, in one embodiment the sensor is mounted internally using in a plug which will effectively isolate the sensor from the wave modes propagating through the formation and through the fluids. For example, in the case of accelerometers, the accelerometers are mounted within the bit sub (or other component). External waves that are carried by the formation and fluid in the annulus 36 will therefore be mechanically isolated by the internal position of the sensor relative to the other waves. An internally mounted sensor will be respond mainly to the vibrations related to direct tool arrival (the vibrations caused by the interaction of the bit with the formation).

Because the steel used in the construction of a bottom hole assembly has a significantly higher quality factor, Q of 10,000, that rock formations and fluids where Q ranges are typically from 1 to 100, the waves that propagate along the collar known as the collar wave or tool arrival will experience much less attenuation than signals recorded by the sensor that have travelled through the formation or the fluid media. Because of the low attenuation of the waves propagating along the steel drill collar relative to the waves propagating through the formations and drilling fluids, it could also be possible to naturally attenuate the various unwanted modes of propagation by placing the receiver at a distance that is far enough away from the bit to attenuate the other unwanted formation and fluid wave modes that will interfere with the direct tool arrival, but not so far as to lose the valuable high frequency information carried by the collar wave or tool wave that is needed to be recorded to calculate the size and displacement of the fracturing. The distances needed to attenuate the unwanted modes can be determined by expressions that relate the energy loss per cycle during transmission for a given quality factor. In this embodiment the location of the bottom hole assembly in relation to the location of the bit is placed at a distance behind the bit to achieve this attenuation. In certain applications in which the bottom hole assembly further includes a mud motor, the motor may attenuate at least a portion of the desirable high frequency modes. Accordingly, to avoid losing information due to such attenuation and to the extent the bottom hole assembly includes a mud motor, in at least some implementations of the present disclosure, the receiver may be located between the motor and the drill bit. Alternatively or additionally, to avoid the interference of fluid and formation arrivals with the collar arrival the receiver may be coupled to the drill string at the surface of the well where the drill string has yet to enter the subsurface and borehole that contains the drilling fluids.

In another embodiment, a plurality of receivers are arranged along the string behind the drill bit to record the acoustical signals generated by the release of elastic energy at the cutting face. In one example, the receivers are spaced within centimeters or millimeters and may be placed in array mounted on a portion of the bit sub. The spacing of the receivers may form an array, with the spacing determined by the frequency range of the acoustical signals and the velocity of propagation in the steel or other material. Because the velocity of propagation in the steel is typically much faster than most formation and fluid velocities which control the nature of propagation of the interfering modes and is known with a high degree of certainty, spatial filters such as FK filters that pass signals that propagate at velocities that are consistent with wave transmission through steel and attenuate events with slower velocities. These filters can be used to separate the direct tool arrival waves from the other interfering waveforms. When using an array of receivers the signal-to-noise ratio can be further increased by using geophysical signal processing techniques to filter the data by stacking the signals over the array. The nature of the stacking depends on the configuration of the array of receivers and whether the receivers are deployed in a linear, bipolar or radial array.

In some instances the stacking can be used to isolate various modes that are propagated by the tool arrival, such as the compressional wave, transverse wave or quadrupole wave. These other tool modes can also be processed using geophysical signal processing techniques to determine the fracturing of a rock formation in relation to the cutting action of the bit.

The quadrupole is a direct tool mode that does not propagate direct tool arrivals above a cutoff frequency, where the cutoff frequency is related to the diameter and thickness of the steel. The amplitudes and frequencies of these other modes provides useful though otherwise band limited information that may be used in relation to the geophysical signal processing techniques in order to specify how the size and displacement of the fractures would be responsible for generating these other wave modes.

The analysis of the signals extracted from the drilling vibrations should not be limited to the case of the amplitudes and frequencies of the direct tool arrival. In another embodiment, the receiver array can be used to reject the direct tool arrival or collar wave and pass other arrivals related to other modes of transmission, where the modes of transmission may be through the rock formation or drilling fluid system based on their velocity of propagation and frequency content through the media. These other modes of propagation may be used in preference to the direct tool arrival when the other modes of wave propagation contain signals related to the fracturing of the formation in relation to the cutting action of the bit that are of interest to the variations of the RMS levels of the signals as described herein.

Microseismic Signal Processing Techniques

In certain implementations of the present disclosure, drilling vibrations generated by the cutting action of the bit may be processed using geophysical signal processing techniques that are conventionally recognized as appropriate for the analysis of microearthquake source mechanisms. In one example, the depth of cut or penetration per revolution of the bit is obtained by using signal processing techniques to measure the sizes and displacements of fractures that result from the deformation and failure of the rock formation in relation to the cutting action of the bit, to estimate the zero-frequency level (ZFL) of the displacement spectra. The ZFL of the displacement is the static offset level here understood to represent the penetration of the bit. Under this consideration and in relation to the technique it provides high resolution motion of the bit that is taken to be the depth of cut or penetration of the bit for each revolution of the bit.

FIG. 5 illustrates a displacement spectra for one revolution of a bit. The graph of displacement spectra has a y-axis of rotary displacement amplitude at a range of frequencies (x-axis) over which amplitude measurements are taken (based on the sampling frequency.) The displacement spectra is obtained by integrating the Fourier transform of the time domain measurement of the acceleration data (bit vibration data) twice in the frequency domain, or the angular velocity time series once where it is understood that the accelerations may be axial, lateral, rotary or a combination of those channels. Rotary displacement may be in the form of radians. The ZFL (zero frequency level) determined by approaches discussed herein is where the displacement spectra theoretically intersects the zero-frequency axis. The ZFL is directly proportional to the average displacements of the fractures initiated by the bit, and to the displacement of the bit per revolution.

Generally speaking, because high resolution displacement data is achievable, the system can detect relative changes of displacement per revolution of the bit or displacement per time or rate of penetration (ROP) and thereby determine when a fracture is encountered as the displacement will be greater relative to areas where fractures are not encountered. With respect to implementations of the present disclosure, the increase in displacement would indicate a change in the hardness of the formation, where hardness increases with displacement all other things being equal.

Observations of strong ground motion generated by earthquakes suggest that the time series recordings of the P- and S-wave signals can be reasonably and effectively treated as band limited white noise where the amplitudes and frequencies of the signals are controlled by the interaction of many smaller faults and fracture patches rupturing simultaneously and the band limitations in the absence of attenuation are related to the displacements and rupture dimensions of earthquake source. Therefore, observations involving the simultaneous occurrence of multiple acoustic emissions generated by fracturing at the face of the bit and then transmitted through steel where the signals are expected to undergo little attenuation indicates that the application and use of models that are generally used to describe the source mechanisms of an earthquake can be used to describe the aggregate sizes and displacements of many fractures being generated simultaneously from the repeated turning of the many cutters on the drill bit when overcoming the rock strength to accommodate the motion of the bit.

The estimation of the size and displacement of the fracturing generated by the cutting action or motion of the bit follows directly from the display of the data that is present in FIG. 5. Here the model for representing the microearthquake source mechanism that estimate the size and displacement of the fracturing typically involves the application and use of a two parameter spectral model where here the two parameters used to describe the model are the zero-frequency level of the displacement spectra and the corner frequency. It should be appreciated that other source parameter models such as the RMS stress drop that also use the amplitude and frequencies of the signals to estimate the size and displacement of fracture and can also be considered and the use of a two parameter spectral model should not limit the scope of this disclosure.

In the two parameter model of the microearthquake source mechanism, the displacement of the bit as determined by the displacements on the fractures is related to the zero frequency level of the displacement spectra. The zero frequency level of the displacement spectra is the same as the static offset. The low-frequency content may be modified by the receivers and electronics used to record the signals on the MWD assembly and the possible use of band pass filters in relation to the signal processing to eliminate the low frequency drilling harmonics. The technique has advantages because it (i) provides an estimate of the ZFL using band-limited data that may not always be reliable in the lower end of the amplitude spectra, (ii) provides a reliable technique to objectively select the ZFL without manual or visual biases, (iii) can be automated in a computer implemented fashion to handle the large volumes of data typically collected by an MWD apparatus as used to collect data in relation to this disclosure.

Geophysical signal processing techniques that use relationships between the power spectral density of the displacement and velocity spectrum are used to calculate this value (ZFL) based on the functional mathematical representation of the earthquake source spectra as described by the two parameter spectral model of the earthquake source provide an objective technique to overcome the expected poor signal to noise ratios of the low frequency acoustical information.

Thus, ZFL is understood to represent the displacement or penetration of the bit. In one specific example, the ZFL is a measurement of the displacement (e.g., in millimeters or inches or angular displacements such as a radians) of the bit per turn of the bit. If the displacement is the axial displacement per revolution of the bit, then this displacement can be used to inform a depth of cut in terms of displacement per revolution. If time taken to make the one turn of the bit is used to describe the axial displacement or penetration of the bit, then it is taken that this may be used to determine rate of penetration. Typical depths of cut as estimated by the technique range from 0.01 inches per revolution to 0.1 inches per revolution

The size or the radius of the fractures can also be determined by the frequency content of the acoustical signal. Signals with higher frequency content generally correspond to smaller fracture areas. This characteristic of the frequency spectra that is used to determine the size of the event is typically referred to as the corner frequency (FIG. 5). There is a linear, relationship between the corner frequency of the displacement spectra and the size of the event. Because of attenuation of the signal during the transmission, estimates of the corner frequency may be compromised if the sensor is placed too far from the bit. Thus, having the sensors as close to the bit as possible, or at least not so far that attenuation is significant, may be a consideration for some implementations discussed herein. Limitations in the sensor and the recording electronics and the filters employed to extract the signals may also limit the useable bandwidth.

Geophysical signal processing techniques that use relationships between the power spectral density of the displacement and power spectral density of the velocity based on the functional mathematical representation of the earthquake source spectra as is described by the two parameter model of the earthquake source that can be used to calculate the corner frequency. A relatively high corner frequency may represent unfractured rock whereas relatively lower corner frequency may indicate the presence of a fracture.

In order to make meaningful comparisons of the fracture sizes and displacements along the well bore, the time window that is used to process the signals extracted from the drilling vibrations is based on the bit speed, in one specific implementation. One way to normalize the time window relative to the bit speed is to specify the time window according to the time needed to make one revolution of the bit. If the bit speed were 120 RPM then the time window would be 500 ms, while if the bit speed were 60 RPM then the time window would need to be 1000 ms to obtain an equivalent measurement of the fracturing generated in response to the cutting action of the bit. Another way to normalize the time window to account for variations in the bit speed along the wellbore would be to normalize the temporal frequency of the spectral density by converting from cycles per second to cycles per revolution. At a bit speed of 120 RPM, the bit would make two revolutions in a second and therefore at cycle per second would be normalized to two cycles per revolution, while a bit speed of 60 RPM would make one revolution in a second therefore one cycle per second would be normalized to one cycle per revolution. In one embodiment, the BHA is instrumented to measure the bit speed at sufficient resolution to specify either a time window equivalent to one rotation of the bit or normalize the spectral density by converting the temporal frequency to cycles per revolution.

As such, the spatial variations in the measurements as obtained through a combination of one or more of the measurements such as the RMS acceleration or the sizes and displacements of the fractures are understood to correspond to the nature and occurrence of deformation and failure in relation to the cutting action of the bit and as such are taken to represent the spatial variations in mechanical rock properties. So, an increasing value of ZFL, relative to a baseline, represents the intersection of the bit with zone of mechanically weaker rock that may be determined to be a fracture, swarm of fractures (e.g., fracture 34) in accordance with for example a stress-strain relationship. Therefore, the geophysical signal processing techniques employed by certain implementations of the present disclosure may involve statistical descriptions such as measuring the RMS level of the acoustical emissions or the application and use of geophysical signal processing techniques that are generally recognized as appropriate for the analysis of microearthquake source mechanisms to describe the deformation and failure of a rock formation in relation to the cutting action of the bit, where the spatial variations in the measurements (e.g., changes in ZFL or corner frequencies) relative to some average or baseline level are used to identify mechanical discontinuities or geological formations as they are encountered or crossed by the bit.

In another embodiment, referring now to FIG. 6, the geophysical signal processing techniques may obtain measurements related to the instantaneous change in the inclination of the bit relative to the average direction of the bit to describe a mechanical discontinuity or geological boundary. Instantaneous change in inclination of the drill bit relative to a long-term average inclination can be obtained using a sensor or an array of sensors configured to record and extract acoustical signals in relation to the independent spatial axes of the drilling vibrations. The magnitude of the deflection of the inclination of the drill bit relative to a long-term trend in the direction of the drill bit depends on the orientation of the mechanical discontinuity or geological boundary with respect to the cutting face of the drill bit and therefore the deflection is understood to indicate a change in the mechanical rock properties. So, for example, with a multi-axis accelerometer measuring axial acceleration and lateral or rotary acceleration of the bit, the ratio of the axial and lateral or rotary displacements may be treated as an inclination, as illustrated in the respective inclination logs 60 (prior to the fracture, while the bit 22 intersects the existing fracture 34). The lateral or rotary acceleration will increase relatively when the bit is deflected from axial movement, such as when the bit encounters an angled discontinuity transverse the borehole.

The Concept of Specific Energy in Relation to Acoustical Processing Techniques

In addition to acoustical information that informs the displacement of the bit, aspects of the present disclosure may further involve force information. While conducting drilling operations the energy defined as the energy needed to remove a volume of rock is useful to describe the efficiency of the drilling operation. The specific energy is a term that describes the minimum amount of work needed to remove a certain volume of rock. Descriptions of the specific energy are useful to understand the variations in rate of penetration or the depth of cut in relation to the forces acting on the bit. Unlike conventional methods, the specific energy here may be calculated from the aforementioned acoustical processing techniques in and of themselves or in conjunction with measurements of torque on bit and/or force on bit.

The size and displacement of the fractures that form in response to the cutting action of the bit controls depth of cut into a rock formation and subsequently the rate of penetration of the drill bit through a rock formation. In at least certain implementations, the work done per the volume of rock removed is determined by the displacement of the fractures as multiplied by the area of the bit. The work is computed from the forces acting on the bit multiplied by the displacement of the bit. As discussed above, the displacements of the bit are obtained through the analysis of the drilling vibration using geophysical signal processing techniques that are appropriate for the analysis of microearthquake source mechanisms as is provided herein.

There are two components of the specific energy: one that is normal to the bit and another that is tangential to the bit or rotary energy. Because the rotary specific energy is proportionally related to the torque per unit of displacement, it provides a measure of the rock strength or the minimum energy needed to drill and the application and use of the rotary specific energy in this regard takes the form of a stress-strain relationship (see, e.g., FIGS. 8A and 8B).

In implementations of the present disclosure, the displacement of the bit as it relates to the fracturing of the formation is evidenced through the RMS level of the measurements or the displacement of the bit is evidenced by the displacement on the fractures as provided through the microearthquake source parameter measurements and the volume of rock removed is proportional to the fracture areas and the fracture displacement are averaged over all of the fractures provides in the period of time processed for the time period that is analyzed for example in a single turn of the bit. In one implementation of the present disclosure, the volume of rock excavated is specified by the product of the aggregate area of the fractures and the average displacement of the fractures removed by one turn of the bit.

For a bit that is turning at a rate of 120 RPM this would involve using a time window of 500 ms to the average area and average the displacement for one revolution of the drill bit. It should be appreciated that in instances when the rock properties are varying slowly or the MWD measurements are updated at rates less than the time period of one bit revolution implementations of the present disclosure are not limited to periods that are specified by the turning rate of the bit.

The rotational specific energy is the torque over the average fracture displacement per revolution as obtained by the analysis of the vibration data disclosed by this technique, provides a novel, innovative stress-strain relationship (see, e.g., FIGS. 8A and 8B). For a given rock type, this is expected to be a linear relationship where the slope of the line is related to an elastic coefficient describing the strength of the rock (e.g., specified rock strength 1 (70), specified rock strength 2 (72), and specified rock strength 3 (74)). Because the measurements are obtained while drilling, the strength of the rock actually determined by and is intrinsically related to the fracturing in response to the cutting action of the bit. When using a polycrystalline diamond compact (PDC) bit, for example, the strength of the rock determined from this stress strain relationship would be directly related to the shear strength of the rock. Stress-strain measurements obtained using approaches disclosed herein can be used to characterize the elastic coefficients of the rock formation.

Role of Stress-Strain Relationships in Implementations of the Disclosure

The stress strain relationships employed by certain implementations of the present disclosure are populated from measurements taken while drilling. In one instance, the strain is understood to be related to the depth of cut or the penetration per revolution of the bit which is determined by differencing the spatial location at two instances in time versus the number of revolutions taken for the bit to travel that distance. In accordance with the present disclosure, where (i) the drilling vibrations are understood to represent the deformation and failure of a rock formation in response to cutting action of the bit in order to accommodate the forward motion of the bit through a rock formation, and (ii) the drilling vibrations as processed through the signal processing techniques to evidence the motion of the bit per turn of the bit are understood to represent strain and (iii) the RMS acceleration as obtained through the geophysical signal processing techniques to evidence the forces acting on the bit per turn of the bit are used to populate the variable in relation to a stress-strain constitutive equation. In a further elaboration, these strain measurements can be related to the forces such as the weight on bit and torque on bit acting on the formation in connection with the bit and drilling fluid system to provide a diagram of a general stress-strain relationship. By relating the orientation and magnitude of the stress with respect to the orientation and magnitudes of the strain, where the orientations and magnitudes of the stress and strain are related to the geographical coordinates of the well, multiple stress-strain relationships can be established to determine the elastic coefficients of a rock formation. Where the well is drill perpendicular to the maximum horizontal compressive stress, it is understood that these stress-strain relationships are expressed in the principle axes.

In general, the deformation of a homogeneous isotropic rock formation can be specified by two elastic coefficients. As the complexity of the rock formation increases through the presence of mechanical discontinuities and geological boundaries, the number of elastic coefficients needed to fully describe the deformation of a rock formation in response to the forces acting on the formation in general increases.

The simplest elastic coefficient would be to relate the WOB to the strain generated by the cutting action of the bit where the strain is the displacement in relation to a length made by one turn of the bit where the time windows used for the geophysical signal processing techniques are related to the bit speed. In one implementation of the present disclosure, the WOB would be obtained from the RMS acceleration, where the component of acceleration is oriented parallel to the borehole and as is illustrated. Another example of an elastic coefficient that may be used is shear modulus, which may be calculated using torque, acceleration, and/or gyroscopic measurements obtained during drilling of the wellbore.

For a transverse isotropic (TI) media, the stress-strain relationship constitutive equation is illustrated in FIG. 8A (using MWD data) with attendant PR intercept and YME slope information illustrated in the curves of FIG. 8B. Where:

E is Young's modulus of elasticity (YME),

ν is Poisson's ratio (PR)

σ1=TOB

σ3=WOB

ε3=Axial displacement spectra ZFL

This stress-strain relationship would in general be proportional to a Young's modulus of elasticity where YME is determined parallel to the direction of drilling. Certain implementations disclosed herein allow for the determination of elastic coefficients for transverse isotropic elastic media, where the assumption of transverse isotropic elasticity is understood to be reasonable approximation to describe the rock deformation by obtaining the stress-strain relationship to the orientation geological boundaries with respect to the inclination of the bit of the well being drilled. This could be accomplished using, for example, FIG. 8A or FIG. 100 where the elastic coefficients as described are proportional to Young's modulus of elasticity and the Poisson's ratio of a rock formation.

In another implementation more particularly shown in FIGS. 9A and 9B, the rock strength is specified by the stress-strain relationships based on the orientation of the well in relation to the bedding planes of the hydrocarbon bearing formation and the orientation of the well with respect to the principle axes of tectonic stress or the state-of-stress acting on the rock formation. This approach allows for the determination of elastic coefficients for a transverse isotropic (TI) elastic media, where the assumption of transverse elasticity is taken as a reasonable approximation to describe the rock deformation by obtaining the stress-strain relationship to the orientation geological boundaries with respect to the inclination of the bit of the well being drilled. This could be accomplished using the above description where the elastic coefficients as described are proportional to Young's modulus of elasticity and the Poisson's ratio of a rock formation.

The loading conditions on a rock formation are given by the forces acting on the rock formation in connection with the drilling apparatus (e.g., weigh on bit and/or torque on bit) and drilling fluid system (e.g., annular pressure) and the deformation of the rock formation are described by the displacements of the bit (e.g., axial and lateral or rotary displacements as measured by accelerometers in the bit sub). The constitutive equations are described with respect to transverse isotropic media. As shown in FIGS. 9A and 9B, transverse isotropic media involves a layering of media (sometimes referred to as “layercake”). Transverse isotropic media involves a layering of media that is normal to a plane of isotropy—meaning the media is relatively uniform around the axis of symmetry. FIG. 9A illustrates a case where the axis 90 of material symmetry is parallel to the borehole 94 (and parallel to the bit axis 96 drilling the borehole). FIG. 9B illustrates a case where the axis of material symmetry 98 is perpendicular to the borehole. In one possible implementation, it is assumed that the media is either vertically transverse isotropic (VTI) or it is horizontally transverse isotropic (HTI). VTI is a case where the axis of symmetry is vertically oriented (layers are horizontal with respect to the free surface). HTI is a case where the axis of symmetry is horizontally oriented (where the anisotropy is understood to involve a layering of the formation that is vertical with respect to the free surface and further where it is understood that fractures represent a case of vertical layering with respect to the free surface). Thus, in the case of a vertical well, FIG. 9A illustrates vertical transverse isotropy (the vertical borehole is parallel to a vertical axis of isotropic layer symmetry) and also illustrates, in the case of a horizontal well, horizontal transverse isotropy (the horizontal borehole is parallel to a horizontal axis of isotropic layer symmetry). In contrast, in the case of a horizontal well, FIG. 9B illustrates vertical transverse isotropy (the horizontal borehole is perpendicular a vertical axis of isotropic layer symmetry) and also illustrates, in the case of a vertical well, horizontal transverse isotropy (the vertical borehole is parallel a horizontal axis of isotropic layer symmetry).

FIGS. 9A and 9B also illustrate the variables for the constitutive stress-strain equation (FIG. 8 and FIG. 100) of a transverse isotropic media. The stress strain variables are populated based on either (i) force measurements (e.g., from strain gauges) and/or (ii) the acceleration measurements (e.g., from accelerometers) or populated only form acceleration data. More specifically, in the first case, the stress-strain variables are populated with data related to WOB, TOB or axial or lateral or rotary displacements. In the second case, the stress strain variables the forces are populated with axial and lateral or rotary acceleration data, and the strain from the axial or lateral or rotary displacements (which may be available by integrating the axial or lateral or rotary acceleration spectra twice in the frequency domain). Generally speaking, Poisson's ratio (and/or Young's modulus of elasticity) is computed from stress-strain constitutive equations populated from measurements taken while drilling along a borehole under the assumption that the transverse axis of material symmetry is parallel to the borehole (FIG. 9A) and under the assumption that the transverse axis of material symmetry is perpendicular to the borehole (FIG. 9B).

When the rock formation (media) is isotropic, the two equations will generate values of PR or YME that generally track each other, meaning that under either assumptive case, the drill bit will react similarly as it drills through relatively uniform rock and therefore the two calculations of PR and YME, while different, will nonetheless track each other. If, however, the borehole intersects discontinuities and or more the media is anisotropic, the computations of PR and/or YME will no longer track each other. The stress-strain relationships may be recast into other equivalent forms, and the particular arrangement shown in FIGS. 8, 9, and 10 are taken for the sake of conveniences and should not be considered limiting. In a particularly useful manner as shown in FIG. 11, the constitutive equations could be re-arranged so that the Young's modulus term is determined using the slope of a linear relationship if the equations were cast in terms of the ratio of the forces acting parallel to the axis of symmetry to the forces acting perpendicular to the axis of material symmetry.

In certain instances, such as drilling under high confining annular pressure, the confining pressure may be more accurately described using the annular pressure instead of WOB. In such a case, AP would be used in place of WOB for the cases illustrated in FIGS. 9A and 9B, and possibly others.

Some of the stress-strain relationships are set out in terms of bi-axial loading accounting for weight on bit and torque on bit. However, it should be understood that an implementation accounting for tri-axial loading may nonetheless use and account for WOB and TOB or the RMS acceleration.

Thus, in the application of the stress-strain relationships for a layered media, where the bedding planes are horizontal a vertical well and a horizontal well are drilled through the same rock formation where:

1. Separate stress-strain relationships that describe the elastic coefficients when the bit is cutting perpendicular to the geological boundaries such as the bedding planes and when the bit is cutting parallel to the geological boundaries would enable at least four elastic coefficients to be determined which can be used to describe in general, stress-strain relationships of transverse isotropic rock formations.
2. Separate stress-strain relationships that describe the elastic coefficients when the bit is cutting parallel to the direction of maximum horizontal compressive stress and parallel to the maximum vertical compressive stress would enable at least four elastic coefficients to be determined.

Referring to FIG. 100, under the two assumptions (TI axis parallel to borehole and TI axis perpendicular to borehole (FIGS. 9A, 10A and 9B, 10B)), there are two values for PR and two values for YME, with PR and YME being elastic coefficients.

In practical application, the rock formation may not be actually horizontal or vertical, but may be tilted. When drilling through titled media, particularly media with small deviated angles of less than 30 degrees, given the expression of the trigonometric variables in relation to a rotation of the principle axis, the equations still produce useful results for PR and YME and may be used to inform the variation in the mechanical rock properties and in particular inform the location of fractures where the variation of the mechanical rock properties is predicted to do so as is described below.

Calculation of the Elastic Coefficients Through Mechanical Rock Property Analysis (MRPA)

FIG. 12 is a diagram illustrating linear stress strain relationships with curves fit to data pairs corresponding to different locations along a bore hole. The slopes of the fit lines in these examples relate to YME in different locations along a bore hole. To obtain data for populating the constitutive equation (or equations), various geophysical data processing techniques are involved, where:

1. Sampling the MWD measurements at a sufficiently high frequency to resolve a small degree of mechanical variability (which may correspond to the nature and occurrence of a discrete fracture a couple of mm wide), the measurements of the forces acting on a rock formation in connection with a drilling bit, (TOB and WOB) and fluid system pressure or annular pressure (Ap), the angular speed of the bit expressed in revolutions per unit of time (RPM) and the 3-components of motion that represent the acceleration of the bit and where each of the measurements taken in time corresponds to a discrete position of the bit along the length wellbore (the MWD data)
2. Processing accelerations may be processed using geophysical signal processing techniques to obtain (i) the average lateral or rotary and vertical displacement of the bit that correspond to a single revolution or turn of the bit and (ii) Root Mean Squared (RMS) amplitude of the acceleration
3. Calculating the RMS averages of the TOB and WOB over time windows that correspond to a single bit turn. Typical penetration rates are usually 0.02 in/rev, while drilling at 240 rpm, would result in a single turn of the bit every 250 ms which if sampled at 1 kHz would provide sufficient data to be able to identify when the bit encounters a single-discrete fracture
4. Obtaining data pairs, which collectively define the linear stress strain relationship, from MWD data using relationships between the rms acceleration and displacement of the bit that are appropriate for the loading conditions and motion of the bit in relation to the axis of symmetry for the constitutive equations used to describe the rock formation where it is understood that the loading conditions are determined with respect to the orientation of the drilling well.
5. Using curve fitting techniques to estimate the two parameters needed to describe a line, the Slope and Intercept and statistical descriptions of the variations of the two elastic parameters YME and PR with respect to each of the linear clusters as were identified with respect to the band-limited MWD data that was used to generate the data pairs
6. Identifying where the distributions of the MWD data pairs form spatially or temporally consistent clusters or otherwise a locus of adjacent points that (i) can be described through the application and use of curve fitting techniques in terms of relationships that are linear and (ii) such that each of the linear relationships can be used to determine the parameters of line such as the slope and intercept in relation to the mechanical rock properties YME and PR.

In one possible example, a set of data pairs are generated for the equation of FIG. 9A (10A) and/or the FIG. 9B (10B). The data pairs are generated along the length of a well bore. In one example, a data pair comprises (y, x) and the set of data pairs may be used to generate the linear stress strain relationships illustrated in graphical form in FIG. 12 for each of the cases shown in FIG. 9A and 9B.

In one embodiment, the variables of the equations are populated using (i) the RMS acceleration data, which is used to describe the forces acting on the formation in connection with the bit, and (ii) the ZFL of the displacement spectrum is used to describe the motion on the bit where the motion understood to be the strain experienced by the rock formation and where the orientations of strain are described by the bit displacements according to the cutting direction of the bit and the orientation of the borehole in relation to the orientation of the axis of material symmetry as is shown in FIG. Thus, the acoustical acceleration measurements may use to specify the forces acting on the formation. The approach essentially conforms with Newton's second law and balance of forces.

Special Cases to Consider

As shown, the MWD parameters may be expressed as a function of the frequency content. In particular, where the frequency content is limited through the application and use of a band-pass filter to generate band-limited MWD data pairs to populate the terms and conditions of the constitutive equations of linear elasticity, using band-limited MWD data to form frequency-dependent, data pairs corresponding to data that generally describes the stresses acting on the material and the deformation of the material to populate the terms and conditions of the constitutive equations of linear elasticity. Using geophysical signal processing techniques to obtain zero-frequency levels (ZFL) of the displacement spectra where the ZFL corresponds of depth of cut per revolution of the bit or the depth of cut per unit of time or the rate of penetration (ROP) in relation to the width of the bandpass filter used to window the MWD data (the band-limited MWD data). That is the ZFL is determined from a specified range of frequencies or is otherwise calculated from bandlimited data.

Using the two parameters YME and PR and the statistical descriptions of the variations of the two parameters as can be obtained through the curve fitting technique for each liner cluster to diagnose the drilling conditions for each of the depths as a function of the frequency used to form the data pairs.

Fracture Identification from the Elastic Coefficients

It is usually not known a priori what the appropriate angular relationships between the axis of symmetry of the constitutive elastic equations used to describe the rock formation and the axis of symmetry of the wellbore are. In practice, most horizontal wells are drilled parallel to bedding and most vertical wells are drilled perpendicular to bedding. Further, in basins, most natural fractures are vertical, and thus most horizontal wells are drilled perpendicular to fractures and most vertical wells drill parallel to fractures.

Referring again to FIGS. 9A and 9B, the MRPA technique is used to populate the terms and conditions of the two assumptive cases where the axis of symmetry of the constitutive media used to describe the rock formation are examined for the cases when (i) the axis of material symmetry is perpendicular to the axis of drilling and (FIG. 9A) (ii) the axis of material symmetry is parallel to the axis of drilling (FIG. 9B). In an isotropic media, the determination of YME and PR from the MRPA analysis for the two constitutive stress strain relationships will result in equally or closely spaced values of YME and PR that tend to track each other in space and time. In anisotropic media, the two computations of YME and PR as provided by the MRPA will deviate in predictable ways that can be used to identify the nature and occurrence of fractures in relation to (i) the differences in the elastic coefficients as specified by the type of anisotropy, either HTI or VTI, that is encountered relative to the orientation of the drilling well and (ii) the reduction in strength of the rock as is provided by variations in HTI YME and the VTI YME.

In the case of a horizontal well when the constitutive equations describe a variation in the end members situations (i) the rock may be understood to be fractured when PR VTI is lower than PR HTI and the magnitude of the fracturing is related to the decrease in the calculation of YME. Typically, in practice, fractured zones can be discerned in horizontal wells when the media is VTI PR is lower than the HTI PR. The difference in the values of the elastic coefficients YME and PR between the HTI and VII represent two cases of a cross-over curve or an anisotropic cross-over curve.

In the case of a vertical well, the rock is understood to be fractured when HTI PR is less than VII PR (e.g., FIG. 13, discussed in more detail below). An objective way to determine these cross-over relationships (PR or YME for the two assumptive cases) is by calculating a series in time or a series in depth (the logs) of the YME or the PR for both the VII and HTI solutions. Time or depth may be correlated to length along the well bore from which the measurements were made.

Mechanical rock property logs (the “Logs”) as calculated using the equations set out by one or more of the approaches discussed herein can be smoothed by averaging the PR and YME values by (i) using the statistics in relation to the goodness of fit to the curve where one such statistic is known as the R statistic as provided by standard least-squares linear curve fitters such as LINEST in Excel, to filter out data with poor statistical evidence for a linear relationship between the data pairs or (ii) to weight the values of the curve at a particular location. Using a curve smoothing technique that would average the data over a time window where the length of the time window corresponded to the variation of the data and then resampling the time window in either time or depth according to its position in the subsurface. Processing the Logs using smoothing techniques can improve the ability to identify the relative variations in the elastic coefficients.

Further processing the Logs by subtracting the mean value from the HTI and VII YME Logs where the mean value is a running average of the data along the Logs where the length of the running average is related to:

• • (1) the measure of the smallest length of mechanical anisotropy variation of interest which in terms of the practice and application of implementations of the present disclosure may involve spatial distances as small as 1 inch for the sample rate and frequency content that is afforded by the application of use of state-of-the-art MWD in relation to the implementations AND
  • (2) of sufficient resolution to document the nature and occurrence of changes in rock properties needed to identify fractures at the resolution need for the commercial exploitation of commercial hydrocarbons from unconventional reservoirs.

The subtraction of the mean value of from the HTI and VTI PR Logs provides a baseline from which to compare the variations between the curves in ways that can be used to identify the locations of fractures. In practice the mean values of the Logs can be calculated as the average of the data values in the logs over a certain time or distance specified by the spatial position in the subsurface along the length of the wellbore from which the measurements were made. When this mean value is subtracted from the Logs it provides the mean-subtracted Logs from which to make it convenient to obtain relative comparisons (e.g., PR HTI to PR VTI and/or YME VTI to YME HTI).

Taking differences of the mean-subtracted YME HTI Log and the mean-subtracted YME VTI Log and taking the differences of the mean-subtracted PR HTI log and the mean-subtracted PR VTI log provides an identification of the type of rock anisotropy based on the orientation of the well in relation to the axis of material symmetry. The differences in these Logs as evidenced by the behavior of the elastic coefficients in relation to the orientation of media symmetry with respect to the orientation of the wellbore can be understood in predictable ways to describe the location of a zone of weakness understood to be a fracture. For the cases presented here, these relationships provide a predictable way to identify fractured rock formations, among other advantages

Specific Case 1: Vertical Well, Horizontal TI Media

In typical VTI media, the vertical PR (FIG. 9A—TI axis parallel to borehole) typically has lower values than horizontal PR (FIG. 9B TI axis perpendicular to borehole). That is, the material is more compliant to a load that is applied perpendicular to the axis of material symmetry than to a load applied parallel to the axis of material symmetry. Loading the rock formation in the same direction as the axis of material symmetry will result in less horizontal deformation, decreased horizontal compliance and/or higher ratios of horizontal to vertical stiffness. Conversely, loading the rock formation perpendicular to the axis of material symmetry, when the material symmetry is governed by fractures, will have higher compliances, higher PR and lower YME.

Stated differently, in a vertical well when the media is layercake (the axis of media symmetry is parallel the borehole), the weight on bit and axial displacements are typically parallel to the axis of material symmetry. In this case the VTI PR is typically less than HTI PR. Therefore, detection of zones where PR VTI is greater than PR HTI implies that the material behavior under the loading conditions of the bit is stiffer or less compliant in the horizontal direction as opposed to the vertical direction. Referring to FIG. 13, when a vertical well encounters HTI media (in a formation expected to be VII), the mean-subtracted PR HTI log becomes less than the mean-subtracted PR VII log, and a Fracture ID flag may be generated. This flag means the ratio of the displacement parallel to the axis of material symmetry increases relative to the displacement perpendicular to the axis of material symmetry. This is evidenced through the application and use of MWD data to describe the constitutive behavior of a rock formation and demonstrate an increase in PR VII relative to PR HTI. This increase in PR evidences the presence of vertical fractures or a vertically fractured rock formation when drilling a vertical well. This logic is generally true is most circumstances, because most vertical wells are drilled perpendicular to the bedding planes of the rock formation where the well bore axis is parallel to the axis of material symmetry and so the presence of high VTI PR values is probably not related to vertical bedding planes. This behavior of a vertically drilling well encountering a set of vertical fractures can be further corroborated by a similar examination of the differences between the VTI YME and the HTI YME. Because the torque on bit is acting as a body force parallel to the axis of material symmetry as would be expected in the case of a vertically drilling well encountering a set of vertical fractures where the fractures control the axis of material symmetry, the VTI YME decreases and the HTI YME increases.

When the YME VTI log crosses or decreases relative the YME/HTI log for a horizontal well then it is likely a zone of weakness that is more compliant in the direction of loading that is parallel to the WOB or perpendicular to the axis of material symmetry has been detected which would again be consistent with a zone of vertical fractures. So, in some instances, PR crossover would generate a flag, YME crossover would generate a flag, and in some instances the presence of both flags indicate a fracture. Moreover, in practice, a threshold may be applied that would need to be met before generating a flag. In one example, for PR data, a distribution curve may be generated for all positive crossovers, and only crossovers exceeding 68th or 90th or 95th percentile may be flagged. Other thresholds or data technique may also be used to eliminate data points that may be attributable to noise.

When the relationship between the HTI and VTI elastic constants returns to a pre-crossover relationship for a vertical well in a VTI media, the crossover fracture ID flag is set back to zero. The flag may remain set, however, for as long as the log indicates. So, as shown in the example of FIG. 13, there are six sections of the 100 foot illustrated borehole where fractures are identified over several feet for each section.

Specific Case 2: Horizontal Well, Horizontal TI Media

FIG. 14A illustrates the two YME computations (FIG. 10A and FIG. 10B) over about 100 feet of horizontal well. FIG. 14B illustrates PR ratio computations (also FIG. 10A and FIG. 10B) over the same 100 fee of horizontal well. The YME and PR computations are based on data obtained while drilling. In this case, only the geophysical processing of the acoustical data as obtained from the accelerometers are used to populate the equations, so not direct measurements of WOB or TOB are provide in this example, but they could be included in another application and use of the equations as provided herein. In the example illustrated in FIGS. 14A and 14B, a fracture ID may be generated where shown. In this case, effectively the opposite behavior than was described for a vertical well drilling parallel to the axis of symmetry or VTI or layercake, is expected because now, in the case of a horizontal well in TI media, the axis of drilling is perpendicular to the axis of media symmetry. In this case, the loading conditions (torque on bit) are parallel to the axis of material symmetry conditions and rotational displacements (revolutions of the bit) are also parallel to the axis of material symmetry and HTI calculation will result in a lower HTI PR.

In the case of a horizontal well encountering a vertical fracture, the relationship is expected to be similar to a vertical well drilling in a VTI media—the PR HTI will decrease relative to the mean-subtracted PR VTI. More specifically, in the event with the mean-subtracted VTI PR for a horizontal well drilling in a layercake media is found to be less than the mean-subtracted HTI PR, then it is likely that the axis of material symmetry relative to the orientation of the drilling well and loading conditions of the bit has been rotated by 90 degrees. This can occur when the axis material symmetry is defined by a set of vertical fractures being intersected by a horizontal well. Thus, as shown, for example in FIGS. 14A and 14B, a fracture flag may be identified in the identified areas, as well as possibly other areas.

Likewise, the mean subtracted VTI YME will increase relative to the mean-subtracted HTI YME when the material symmetry is as defined by a set of vertical fractures because torque on bit will be loading perpendicular to the axis of material symmetry. In this example the cross-overs are not always synonymous as can be expected for real rock where a continuum of mechanical rock properties will occur based on the natural heterogeneity of a complex, natural system where it is understood that the detection of fractures from among various the relationships between these cross-overs is just one implementation among others.

Among the advantages of the approaches disclosed herein is to use the values of the differences in the elastic coefficients between the various types of cross-over than can be expected from the disclosed approaches where in simple cases fracture flags identified by the simultaneous cross of both YME and PR curves, to other cases of mechanical rock heterogeneity that is caused by the occurrence of only one curve crossing over the other or vice versa. Here this may lead to additional fracture classification schemes that would involve the cross-over of one set of coefficients relative to the other.

Elastic coefficients and the variation in the elastic coefficients where the variation of the elastic coefficients is determined in one particular application though the differences as obtained by a subtraction and of the elastic coefficients for the HTI and VII instances as is specified by the orientation of the well in relation to the axis of material symmetry. When the variations determined from the stress-strain relationships based on the geographical orientation of the well relative to the bedding planes provide an indication of fractures in accordance with the manner described here would provide useful information for the design the emplacement of hydraulic fractures treatments and the selection of hydraulic fracture initiation points.

Other Improvements

In one embodiment, the wellbore is drilled laterally through an unconventional shale reservoir. The natural variations in the strength of an unconventional shale reservoir could be viewed by plotting a plethora of stress-strain relationships that could be derived in association with every single turn of the bit (FIG. 8). This would create a scatter plot that can be analyzed by using statistical methods to find significant relationships in the data that are used to classify the nature of the rock deformation based on the groupings of the mechanical rock property data on the stress-strain diagram.

These classified rock strength measurements may be indexed according to their spatial location along the well bore trajectory. Classification of the type of rock deformation such as strong or weak, brittle or ductile based on the mechanical rock properties and in particular the elastic coefficients in the stress-strain diagrams when logged using the MWD system could be used to identify and select zones along the wellbore selection of hydraulic fracture initiation points for the emplacement of hydraulic fractures.

Other stress-strain relationships can be developed in the manner described here where the forces acting the formation in connection with the bit are the effective stress obtained by differencing the force and the fluid pressures in the drilling system or the Torque acting on the bit. In another embodiment the orientation and geometry of the cutters in relation to the WOB and Torque can be used to describe tractions that are normal and tangential to the cutting face and be used to specify additional coefficients of elasticity as provided by implementations of the present disclosure. The normal traction can be modified by the drilling fluid pressures to calculate effective normal stress acting on the cutting face of the rock formation. When the effective normal stress and shear stress can be projected onto a fault place where the fault plane undergoes reactivation as evidenced by the techniques herein and when used in conjunction with a failure criterion, critical information on the state-of-stress in the reservoir can be provided.

Discrete Microearthquake Detection to Identify Faults

Referring now to FIG. 15 if the forces acting on the formation in connection with the drill bit and drilling fluid system when conducting drilling operations are sufficient to overcome the failure criteria of a pre-existing fault, then the fault will slip or fail. Reactivation of a fault or pre-existing fracture can be evidenced by extracting a signal from the drilling vibrations that is related to a microseismic event with attendant primary, compressional (P) and secondary, or shear (S) arrivals. In the case where the fault is perpendicular to the trajectory of the wellbore the P-wave arrival is related the particle motion parallel to the axis of the drill string and the S- or transverse wave is the particle motion parallel to the lateral and torsion motion of the drill string. Deviations of the particle motion of the P- and S-waves can be used to determine the orientation of the fault relative to the trajectory of the wellbore.

Reactivation of a fault is expected to create much larger signals are typically expected from the acoustic emissions generated by the fracturing of a rock formation in relation to the cutting action of the bit. By looking at the instantaneous amplitude levels relative to a long-term trend, where the temporal windows used to select the instantaneous amplitudes are related to the bit speed and the long-term window is related to the spatial distribution of the faults in the rock formation, it is possible to identify the location where the bit encountered and crossed a fault.

In the special case of a fault reactivation while conducting drilling operations, the sensors deployed on the bottom hole assembly act like an earthquake seismometer where the geophysical signal processing techniques identify the discrete arrivals of P-waves and S-waves with orthogonal particle motions to detect the presence and reactivation of a pre-existing fault. If the orientation of the fault with respect to the orientation and magnitude of the forces acting on the formation in connection with the bit and drilling fluid system can be determined then this would enable a technique to specify in 3-dimensions a failure criterion of the fault.

Calculation of Scalars and Calibration Values

Aspects of the present disclosure may also involve the determination of absolute values of mechanical rock properties. The systems and techniques involve the application and use of (i) the forces or accelerations of the drill bit and (ii) the displacements or motions of the drill bit. Such force or acceleration and displacement data may be obtained from recording the near-bit mechanical drilling vibrations in relation to the drill bit interacting with a material having known properties and with a rock formation. In one specific example, drilling vibrations experienced by the drill bit from its breaking of rock while drilling, propagate as acoustical signals that are translated into data by accelerometers or other sensors positioned proximate the drill bit. The acoustical signals are translated into mechanical rock properties according to the techniques discussed herein. Further, by first drilling through some known media with known mechanical rock properties, such as a cement 38 (see FIG. 2B) in the well, the system may capture the vibration data and generate scalars. In turn, the scalars can be used to transform derived mechanical rock properties, from data captured in the same manner but for an unknown media, such as a formation being drilled through, into mechanical rock properties for that formation. In one example, the mechanical rock properties for the formation are considered absolute values in that the mechanical rock properties have been normalized by first obtaining scalars for the mechanical rock property computation, which may be based on using the same drill bit and related components to drill through a media with known properties.

Stated differently, the present disclosure outlines an innovative technique, and associated systems and apparatus, to obtain at least one set of calibration values, which in one specific case are scalars, in relation to drilling a rock formation or material with known mechanical properties. Scalars obtained in the manner presented here provide for a way to transform MWD data to stress and strain experienced by a rock formation or material when interacting with a drill bit.

An MWD device, as disclosed, measures and obtains data pertaining to drilling forces, such as the weight acting on the bit (WOB) or the torque acting on the bit (TOB) and/or accelerations that describe the angular and linear motions of the bit. Mechanical rock properties may be described using stiffness coefficients, colloquially known as the Cij's, that are expressed in terms of stress represented generally as force per area and more specifically as pounds per square inch.

Stiffness coefficients of rock formations are expressed in terms of force per unit area and are on the order 1e10 Pascal's or several Mpsi. Downhole measurements of forces such as weight on bit are on the order of 10's of kilopounds and torque on bit on the order of a few kilopounds per foot. Typical values of near-bit accelerations used to represent the forces acting on the formation can be on the order of several g's. And, typical values of the displacement of the bit as determined from processing the near-bit accelerations are on the order of several micrometers.

Because the MWD data represent forces or accelerations, and displacements as opposed to stress (a) and strain (c), respectively, corrections that take into account the physics of the elastic radiation of the vibrations generated by the drill bit, the transmission of vibrations as acoustical waves to the MWD data recorder 36, and the lengths and areas over which the radiation and transmission occur need to be obtained to transform the MWD data to units that can be used to obtain the stiffness coefficients or the material or rock formation being drilled. Likewise, the forces obtained from MWD data, such as weight on bit and torque on bit can be transformed to stresses with the understanding that a geometric correction factor in relation to the effective contact area is also involved. One example of such a contact area is the area of the bit.

In practical terms, it may be difficult or impossible to know, in-situ, the actual and precise contact area for any given turn of the bit. The area of the bit in contact with the rock formation depends on the various configurations of the cutters on the bit with respect to the weight on the bit and the wear of the cutters on the bit. More weight on the bit presses the cutters further into the formation and results in an increase in contact area as a function of the weight on bit and in a complex fashion. Also, the loss and wear of cutters during normal drilling operations will also have unpredictable and hard to determine changes in the contact area of the bit.

The motion of the bit also needs to be transformed or otherwise scaled to a strain in order to process the data using a stress-strain relationship and to obtain absolute values for the mechanical rock properties. The displacement of the bit that occurs relative to a given segment or reference length of the rock formation may be used to obtain the strain experienced by the rock formation when interacting with a drill bit. One such example of the reference length of the rock could be the circumference of the borehole. Using the circumference of the bit used to obtain the reference length would have to account for changes with respect to the radius of the bit or possibly in an arbitrary manner depending on the geometry and configurations of the cutters on the bit.

The use of geometrical constructs and analytical expressions of wave generation and propagation to transform the MWD data to obtain stress and strain of a rock formation interacting with a drill bit is unsatisfactory because, (i) the reference lengths of the rock with respect to the displacement of the bit is generally unknown, (ii) the area of the bit is rugose and variable, and (iii) the radiation and scattering of energy from the bit rock interface and the transmission of the energy to the MWD recorder is difficult if not impossible to accurately predict in-situ and under changing drilling conditions.

Referring to FIG. 16, a method is provided for obtaining a set of scalars from drilling a rock formation or material with known mechanical properties that transforms the accelerations or forces, and displacements obtained from MWD data to corresponding stresses and strains. The method involves (i) drilling through a material with known mechanical properties and obtaining signals, such as acoustical signals from the drill bit interacting with the material (operation 1610), (ii) processing the MWD data to obtain the forces on the bit and the motions of the bit interacting with the material (operation 1620), (iii) populating a general stress-strain relationship using (a) the forces on the bit, (b) the motions of the bit, (c) the values of the known mechanical rock properties and (d) the scalars (unknown) that are to be determined according to each force and displacement (operation 1630), and (iv) obtaining values of the scalars that satisfy conditions of the stress-strain relationship in relation to the knowledge of the mechanical rock properties (operation 1640). The acoustical signals may be stored in memory of the bit sub. These scalars are understood to account for geometrical considerations such as the effective areas and reference lengths in relation to the bit design and take into considerations other complex, hard-to-predict factors, such as the radiation and transmission of the drilling induced vibrations from the cutters to the MWD sensors.

The techniques discussed herein involve processing a material having known mechanical rock properties. Knowledge of the mechanical rock properties, which may be considered absolute values, useful to process drilling vibrations can be obtained independently from common well known methods such as from (i) sonic or acoustical measurements of rock velocities that can be systematically related to mechanical rock properties or (ii) the application and use of techniques that systematically measure the deformation of a rock sample or a core from a rock sample in response to a given load or stress to obtain mechanical rock properties. Other knowledge about absolute values of mechanical rock properties that can be used to scale drilling vibrations may also be obtained from computer generated models of rock mineral stiffness coefficients. In some instances, the absolute values of mechanical rock needed to scale the drilling vibrations can be arrived at through deductive or experiential means or what is commonly understood as a best guess. Cement is one example of a material with known mechanical properties that is commonly encountered in a wellbore during the drilling process.

The process of drilling a horizontal well may be completed in a series of stages. The first stage may involve the drilling of a vertical well to a target formation 30, also referred to as a hydrocarbon bearing formation, which is part of a larger reservoir. After the first stage has been drilled, the wellbore is typically cased with steel tubes, which are held or suspended in place by cement between the outside of the steel and the wellbore. After setting the casing, there is typically cement 38, referred to as a cement plug, left in the bottom of the well. The second stage would then involve sidetracking to a lateral or drilling through the cement plug in order to begin drilling directionally forming a horizontal well 38 towards the reservoir in order to intersect and target hydrocarbon formations of the reservoir.

In the instance where the rock formation or material being drilled includes a cement plug with known mechanical rock properties per or given the constituents of the cement or other techniques that can be used to investigate the mechanical properties of a cement as described above, then the processing of the acoustical signals generated by drilling vibrations involves an innovative way to obtain stress and strain from corresponding forces or accelerations of the bit, and the motions or displacements of the bit in relation a set of scalars (A, B, C, D, E and F) with respect to or representative of the known mechanical rock properties.

The mechanical properties of the cement can be expressed as an isotropic media in terms of two elastic stiffness coefficients C₁₁ and C₁₂. The isotropic stiffness coefficients and the forces or accelerations and displacements of the bit in relation to the scalars (A, B, C, D, E, F) that are required to transform the MWD data to stress and strain can be arranged through a system of linear equations as follows:

$\left[\begin{array}{cccccc}-a_1& 0& 0& d_1C_{11}& d_2C_{12}& d_3C_{12}\\ 0& -a_2& 0& d_1C_{12}& d_2C_{11}& d_3C_{12}\\ 0& 0& -a_3& d_1C_{12}& d_2C_{12}& d_3C_{11}\end{array}\right]\left[\begin{array}{c}A\\ B\\ C\\ D\\ E\\ F\end{array}\right]=\left[\begin{array}{c}0\\ 0\\ 0\end{array}\right]$

This stress-strain relationship represents a homogeneous system of linear equations that can be solved for the scalars (A, B, C, D, E, and F) from measurements while drilling a cement or other isotropic material where the mechanical properties are known. Here the displacements or “d” terms and the accelerations or “a”, terms are specified respectively in terms of (i) the axial, angular, centripetal and/or lateral displacement of the drill bit and (ii) the axial, angular, centripetal and/or lateral accelerations of the bit when interacting with the cement. Values for the stiffness coefficients, C₁₁ and C₁₂, are known for the sample media. In various possible examples, the “d” terms may be any combination of displacements (e.g., d₁ is axial, d₂ is angular, and d₃ is centripetal, or d₁ is angular, d₂ is centripetal and d₃ is axial, or d₁ and d₂ are axial and d₃ is angular, etc.), and the acceleration or the “a” terms may similarly be any combination of accelerations.

It would also be possible to represent the “a” terms using dynamic data, such as from strain gauges associated with the MWD, of the WOB and the TOB where the solution to the scalars would represent the effective contact areas of the bit with respect to the orientation of the forces being applied to the bit.

Stated independently, the effective contract areas of the bit acted on by the torque is not expected to be the same as the effective contact area acted on by the weight on bit. The differences in these contact areas depend on the geometry and configuration or arrangement of the cutters and the exposure of the cutters. In the case of scaling the weight and torque forces to stress, the scalars describe how the cutter exposure with respect to a particular bit design transform the forces to the stresses normal and perpendicular to the motion of the bit.

The technique can also be extended to include the cases when the forces on the bit are obtained from surface drilling dynamics measurements, such as WOB and TOB, where the torque is obtained from surface measurements such as the rate of fluid flow through the motor.

The values of the scalars can also be obtained, in relation to other more exotic material symmetries as well such as transverse isotropy, orthotropy, triclinicity, etc.

Let it also be understood that any assumptions placed on the scalars such as B=C or A=1 or B=0 represent degenerative cases of the stated solution.

This system of equations can be modified to obtain scalars using other isotropic representations of the elastic coefficients, where the stress-strain relationship may be expressed using elastic coefficients such as Young's modulus of elasticity and Poisson's ratio and therefore the formulation as shown above in terms of the stiffness coefficients C₁₁ and C₁₂ should not be considered limiting.

Because the cement plugs can be of hundreds of feet in length, drilling this cement casing plug with respect to time may take several minutes or even hours to drill. Therefore, MWD data obtained from of a drill bit interacting with cement may result in multiple measurements of the forces or accelerations, and displacements allowing the population of these equations under different drilling conditions and thereby determining a single set of scalars that satisfy a solution to the homogeneous stress-strain equations.

The application and use of a single set of scalars to transform MWD data to stress and strain can be thought of like this: if the cement properties along the length of the lateral are constant and isotropic, and the drilling conditions vary, which they will, then the variations in the stress and strain experienced by the cement that are caused by the variations in the applied drilling behavior, are understood to fall along a locus of points that defines or otherwise describes the elastic portion of the stress-strain relationship as is given the isotropic stiffness coefficients of the cement or other isotropic material with known mechanical properties.

The homogeneous system of equations can be solved using algebraic techniques to construct the inverse of a matrix, e.g., by using a singular value decomposition (SVD). The SVD describes a family of solutions to the given system of equations in terms of linear combinations of the eigenvectors that correspond to the zero valued eigenvalues of the SVD decomposition. The set of scalars obtained from a solution to the homogeneous equations is given by a linear combination of the eigenvectors, (U, V and W) associated with the zero-valued eigen values as follows:

${\rho }_1\left[\begin{array}{c}{u}_1\\ u_2\\ u_3\\ u_4\\ u_5\\ u_6\end{array}\right]+{\rho }_2\left[\begin{array}{c}{v}_1\\ v_2\\ v_3\\ v_4\\ v_5\\ v_6\end{array}\right]+{\rho }_3\left[\begin{array}{c}{w}_1\\ w_2\\ w_3\\ w_4\\ w_5\\ w_6\end{array}\right]=\left[\begin{array}{c}A\\ B\\ C\\ D\\ E\\ F\end{array}\right]$

A solution for the scalars is obtained by ρ₁=1 and ρ₂=ρ₃=0.

While cement can be represented as an isotropic material, other measurements, like acoustical measurements or sonic measurements of rock velocity can be used to specify, a priori, the absolute values of the mechanical rock properties in terms of stiffness coefficients or other elastic coefficients for a transverse isotropic media or other more exotic media symmetries. In the case of a transverse isotropic media, the stress-strain equations in relation to the scalars are:

$\left[\begin{array}{cccccc}-a_1& 0& 0& d_1C_{11}& d_2C_{12}& d_3C_{13}\\ 0& -a_2& 0& d_1C_{12}& d_2C_{11}& d_3C_{13}\\ 0& 0& -a_3& d_1C_{13}& d_2C_{13}& d_3C_{33}\end{array}\right]\left[\begin{array}{c}A\\ B\\ C\\ D\\ E\\ F\end{array}\right]=\left[\begin{array}{c}0\\ 0\\ 0\end{array}\right]$

This stress-strain relationship represents a homogeneous system of linear equations that can be solved for the scalars (A, B, C, D, E, and F) from measurements while drilling cement. Here the displacements (“d” terms) and the accelerations (“a” terms) are specified respectively in terms of (i) the axial, angular, centripetal and/or lateral displacement of the drill bit and (ii) the axial, angular, centripetal and/or lateral accelerations of the bit when interacting with the cement. Unlike the isotropic case, for transverse isotropic media, it should be appreciated that the terms for the accelerations and displacements are arranged according to the orientation of the drilling well with respect to the axis of material symmetry. For example, laterally drilling a vertically transverse isotropic (VTI) media would place the axial acceleration and axial displacement with respect to the a₁ and d₁ terms.

It would also be possible to represent the “a” terms using MWD of the weight on bit and the torque on bit where the solution to the scalars would represent the effective contact area of the bit. Again, for laterally drilling a VTI media the weight on bit measurement would be used to inform the a₁ term and the scalar A, would be understood to reference the effective area of the cutter exposure acting in the direction of drilling. The scalars obtained in this manner describe how the geometry and configuration of the cutters convert the drilling forces to stresses.

The solutions to the scalars in this example would follow as for the isotropic example where a linear combination of the three eigenvectors associated with the zero eigenvalues obtained from a SVD of the coefficient matrix satisfies the homogenous equations and can be used to obtain the scalar values.

As discussed, a set of scalars that when applied to the motions and accelerations or forces of a drill bit interacting with a rock formation or other material with known values of mechanical properties, and, in particular, a cement, can be used to transform MWD data to obtain stress and strain experienced by the material or rock formation.

In another implementation of the present disclosure, a computer model may be used to describe the stresses and strains of a material with known mechanical properties interacting with a drill bit. These stresses and strains can be obtained using a computer aided drafting of the arrangement and geometries of the cutters in contact with a computer description of an elastic material undergoing elastic deformation when subjected to computer generated forces and motions of the bit. The forces and motions of the bit can be converted to stress and strain with respect to the arrangement and geometry of the cutters on the bit. Accordingly, when actually drilling a material with unknown properties but with the same forces and displacements or stresses and strains as informed by the computer model, those stresses and strains can be used to scale the MWD data and in particular the accelerations and displacements directly to obtain the scalars according to:

$A=\frac{\sigma }{a_1},B=\frac{\tau }{a_2},C=\frac{\tau }{{\alpha }_3},D=\frac{{\varepsilon }_{\mathrm{axial}}}{d_1},E=\frac{{\varepsilon }_{\mathrm{radial}}}{d_2},F=\frac{{\varepsilon }_{\mathrm{angular}}}{d_3}$

where:

the σ and t terms represent one of a stress acting perpendicular or parallel to the axis of drilling and the ε terms represent the axial or one of the angular or linear strains perpendicular to the axis of drilling, and

the “a” terms represent the acceleration of the bit with respect to the axis of drilling and the “d” terms represent the displacements of the bit with respect to the axis of drilling.

Note, the motions and accelerations of the bit are scaled with respect to the corresponding stresses and strains obtained from the computer model of the bit interacting with the rock formation to obtain meaningful transformations of the MWD data to stress and strain.

Next, a further elaboration is provided on the techniques through the application and use of the scalars to (i) not only transform the MWD data to stresses and strains but also, (ii) to obtain a general stress-strain relationship when drilling a rock formation with unknown mechanical rock properties (operation 1650). Using the scalars to transform the MWD data to stress and strain, the isotropic stress-strain relationship is described as follows:

$\left[\begin{array}{cc}{\mathrm{Dd}}_1& \left({\mathrm{Ed}}_2+{\mathrm{Fd}}_3\right)\\ Ed_2& \left({\mathrm{Dd}}_1+{\mathrm{Fd}}_3\right)\\ Fd_3& \left({\mathrm{Dd}}_1+{\mathrm{Ed}}_2\right)\end{array}\right]\left[\begin{array}{c}{C}_{11}\\ C_{12}\end{array}\right]=\left[\begin{array}{c}{\mathrm{Aa}}_1\\ {\mathrm{Ba}}_2\\ {\mathrm{Ca}}_3\end{array}\right]$

FIG. 17 is a diagram depicting the path of an accelerometer 1702 as a drill bit bores a hole along an axis of drilling 1704. The accelerometer may be positioned on or otherwise form a part of an MWD assembly. In any event, the accelerometer is positioned proximate a drill bit. The diagram is an isometric view of a path of the accelerometer as the drill bit rotates clockwise 1708. In this representation, the axial (d₁), lateral (d₂) and centripetal (d₃) displacements when multiplied by the respective scalars, D, E, F are the axial, lateral and centripetal strains, and likewise the axial (a₁), lateral (a₂) and centripetal (a₃) accelerations when multiplied by the respective scalars, A, B, C are the axial, lateral and centripetal stresses.

This system of equations can be solved using well known algebraic techniques to obtain a solution for the isotropic stiffness coefficients (e.g., C₁₁ and C₁₂). Because, the MWD data have been transformed to stress and strain through the application and use of the scalars, the isotropic stiffness coefficients obtain in this manner represent absolute values of the mechanical rock properties.

Note, the coefficients can be applied to obtain the absolute values of mechanical rock properties for other media symmetries such as transverse isotopic media with the media being horizontally transverse isotropic (HTI) or vertically transverse isotropic (VTI) as specified by the axis of material symmetry in relation to the orientation of the drilling well, such as follows:

$\left[\begin{array}{cccc}{\mathrm{Dd}}_1& {\mathrm{Ed}}_2& {\mathrm{Fd}}_3& 0\\ Ed_2& {\mathrm{Dd}}_1& {\mathrm{Fd}}_3& 0\\ 0& 0& \left({\mathrm{Dd}}_1+{\mathrm{Ed}}_2\right)& {\mathrm{Fd}}_3\end{array}\right]\left[\begin{array}{c}{C}_{11}\\ C_{12}\\ C_{13}\\ C_{33}\end{array}\right]=\left[\begin{array}{c}{\mathrm{Aa}}_1\\ {\mathrm{Ba}}_2\\ {\mathrm{Ca}}_3\end{array}\right]$

This technique further advances the application and use of the absolute values of the elastic coefficients in predictable and systematic ways to inform the identification of fractures, bedding planes, material boundaries or other discontinuities that act to separate or other offset rock formations with different mechanical rock properties with respect to axis of media symmetry in relation to the orientation of the drilling well as follows.

Formation Mechanical Quality

The processing and analysis of drilling induced vibrations, which may be captured or otherwise considered as acoustic emissions, while drilling a wellbore through a formation, under implementations disclosed herein, can be used to determine various mechanical rock properties and, based on the measured values of such mechanical rock properties, to classify the mechanical quality of the rock formation being drilled, which may involve classifying the mechanical quality continuously or discretely along one or more portions of the wellbore and otherwise. The mechanical quality generally corresponds to a collection of mechanical rock properties that, when taken together, provide an indication of the susceptibility of a given portion of the formation to various operations and/or may provide an indication, alone or in combination with other information, of the possible production performance of the well and associated formation. Such operations may include, but are not necessarily limited to, drilling and the various aspects of hydraulic fracturing of the formation including proppant selection and concentration, perforation placement, etc. Such information may further be relevant to well spacing, well placement, in-fill drilling, depletion modelling, and order of operations, among other things.

Implementations of the present disclosure may obtain and process drilling vibration information to determine a mechanical quality of rock within a formation and quantify or characterize the mechanical quality according to a mechanical quality index. For example, in certain implementations, the mechanical quality index may indicate whether rock is “good” or “bad” with respect to some aspect of drilling or completing a well. Alternatively, the mechanical quality index may be a numerical score or similar metric. In one specific implementation, mechanical rock quality may correspond to the relative ease with which rock may be drilled and fractured. For example, with respect to drilling, the mechanical quality may be based on some mechanical rock property associated with the strength of the rock (e.g., YME). Similarly, fracturability may be based on some mechanical rock property associated with the degree of layering within the formation (e.g., a metric representative of the brittleness of the formation). In such an implementation, highly drillable and highly fracturable rock may be characterized by the system to have high mechanical quality while rock having both low drillability and fracturability may be characterized by the system to have low or poor mechanical quality. Rock having varying degrees and combinations of drillability and fracturability (e.g., high drillability but low fracturability) may be ascribed various levels of intermediate mechanical quality. The resulting mechanical quality metric may then be used to, among other things, determine a path of a well, select drilling parameters for drilling of the well, select a particular drill bit (e.g., by selecting a drill bit that provides high stress for the particular formation or includes a downhole motor as discussed above), identify locations for fracturing, determine fracturing operation parameters, or other aspects related to drilling and completion of a well. It should be appreciated that the foregoing example in which drillability and fracturability provide the basis for mechanical quality is provided merely as an example. In other implementations, any of drillability, fracturability, or any other characteristic of the formation, alone or in combination, may provide the basis of the mechanical quality assigned to a given portion of formation.

The mechanical rock quality data obtained from data captured (e.g., recorded) during drilling (and possibly supplemented with additional data regarding the formation obtained during drilling or otherwise available) may be used to inform various drilling or completion related operations. For example, in one implementation, the mechanical rock quality data may be used to control or modify one or more aspects of a fracturing operation for the well or other wells extending through the subterranean formation. Such aspects may include, without limitation, one or more of: determining the locations of fracture stages; determining the spacing of perforations within any stage, which may differ between stages along the length of horizontal well bore; determining the type of fracturing fluid used during fracturing operations; determining the amount of proppant and/or additives included in the fracturing fluid; and determining one or more of the injection rate, injection pressure, and/or injection duration of fracturing fluid. As another example, the mechanical rock quality data may be used to inform one or more aspects of drilling operations of the well or subsequent wells extending through the formation. Such aspects may include, without limitation, one or more of: drill bit selection, drilling path/direction; bit rotational speed; penetration rate; and weight-on-bit.

In one example, the vibration data, which may be from one or more accelerometers positioned on a component of a drilling assembly, is collected and stored during drilling. The vibration data may then be captured by some component of the drilling assembly and mechanical rock quality data may be collected and processed by a computing system to facilitate analysis of the mechanical rock quality data. The computing system may be any device sufficient to access the data and process it as described herein. In one example implementation, the computing system may be a server or other computing system. In the case of a server, it may be accessible by some form of computing device, in communication with the server, that provides a user interface to facilitate interaction with and presentation of the information from the data processed on the server. The computing system may also be in communication with or integrated with and be configured to automatically control one or more systems to perform well-related operations based on the mechanical rock quality data. For example, in certain implementations the computing system may be communicatively coupled to one or more of a hydraulic fracturing system for controlling fracturing operations and/or a drilling system for controlling drilling operations.

FIG. 18 is a representative illustration of operative components of a drill string assembly 1800. The drill string assembly 1800 includes a drill string 1802 to which one or more accelerometers, such as accelerometer 1804, are coupled. Although not illustrated, the drill string assembly 1800 further includes a drill bit disposed on a distal end of the drill string 1802 for drilling through a subterranean formation. More specifically, during drilling operations, weight is applied to the drill string 1802 such that the bit contacts a portion of the subterranean formation. The drill string 1802 is then rotated (such as indicated by arrows 1806A, 1806B) such that the drill bit engages and cuts into the subterranean formation. The accelerometer may be a component of a measurement while drilling (MWD) assembly or may otherwise be integrated with some component of the drill string assembly. In one possible implementation, the accelerometers are positioned proximate the drill bit and between the drill bit and any mud motor positioned behind the drill bit. For example, the accelerometer, a memory, and a processor communicatively coupled to each of the accelerometer and the memory may be part of the measurement components 36 illustrated in FIG. 2B. During operation, the processor may receive accelerometer data from the accelerometer, process the data, and store the data in the memory. In certain implementations, the accelerometer, memory, and processor may be integrated into a computing system mounted on the drill string 1802. For example, such computing systems may generally corresponding to the computing system illustrated in FIG. 25 and discussed below in further detail, with some of the components omitted if unnecessary in certain applications (e.g., human input device(s) 2514, human output device(s) 2516 may not be necessary in certain downhole applications).

During drilling operations, the accelerometer 1804 measures vibrations of the drill string assembly 1802. As indicated in FIG. 18, the accelerometer 1804 may be a multi-axial accelerometer configured to measure vibrations in each of a torsional direction (as indicated by arrow 1810 and as also referred to herein as the centripetal direction), a linear direction (as indicated by arrow 1812 (extending outwardly from the page)), and a lateral direction (as indicated by arrow 1814). In specific examples, the accelerometer may be a three- or six-axis accelerometer sensor arranged to capture axial, lateral and centripetal accelerations, collectively understood as drill bit vibrations, from the drill bit interacting with the formation during drilling operations. In other implementations, an array of multi-axial accelerometers may be implemented. For example, and without limitation, two two-channel accelerometers may be coupled together to form a four channel accelerometer. Such arrays of accelerometers may be combined to separate angular and linear motions of the bit. It should also be appreciated that, in certain implementations, measurements of the rotation (including any of rotational position, speed, and/or acceleration) of the drill string 1802 may be provided by a gyroscope 1816 positioned in-line with the drill string 1802 and which may be used in addition to one or more additional accelerometers for measuring other dynamics (e.g., axial acceleration) of the drill string 1802 as described herein.

As previously discussed in detail, the vibrations captured by the accelerometer 1804 may be processed to obtain various mechanical rock properties of the subterranean formation. Such mechanical rock properties may include, without limitation, stresses, strains, stiffness coefficients, elastic coefficients, unconfined compressive strength (UCS) (or other measurement of strength), process zone strength, Young's Modulus of Elasticity (YME), and Poisson's Ratio (PR).

After deriving the mechanical rock properties from the vibration data, the mechanical rock properties may be used to determine additional characteristics of the subterranean formation at the location where the corresponding mechanical rock property data was obtained. Such characteristics may correspond to the mechanical quality of the subterranean formation, where the mechanical quality generally refers to the relative susceptibility of the subterranean formation to various mechanical operations.

In one specific example, mechanical quality is a function of the subterranean formations susceptibility to each of drilling and fracturing operations. In general, subterranean formations exhibiting relatively low strength (e.g., as measured using UCS) and relatively high brittleness (e.g., as measured using a brittleness index) are highly susceptible to both drilling and fracturing operations and may be characterized as having high mechanical quality. Similarly, rock exhibiting high strength and low brittleness may be characterized as having low mechanical quality and rock exhibiting some mix of favorable strength and brittleness may be characterized as having intermediate quality. In other implementations, other characteristics regarding the general fracturability of the formation may be used instead of brittleness including, without limitation, one or more of laminarity, complexity, anisotropy, or breakdown.

During drilling operations, the mechanical quality may be assessed at various locations along the wellbore. Accordingly, a profile of the mechanical quality of the formation along and surrounding the wellbore may be established. Such mechanical quality data, which is generally obtained from drilling dynamics measurements (e.g., accelerometer, gyroscope, or strain gauge measurements obtained during drilling), may also be incorporated into or correlated with other formation data to confirm and/or supplement such additional data. Without limitation, the additional data may include seismic data, logging while drilling (LWD) data, measurement while drilling (MWD) data, wireline data, core sample data, production data, tracer data, completion diagnostics data, or any other data related to the formation. The collective data may then be used to inform or control subsequent operations, such as subsequent drilling and/or fracturing operations.

In one specific example, determining the mechanical quality for a given portion of the subterranean formation may include calculating each of a UCS value and a value representative of brittleness for a given location along the wellbore. To do so, the accelerometer 1804 may measure vibrations of the drill string assembly 1800 induced by the drill bit cutting into the subterranean formation during drilling. Based on the vibrations, various parameters may be calculated or measured at different locations and times during the drilling operation. In the following example, such parameters include YME, PR, and zero-frequency level (ZFL), the collection and measurement of each of which are previously described in detail. In other implementations of the present disclosure, however, any suitable parameter derived from measurements of the drill bit vibration obtained during drilling may be used in determining mechanical quality for a portion of the subterranean formation. Other parameters and values that may be used in other implementations to determine mechanical rock quality include, without limitation, strain measurements, stress measurements, and elastic coefficients. Various drilling-related parameters may also be used including, without limitation, any of WOB, TOB, ROP, or any of displacement, velocity, or acceleration in any of the axial, lateral, angular, or torsional (centripetal) directions as measured during drilling.

In general, the UCS of a given rock type is empirically related to its YME. For example, in one example implementation, UCS may be calculated according to the following formula:

UCS=aE^b+c

where E is the YME value obtained during drilling and a, b, and c are coefficients that may be empirically determined, such as through laboratory testing of samples of the formation, from previously collected formation data, or from relevant tables of mechanical rock properties. For example, in one implementation, UCS may be considered to be proportional to the square of YME (e.g., UCS=aE²).

Values or metrics representative of brittleness may be calculated in various ways, however, in at least one implementation, a metric representative of brittleness may be calculated as a function of the volumetric proportion, orientation, and/or distribution of one or more of clay, kerogen, or other organic matter and/or minerals in the subsurface formation. For example, shear modulus may be used as a way of approximating brittleness based on the layering of clay within the formation. In one implementation, calculation of the brittleness metric may first include calculating various shear moduli for the subterranean formation. Such shear moduli may include, without limitation, one or more of an isotropic shear modulus (μ_iso) and a torsional shear modulus (μ_tor).

μ_iso may be calculated as a function of the calculated YME and PR values at a given point in the formation. Additional shear moduli may be calculated based on the various signals and measurements from the accelerometer 1804. For example, μ_tor may be approximated using the following formula:

μ_tor≅A_ang-RMS/ZFL_ang

where A_ang-RMS is the root mean square of the angular acceleration as measured by the accelerometer 1804 and ZFL_ang is the zero-frequency level (ZFL) of the angular displacement measured by the lateral channel of the accelerometer 1804 (and further indicated in FIG. 2A).

The various shear moduli may be used to characterize or otherwise quantify the anisotropy of the subterranean formation. For example, in one implementation, a parameter referred to herein as γ_iso may be defined as the relative anisotropy of the subterranean formation and may range from a value of 0 (e.g., fully isotropic) to 1 (e.g., fully anisotropic). In one example implementation, γ_iso may be calculated as follows:

γ_iso=f(YME,PR, μ)˜μ_iso−μ_tor

The brittleness metric of the subterranean formation may then be quantified based, in part, on the calculated anisotropy of the subterranean formation. In one specific example, a brittleness metric (Br) may be calculated according to the following formula:

Br=1−V_cl(γ_iso)−V_ker

where V_cl is a relative volume of clay in the formation as determined using the value of γ_iso and V_ker is a relative volume of kerogen in the subterranean formation.

Based on the calculated brittleness metric and UCS values, the corresponding portion of the subterranean formation may be assigned a mechanical quality. The particular scale used to assign mechanical quality may vary. For example, mechanical quality for a location in the formation may be a numerical score based on each of the UCS and brittleness metric values calculated for that location. The mechanical quality may also be based on a comparison of such values to similar values calculated for other locations in the formation including other locations along the same wellbore and/or locations along other wellbores within the formation. In another example, mechanical quality for the location may be characterized as having “good” quality, “bad” quality, “intermediate”/“mixed” or variations thereof. For example, a label of “good” may be assigned to locations of the formation exhibiting relatively low UCS and having a relatively high brittleness metric, which are general indications that the formation in that location is easily drilled and easily fractured. Conversely, if UCS is high and the brittleness metric is low, the location may be characterized as have “bad” quality and “intermediate”/“mixed” quality may be assigned for other cases in which favorable UCS or favorable brittleness is observed. As previously noted, use of UCS and a brittleness metric as the basis for mechanical quality is provided herein merely as an example and, in other implementations, mechanical quality may be based on any other suitable property of the subterranean formation derived, at least in part, from measurements obtained during drilling of a wellbore.

The foregoing examples in which mechanical quality is classified by a numerical value or a “good”/“bad” rating system are intended as being illustrative only and should not be regarded as limiting. Rather, mechanical quality may be classified using any suitable system for indicating the susceptibility of portions of the formation to certain drilling, fracturing, and completion operations. Implementations of this disclosure are not limited to any specific number of mechanical quality classifications. For example, one implementation may simply identify locations as having “high” or “low” mechanical quality while others may use a more granular system, e.g., a system that classifies locations as having one of “high”, “high-medium”, “medium”, “medium-low”, and “low” mechanical quality.

Although mechanical quality classification may include assigning locations along the wellbore to particular quality “tiers” or groups (e.g., “good”, “bad”, “medium”), mechanical quality may more generally be expressed using any suitable qualitative or quantitative system. As one alternative to the previously discussed “good”/“bad” classification, locations along the wellbore may instead be assigned a score calculated based on the various mechanical property measurements. For example, each location may be assigned a score from 0 to 100, with 100 indicating the highest quality. Such an approach provides a more continuous quality metric and facilitates the use of quantitative curves or similar visualizations of mechanical quality along the wellbore.

Regardless of how mechanical quality is ultimately indicated, the mechanical quality data may be used to facilitate various well-related operations. For example and without limitation, the mechanical quality data may be processed and analyzed by a control system or presented to a user via a computing device to facilitate, among other things, steering of the drill bit, planning drill paths for and/or controlling drilling direction for subsequent wells, identification and demarcation of fracturing stages, and the like. The mechanical quality information may also be compiled and stored to supplement other information (e.g., seismic data, core sample data, MWD/LWD data, and the like) regarding the subterranean formation. Accordingly, in addition to informing and/or controlling subsequent operations, the mechanical quality information may be used, at least in part, to generate one or more of data tables, maps, graphs, seismic inversions, computer models, AI applications, or other visualizations that may be used for further analysis of the subterranean formation.

As indicated above, the mechanical quality information is generally associated with a particular location within the subterranean formation. Accordingly, implementations of the present disclosure may further include functionality directed to identifying the location of the drill bit when corresponding vibration data is collected. In one example, the drill string may include an MWD or similar tool that collects drill bit dynamics data using one or more sensors (e.g., gyros, accelerometers, etc.) and that may be used to determine the path of the drill bit during drilling. In such implementations, the MWD data may be synchronized or otherwise correlated to the mechanical quality information (or any data underlying the mechanical quality information) such the location within the rock formation from which the mechanical quality information was obtained may be readily ascertained. For example, each of the MWD tool and the electronics associated with collecting and processing vibration/acoustic data may share a common clock. In other implementations, each of the MWD data and the vibration/acoustic data may be transmitted uphole (e.g., by mud pulse telemetry or another suitable communication technique) and correlated based on a known temporal offset between the MWD data and the vibration/acoustic data. It should be appreciated that determining location based on MWD data is provided merely as an example and should not be considered limiting for purposes of the present disclosure. Rather, any suitable tool or technique for locating the drill bit may be used in implementations of this disclosure as can any suitable method for synchronizing or correlating such location data with mechanical quality information or any data from which mechanical quality information may be derived.

FIG. 19 is a flowchart illustrating one method conforming to aspects of the present disclosure. It should be recognized that the elaborate detail and various alternatives and embodiments set out herein may constitute other methods, alone or in combination with that set forth in FIG. 19. Moreover, various steps of the method of FIG. 19, as well as other methods, may be performed within a computer system such as set out in FIG. 16 or may be performed, in whole or in part, in a bottom hole assembly associated with or proximate a bit, such as shown in FIG. 2, may form or be used in steering and may therefore be deployed in the geosteering system illustrated in FIG. 2 or may be deployed in various mechanisms associated with completions.

Referring to FIG. 19, the method involves receiving acoustical signals obtained from one or more sensors positioned on a component of a bottom-hole assembly (operation 1910). The sensors (e.g., accelerometers) are in operable communication with at least one data memory to store the acoustical signals (e.g., vibration data) where the acoustical signals are generated from a drill bit interacting with a rock formation while drilling a wellbore. The method further involves processing the acoustical signals to obtain at least one set of data values representative of a mechanical rock property of the rock formation along the wellbore created by the drill bit interacting with the rock formation for a period of time (operation 1920). After which, the method may further involve determining a mechanical quality of the rock formation based, at least in part, on the set of data values representative of the mechanical rock property (operation 1930). In some instances, the method may further involve using the mechanical quality data to drill the well, complete the well, or otherwise perform well-related operations (operation 1940).

FIG. 20 is an example graphical visualization 2000 generated by the systems and methods set out herein for depicting the mechanical quality data discussed above. In particular, the visualization 2000 includes horizontal axis 2002 indicating measured depth in feet and a vertical axis 2004 indicating a proportion of total mechanical quality classifications. The visualization 2000 further includes a series of vertical bars, such as vertical bar 2006, that summarize mechanical quality classifications for portions of the wellbore. In the specific example of FIG. 20, the wellbore is divided into 20 foot portions and, as a result, each vertical bar represents mechanical quality data for a respective 20 foot increment of the wellbore.

Although illustrated as being grouped into 20 foot portions or “buckets”, it should be appreciated that the mechanical quality classifications may be collected and grouped in any sized interval, including intervals that are greater than or less than 20 feet. For example, in at least one specific implementation, mechanical rock property data may is calculated every 0.5 feet of the wellbore. Accordingly, mechanical quality data may be determined at similar 0.5 intervals. In other words, mechanical quality data may be provided for each location along a wellbore for which mechanical rock property measurements are obtained and/or may be combined into any suitable increment of the wellbore.

For purposes of the current example, the mechanical quality data is classified using a two-variable, binary classification system in which the variables of interest are brittleness and UCS and each of brittleness and UCS are assessed as being high or low for each location along the wellbore at which brittleness and UCS are calculated. As previously discussed, high brittleness and low UCS are typically desirable as they generally correspond to ease of fracturing and ease of drilling, respectively. To that end, locations at which brittleness is above a threshold brittleness value and UCS is below a threshold UCS value may be classified as having the highest quality. Other locations at which only one of brittleness exceeds the threshold brittleness value or UCS falls below the threshold UCS value

Each vertical bar may provide a cumulative representation of the mechanical quality data for its respective portion of the wellbore. Referring specifically to vertical bar 2006, the vertical bar 2006 as a whole represents the mechanical quality data for the portion of the wellbore corresponding to a measured depth (e.g., depth along the wellbore) of 5860 feet to 5880 feet, with portions 2008A-C representing different quality classifications obtained within that depth range. Each portion 2008A-C may be assigned a particular color, fill pattern, or other visual property to facilitate distinguishing between portions. In particular, bar portion 2008A indicates that approximately 50% of samples obtained along a section of the wellbore between 5860 feet and 5880 feet indicated both high brittleness and low UCS, approximately 15% of samples within the same section indicated low brittleness and low UCS, and approximately 20% of samples within the same section indicated high brittleness and high UCS. The remaining 15% of samples, by default, indicated both low brittleness and high UCS. As illustrated in FIG. 20, such samples may be omitted from the bar 2006. Of course, it is possible to represent the fourth mechanical quality zone (low brittleness and high UCS) with some form of distinct representation.

The visualization 2000 may be used to identify various trends within the formation and, as a result, may be used to inform subsequent drilling, fracturing, or other operations. For example, in the visualization 2000 relatively high mechanical quality was observed between 5800 feet and 5920 feet and between 6120 feet and 6420 feet. The definition of “high” or “highest” mechanical quality may vary from application to application. For example, in certain applications, the portions of the wellbore having the highest mechanical quality may be identified as those having the highest overall proportion of measurements indicated both high brittleness and low UCS (e.g., as indicated by the solid portions of the bars of the visualization 2000). In another example, portions of the wellbore exhibiting the highest mechanical quality may be instead defined as those sections exhibiting the highest proportion of measurements indicating at least one of high brittleness, low UCS, or both high brittleness and low UCS (e.g., as indicated by the overall height of the bars of the visualization 2000). In the example visualization 2000, the portions of the wellbore from 5800-5920 feet and 6120-6420 would be considered to have the highest mechanical quality under both of these example standards, suggesting that such ranges may be particularly susceptible to fracturing operations or may correspond to portions of the formation through which subsequent drilling operations may be directed. In contrast, areas having poor mechanical quality, such as the range between 5980 feet and 6080 feet may be avoided or may be targets for additional treatment, such as acidization, prior to any subsequent completion operations. Again, what constitutes “poor” mechanical quality may vary but, in this example, generally correspond to portions of the wellbore for which relatively low proportions of measurements indicate at least one of brittleness and UCS being within a desirable range. While the foregoing example describes fracturing portions of the well represented by the visualization 2000, it should be appreciated that the data underlying the visualization 2000 may also be used to inform drilling, fracturing, or other operations for other wells in the formation.

A bar-type graph as in the visualization 2000 of FIG. 20 is simply one approach to presenting the mechanical quality data. For example, in other implementations, the bars may be replaced with lines, a heat map, or other visual representation of the mechanical quality data. Moreover, visualizations in accordance with the present disclosure may include more than two variables and more than two possible values per variable. In such implementations, the bar graph (or other indicator) may be further subdivided into additional portions to provide additional detail regarding the underlying mechanical quality data.

In certain implementations, the visualization 2000 may be presented through a user interface of a computing device. Through the user interface, a user may be able to modify parameters of the visualization (e.g., the colors or patterns used to indicate the different quality classifications), drill into or out of the data (e.g., by changing the range of the measured depth being displayed), modify the variable illustrated in the visualization, modify the threshold values used to classify mechanical quality data, and perform other similar tasks. In addition to the graphical depiction shown in FIG. 20, it is possible to further include programmable or user selectable or settable thresholds such that indications of high and low quality, for example, may further be depicted in the user interface by comparing the proportional classifications or discrete attributes of any given proportional classification to any applicable threshold or thresholds.

FIG. 21 below is a block diagram of a machine in the example form of a computer system 2100 within which instructions 2106 for causing the machine to perform any one or more of the methodologies, and various combinations of the same, discussed herein may be executed by one or more hardware processors 2102. In various embodiments, the machine operates as a standalone device or may be connected (e.g., networked) to other machines. In a networked deployment, the machine may operate in the capacity of a server or a client machine in server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. Further, while only a single machine is illustrated, the term “machine” shall also be taken to include any collection of machines or controllers that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein.

As depicted in FIG. 21, the example computing system 2100 may include one or more hardware processors 2102, one or more data storage devices 2104, one or more memory devices 2108, and/or one or more input/output devices 2110. Each of these components may include one or more integrated circuits (ICs) (including, but not limited to, field-programmable gate arrays (FPGAs), application-specific ICs (ASICs), and so on), as well as more discrete components, such as transistors, resistors, capacitors, inductors, transformers, and the like. Various ones of these components may communicate with one another by way of one or more communication buses, point-to-point communication paths, or other communication means not explicitly depicted in FIG. 21. Additionally, other devices or components, such as, for example, various peripheral controllers (e.g., an input/output controller, a memory controller, a data storage device controller, a graphics processing unit (GPU), and so on), a power supply, one or more ventilation fans, and an enclosure for encompassing the various components, may be included in the example computing system 200, but are not explicitly depicted in FIG. 21 or discussed further herein.

The at least one hardware processor 2102 may include, for example, a central processing unit (CPU), a microprocessor, a microcontroller, and/or a digital signal processor (DSP). Further, one or more hardware processors 2102 may include one or more execution cores capable of executing instructions and performing operations in parallel with each other. In some instances, the hardware processor is within the bit sub, and others it is part of another separate processing system.

The one or more data storage devices 2104 may include any non-volatile data storage device capable of storing the executable instructions 2106 and/or other data generated or employed within the example computing system 2100. In some examples, the one or more data storage devices 2104 may also include an operating system (OS) that manages the various components of the example computing system 2100 and through which application programs or other software may be executed. Thus, in some embodiments, the executable instructions 2106 may include instructions of both application programs and the operating system. Examples of the data storage devices 2104 may include, but are not limited to, magnetic disk drives, optical disk drives, solid state drives (SSDs), flash drives, and so on, and may include either or both removable data storage media (e.g., Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc Read-Only Memory (DVD-ROM), magneto-optical disks, flash drives, and so on) and non-removable data storage media (e.g., internal magnetic hard disks, SSDs, and so on).

The one or more memory devices 2108 may include, in some examples, both volatile memory (such as, for example, dynamic random access memory (DRAM), static random access memory (SRAM), and so on), and non-volatile memory (e.g., read-only memory (ROM), flash memory, and the like). In one embodiment, a ROM may be utilized to store a basic input/output system (BIOS) to facilitate communication between an operating system and the various components of the example computing system 2100. In some examples, DRAM and/or other rewritable memory devices may be employed to store portions of the executable instructions 2106, as well as data accessed via the executable instructions 2106, at least on a temporary basis. In some examples, one or more of the memory devices 2108 may be located within the same integrated circuits as the one or more hardware processors 2102 to facilitate more rapid access to the executable instructions 2106 and/or data stored therein.

The one or more data storage devices 2104 and/or the one or more memory devices 2108 may be referred to as one or more machine-readable media, which may include a single medium or multiple media that store the one or more executable instructions 2106 or data structures. The term “machine-readable medium” shall also be taken to include any tangible medium that is capable of storing, encoding, or carrying instructions 2106 for execution by the machine and that cause the machine to perform any one or more of the methodologies of the present invention, or that is capable of storing, encoding, or carrying data structures utilized by or associated with such instructions 2106.

The input/output devices 2110 may include one or more communication interface devices 2112, human input devices 2114, human output devices 2116, and environment transducer devices 2118. The one or more communication interface devices 2112 may be configured to transmit and/or receive information between the example computing system 2100 and other machines or devices by way of one or more wired or wireless communication networks or connections. The information may include data that is provided as input to, or generated as output from, the example computing device 2100, and/or may include at least a portion of the executable instructions 2106. Examples of such networks or connections may include, but are not limited to, Universal Serial Bus (USB), Ethernet, Wi-Fi®, Bluetooth®, Near Field Communication (NFC), and so on. One or more such communication interface devices 2110 may be utilized to communicate one or more other machines, either directly over a point-to-point communication path or over another communication means. Further, one or more wireless communication interface devices 2112, as well as one or more environment transducer devices 2118 described below, may employ an antenna for electromagnetic signal transmission and/or reception. In some examples, an antenna may be employed to receive Global Positioning System (GPS) data to facilitate determination of a location of the machine or another device.

In some embodiments, the one or more human input devices 2114 may convert a human-generated signal, such as, for example, human voice, physical movement, physical touch or pressure, and the like, into electrical signals as input data for the example computing system 2100. The human input devices 2114 may include, for example, a keyboard, a mouse, a joystick, a camera, a microphone, a touch-sensitive display screen (“touchscreen”), a positional sensor, an orientation sensor, a gravitational sensor, an inertial sensor, an accelerometer, and/or the like.

The human output devices may convert electrical signals into signals that may be sensed as output by a human, such as sound, light, and/or touch. The human output devices 2116 may include, for example, a display monitor or touchscreen, a speaker, a tactile and/or haptic output device, and/or so on.

The one or more environment transducer devices 2118 may include a device that converts one form of energy or signal into another, such as from an electrical signal generated within the example computing system 2100 to another type of signal, and/or vice-versa. Further, the transducers 2118 may be incorporated within the computing system 2100, as illustrated in FIG. 21, or may be coupled thereto in a wired or wireless manner. In some embodiments, one or more environment transducer devices 2118 may sense characteristics or aspects of an environment local to or remote from the example computing device 2100, such as, for example, light, sound, temperature, pressure, magnetic field, electric field, chemical properties, physical movement, orientation, acceleration, gravity, and so on. Further, in some embodiments, one or more environment transducer devices 2118 may generate signals to impose some effect on the environment either local to or remote from the example computing device 2100, such as, for example, physical movement of some object (e.g., a mechanical actuator), receiving or processing accelerometer data, strain gauge data, and the like.

Claims

1. A method of classifying mechanical quality of material in subterranean formations comprising:

with a processor, processing signal data to compute displacements of a drill bit, the displacements resulting from interaction of the drill bit with a material of a subterranean formation during drilling of a wellbore through the subterranean formation;

processing the displacements to compute a property of the material along the wellbore; and

generating a mechanical quality metric for a location within the material based on the computed property.

2. The method of claim 1, wherein:

the signal data includes acoustical signal data,

the signal data is stored on a non-transitory computer readable medium,

the signal data is measured using a sensor of a bottom hole assembly in operable communication with the non-transitory computer readable medium to store the signal data, and

processing the signal data comprises processing the acoustical signal data to obtain a property value representative of the property.

3. The method of claim 2, wherein the signal data includes signal data corresponding to at least one of an axial acceleration of the drill bit, a lateral acceleration of the drill bit, a torsional acceleration of the drill bit, a centripetal acceleration of the drill bit, or a rotational acceleration of the drill bit.

4. The method of claim 2, wherein processing the acoustical signal data to obtain the property comprises computing at least one of Poisson's ratio or Young's modulus of elasticity using the acoustical signal data.

5. The method of claim 2, wherein processing the signal data to compute the property value comprises computing shear modulus using the signal data.

6. The method of claim 2, wherein generating the mechanical quality metric includes calculating unconfined compressive strength (UCS) at the one or more locations using the acoustical signal data.

7. The method of claim 6, wherein UCS is calculated as:

UCS=aEb+c

where: a, b, and c are empirically obtained coefficients; and E is Young's modulus of elasticity.

8. The method of claim 2, wherein generating the mechanical quality metric includes calculating a brittleness metric at the one or more locations based on the acoustical signal data.

9. The method of claim 8, wherein the brittleness metric is calculated as:

Br=1−Vcl(γiso)−Vker

where: Br is the brittleness metric, Vcl is a relative volume of clay in the material, γiso is a relative measure of anisotropy of the material, and Vker is a relative volume of kerogen in the material;

γiso is calculated as: γiso=μiso−μtor

where: μiso is isotropic shear modulus, and μtor is torsional shear modulus; and

μtor is calculated as: μtor≅Aang-RMS/ZFLang

where: Aang-RMS is a root mean square value of the acoustical signal data corresponding to an angular acceleration, and ZFLang is a zero-frequency level of an angular displacement spectra.

10. The method of claim 2, wherein generating the mechanical quality metric includes calculating an anisotropy metric for the one or more locations.

11. The method of claim 2 further comprising performing at least one well operation based on the mechanical quality metric at the location, wherein the at least one well operation includes at least one of a drilling operation or a fracturing operation.

12. A method of evaluating mechanical quality of subterranean formations comprising:

using a computing processor operatively coupled to a non-transitory computer readable memory, processing sensor signal data captured during drilling of a wellbore through a subterranean formation using a drill bit, wherein processing the sensor signal data computes property values for a mechanical material property at a sequence of locations in the subterranean formation along the wellbore;

processing the property values to generate a mechanical quality metric for each location of the sequence of locations; and

storing the mechanical quality metrics in the non-transitory computer readable memory.

13. The method of claim 12 further comprising processing location data from a measurement while drilling (MWD) tool captured during drilling of the wellbore to correlate the property values to the sequence of locations.

14. The method of claim 13, wherein storing the mechanical quality data for the sequence of locations comprises aggregating the mechanical metrics for a consecutive subset of locations of the sequence of locations and storing an aggregated mechanical metric for a range corresponding to the consecutive subset of locations.

15. The method of claim 12, wherein the sensor signal data is acoustical signal data.

16. The method of claim 12, wherein the sensor signal data is provided by one or more sensors of a bottom hole assembly.

17. The method of claim 12, wherein the mechanical material property includes Young's modulus of elasticity (YME) and processing the property values to obtain the mechanical quality metric for each location of the sequence of locations comprises calculating unconfined compressive strength (UCS) using the YME value of the location.

18. The method of claim 12, wherein the mechanical material property includes Young's modulus of elasticity (YME) and processing the at least one set of data values to obtain the mechanical quality data further comprises, for each location of the sequence of locations, calculating a brittleness metric of the location based on YME.

19. A method of classifying mechanical quality of material in subterranean formations comprising:

using a computing processor operatively coupled to a non-transitory computer readable memory, computing property values for a mechanical material property at a sequence of locations in the subterranean formation along the wellbore;

processing the property values to generate a mechanical quality metric for each location of the sequence of locations; and

storing the mechanical quality metrics in the non-transitory computer readable memory.

20. The method of claim 19, wherein generating the mechanical quality metric includes calculating unconfined compressive strength (UCS) at each location of the sequence of locations.

21. The method of claim 19, wherein generating the mechanical quality metric includes calculating a brittleness metric at each location of the sequence of locations.

22. The method of claim 19, wherein generating the mechanical quality metric includes calculating an anisotropy metric at each location of the sequence of locations.

23. The method of claim 19 further comprising performing at least one well operation based on the mechanical quality metrics.

24. A tangible, non-transitory, computer-readable media having instructions encoded thereon, the instructions, when executed by a processor, are operable to:

compute property values for a mechanical material property at a sequence of locations in the subterranean formation along the wellbore;

process the property values to generate a mechanical quality metric for each location of the sequence of locations; and

store the mechanical quality metrics in a non-transitory computer readable memory.

25. The tangible, non-transitory, computer-readable media of claim 24, wherein the property values for the mechanical material property are computed from vibration of a drill bit measured during interaction of the drill bit with the subterranean formation while drilling the wellbore.