Method of calibrating fracture geometry to microseismic events

Abstract

A method of performing a fracture operation is provided at a wellsite. The wellsite is positioned about a subterranean formation having a wellbore therethrough and a complex fracture network therein. The complex fracture network includes natural fractures, and the wellsite stimulated by injection of an injection fluid with proppant into the complex fracture network. The method involves generating wellsite data comprising measurements of microseismic events of the subterranean formation, modeling a hydraulic fracture network and a discrete fracture network of the complex fracture network based on the wellsite data, and performing a seismic moment operation. The method involves determining an actual seismic moment density based on the wellsite data and a predicted seismic moment density based on shear and tensile components of the simulated hydraulic fracture network, and calibrating the discrete fracture network based on a comparison of the predicted moment density and the actual moment density.

Description

BACKGROUND

The present disclosure relates generally to methods and systems for performing wellsite operations. More particularly, this disclosure is directed to methods and systems for performing fracture operations, such as investigating subterranean formations and characterizing hydraulic fracture networks in a subterranean formation.

In order to facilitate the recovery of hydrocarbons from oil and gas wells, the subterranean formations surrounding such wells can be hydraulically fractured. Hydraulic fracturing may be used to create cracks in subsurface formations to allow oil or gas to move toward the well. A formation is fractured by introducing a specially engineered fluid (referred to as “fracturing fluid” or “fracturing slurry” herein) at high pressure and high flow rates into the formation through one or more wellbores. Hydraulic fractures may extend away from the wellbore hundreds of feet in two opposing directions according to the natural stresses within the formation. Under certain circumstances, they may form a complex fracture network. Complex fracture networks can include induced hydraulic fractures and natural fractures, which may or may not intersect, along multiple azimuths, in multiple planes and directions, and in multiple regions.

Patterns of hydraulic fractures created by the fracturing stimulation may be complex and may form a fracture network as indicated by a distribution of associated microseismic events. Complex hydraulic fracture networks have been developed to represent the created hydraulic fractures. Examples of fracture techniques are provided in U.S. Pat. Nos. 6,101,447, 7,363,162, 7,788,074, 20080133186, 20100138196, and 20100250215.

SUMMARY

In at least one aspect, the present disclosure relates to methods of performing a fracture operation at a wellsite. The wellsite is positioned about a subterranean formation having a wellbore therethrough and a fracture network therein. The fracture network has natural fractures therein. The wellsite may be stimulated by injection of an injection fluid with proppant into the fracture network. The method involves obtaining wellsite data including natural fracture parameters of the natural fractures and obtaining a mechanical earth model of the subterranean formation and generating a hydraulic fracture growth pattern for the fracture network over time. The generating involves extending hydraulic fractures from the wellbore and into the fracture network of the subterranean formation to form a hydraulic fracture network including the natural fractures and the hydraulic fractures, determining hydraulic fracture parameters of the hydraulic fractures after the extending, determining transport parameters for the proppant passing through the hydraulic fracture network, and determining fracture dimensions of the hydraulic fractures from the determined hydraulic fracture parameters, the determined transport parameters and the mechanical earth model. The method also involves performing stress shadowing on the hydraulic fractures to determine stress interference between the hydraulic fractures and repeating the generating based on the determined stress interference.

If the hydraulic fracture encounters a natural fracture, the method may also involve determining the crossing behavior between the hydraulic fractures and an encountered fracture based on the determined stress interference, and the repeating may involve repeating the generating based on the determined stress interference and the crossing behavior. The method may also involve stimulating the wellsite by injection of an injection fluid with proppant into the fracture network.

The method may also involve, if the hydraulic fracture encounters a natural fracture, determining the crossing behavior at the encountered natural fracture, and the repeating involves repeating the generating based on the determined stress interference and the crossing behavior. The fracture growth pattern may be altered or unaltered by the crossing behavior. A fracture pressure of the hydraulic fracture network may be greater than a stress acting on the encountered fracture, and the fracture growth pattern may propagate along the encountered fracture. The fracture growth pattern may continue to propagate along the encountered fracture until an end of the natural fracture is reached. The fracture growth pattern may change direction at the end of the natural fracture, and the fracture growth pattern may extend in a direction normal to a minimum stress at the end of the natural fracture. The fracture growth pattern may propagate normal to a local principal stress according to the stress shadowing.

The stress shadowing may involve performing displacement discontinuity for each of the hydraulic fractures. The stress shadowing may involve performing stress shadowing about multiple wellbores of a wellsite and repeating the generating using the stress shadowing performed on the multiple wellbores. The stress shadowing may involve performing stress shadowing at multiple stimulation stages in the wellbore.

The method may also involve validating the fracture growth pattern. The validating may involve comparing the fracture growth pattern with at least one simulation of stimulation of the fracture network.

The extending may involve extending the hydraulic fractures along a fracture growth pattern based on the natural fracture parameters and a minimum stress and a maximum stress on the subterranean formation. The determining fracture dimensions may include one of evaluating seismic measurements, ant tracking, sonic measurements, geological measurements and combinations thereof. The wellsite data may include at least one of geological, petrophysical, geomechanical, log measurements, completion, historical and combinations thereof. The natural fracture parameters may be generated by one of observing borehole imaging logs, estimating fracture dimensions from wellbore measurements, obtaining microseismic images, and combinations thereof.

In another aspect, the disclosure relates to a method of performing a fracture operation at a wellsite positioned about a subterranean formation having a wellbore therethrough and a fracture network therein, with the fracture network including natural fractures, and with the wellsite stimulated by injection of an injection fluid with proppant into the fracture network. The method involves obtaining wellsite data including natural fracture parameters of the natural fractures and obtaining a mechanical earth model of the subterranean formation, generating a hydraulic fracture growth pattern for the fracture network over time, performing interpretation of microseismicity on the hydraulic fractures to determine stress interference between the hydraulic fractures, and repeating the generating based on the determined stress interference. The generating involves extending hydraulic fractures from the wellbore and into the fracture network of the subterranean formation to form a hydraulic fracture network including the natural fractures and the hydraulic fractures, determining hydraulic fracture parameters of the hydraulic fractures after the extending, determining transport parameters for the proppant passing through the hydraulic fracture network, and determining fracture dimensions of the hydraulic fractures from the determined hydraulic fracture parameters, the determined transport parameters, and the mechanical earth model.

In another aspect, a method of performing a fracture operation at a wellsite positioned about a subterranean formation having a wellbore therethrough and a fracture network therein is provided. The fracture network includes natural fractures, and the wellsite is stimulated by injection of an injection fluid with proppant into the fracture network. The method involves generating wellsite data including natural fracture parameters of the natural fractures and obtaining measurements of microseismic events of the subterranean formation, modeling hydraulic fractures of the fracture network based on the wellsite data and defining a hydraulic fracture geometry of the hydraulic fractures, generating a stress field of the hydraulic fractures using a geomechanical model based on the wellsite data, determining shear failure parameters including a failure envelope and a stress state about the fracture network, determining a location of shear failure of the fracture network from the failure envelope and the stress state, and calibrating the hydraulic fracture geometry by comparing the modeled hydraulic fractures and the locations of shear failure against the measured microseismic events. The method may also involve measuring the wellsite data and the microseismic events at the wellsite, adjusting the natural fracture parameters operation based on the calibrating, performing a stimulation operation including stimulating the wellsite by injecting the injection fluid into the fracture network, and/or adjusting the stimulation operation based on the calibrating.

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

In at least one aspect, the disclosure relates to a method of performing a microseismic fracture operation for a wellsite having a subterranean formation with a complex fracture network therein. The fracture network includes natural fractures, and the wellsite is stimulated by injection of an injection fluid with proppant into the fracture network. The method involves generating wellsite data including measurements of microseismic events of the subterranean formation, modeling a hydraulic fracture network and a discrete fracture network of the complex fracture network based on the wellsite data, and performing a seismic moment operation. The performing involves determining an actual seismic moment density based on the wellsite data and a predicted seismic moment density based on shear and tensile components of the simulated hydraulic fracture network, and calibrating the discrete fracture network based on a comparison of the predicted moment density and the actual moment density.

In another aspect, the disclosure relates to a method of performing a fracture operation at a wellsite. The wellsite is positioned about a subterranean formation having a wellbore therethrough and a complex fracture network therein. The fracture network includes natural fractures, and the wellsite is stimulated by injection of an injection fluid with proppant into the fracture network. The method involves generating wellsite data including measurements of microseismic events of the subterranean formation, modeling a hydraulic fracture network and a discrete fracture network of the complex fracture network based on the wellsite data, and performing a seismic moment operation. The performing involves determining an actual seismic moment density based on the wellsite data, determining a predicted moment density by defining the shear and tensile components of the simulated hydraulic fracture network and converting the shear and tensile components of the simulated hydraulic fracture network, and calibrating the discrete fracture network based on a comparison of the predicted moment density and the actual moment density.

Finally, in another aspect, the disclosure relates to a method of performing a fracture operation at a wellsite. The wellsite is positioned about a subterranean formation having a wellbore therethrough and a fracture network therein, and the fracture network including natural fractures. The method involves stimulating the wellsite by injecting the injection fluid with proppant into the fracture network, generating wellsite data including measurements of microseismic events of the subterranean formation, modeling a hydraulic fracture network and a discrete fracture network of the complex fracture network based on the wellsite data, and performing a seismic moment operation. The performing involves determining an actual seismic moment density based on the wellsite data and a modeled seismic moment density based on shear and tensile components of the simulated hydraulic fracture network, and calibrating the discrete fracture network based on a comparison of the predicted moment density and the actual moment density. The method also involves adjusting the stimulation operation based on the calibrating.

In another aspect, the disclosure relates to a method of performing a microseismic fracture operation for a wellsite having a subterranean formation with a fracture network therein involving describing a relationship between microseismic events of the complex fracture network of the subterranean formation, generating a discrete fracture network including discrete fractures from the complex fracture network, determining fracture attributes of the discrete fractures, and determining an estimated production rate based on the fracture attributes.

Finally, in another aspect, the disclosure relates to a system for performing a microseismic fracture operation for a wellsite having a subterranean formation with a fracture network therein.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the system and method for characterizing wellbore stresses and/or microseismic fracture techniques are described with reference to the following figures. The same numbers are used throughout the figures to reference like features and components. Implementations of various technologies will hereafter be described with reference to the accompanying drawings. It should be understood, however, that the accompanying drawings illustrate only the various implementations described herein and are not meant to limit the scope of various technologies described herein.

FIG. 1.1 is a schematic illustration of a hydraulic fracturing site depicting a fracture operation;

FIG. 1.2 is a schematic illustration of a hydraulic fracture site with microseismic events depicted thereon;

FIG. 2 is a schematic illustration of a 2D fracture;

FIG. 3 is a schematic illustration of a stress shadow effect;

FIG. 4 is a schematic illustration comparing 2D Displacement Discontinuity Method (DDM) and Flac3D for two parallel straight fractures;

FIGS. 5.1-5.3 are graphs illustrating 2D DDM and Flac3D of extended fractures for stresses in various positions;

FIGS. 6.1-6.2 are graphs depicting propagation paths for two initially parallel fractures in isotropic and anisotropic stress fields, respectively;

FIGS. 7.1-7.2 are graphs depicting propagation paths for two initially offset fractures in isotropic and anisotropic stress fields, respectively;

FIG. 8 is a schematic illustration of transverse parallel fractures along a horizontal well;

FIG. 9 is a graph depicting lengths for five parallel fractures;

FIG. 10 is a schematic diagram depicting UFM fracture geometry and width for the parallel fractures of FIG. 9;

FIGS. 11.1-11.2 are schematic diagrams depicting fracture geometry for a high perforation friction case and a large fracture spacing case, respectively;

FIG. 12 is a graph depicting microseismic mapping;

FIGS. 13.1-13.4 are schematic diagrams illustrating a simulated fracture network compared to the microseismic measurements for stages 1-4, respectively;

FIGS. 14.1-14.4 are schematic diagrams depicting a distributed fracture network at various stages;

FIG. 15 is a flow chart depicting a method of performing a fracture operation;

FIGS. 16.1-16.4 are schematic illustrations depicting fracture growth about a wellbore during a fracture operation;

FIG. 17 is a schematic diagram depicting stresses applied to a hydraulic fracture;

FIG. 18 is a graph depicting a Mohr-Coulomb envelope and a Mohr circle for a rock medium;

FIGS. 19.1 and 19.2 are schematic diagrams illustrating cross-sectional and map views, respectively, of stresses applied to a hydraulic fracture;

FIG. 20 is a schematic timeline illustrating interaction of hydraulic and natural fractures with seismic events;

FIG. 21 is schematic diagram illustrating a progression of hydraulic and natural fracture interaction;

FIGS. 22.1 and 22.2 are schematic diagrams depicting a discrete fracture network and a fracture network with simulated hydraulic fractures, respectively;

FIGS. 23.1 and 23.2 are flow charts depicting methods of performing a fracture operation;

FIG. 24 is a schematic diagram depicting a fracture plane about a coordinate axis;

FIGS. 25.1-25.5 illustrate simplified, schematic views of an oilfield having subterranean formations containing reservoirs therein in accordance with implementations of various technologies and techniques described herein;

FIG. 26 illustrates a schematic view, partially in cross section, of an oilfield having a plurality of data acquisition tools positioned at various locations along the oilfield for collecting data from the subterranean formations in accordance with implementations of various technologies and techniques described herein;

FIG. 27 illustrates a production system for performing one or more oilfield operations in accordance with implementations of various technologies and techniques described herein;

FIG. 28 is a schematic diagram illustrating shear and tensile stresses on a fracture;

FIGS. 29.1-35.1 are graphs depicting fracture growth with various shear stresses applied thereto, and FIGS. 29.2-35.2 are graphs depicting fracture growth with various tensile stresses applied thereto;

FIG. 36 is a graph depicting microseismic mapping about a fracture network;

FIG. 37 is a graph illustrating a simulated hydraulic fracture network;

FIGS. 38.1 and 38.2 are graphs illustrating stress and strain, respectively, of the simulated hydraulic fracture network of FIG. 37;

FIGS. 39.1 and 39.2 are graphs illustrating modeled deformation of FIGS. 38.1 and 38.2, respectively;

FIG. 40 is a graph illustrating a cumulative seismic moment density;

FIGS. 41.1 is a graph of a portion 41.1 of the simulated hydraulic fracture of FIG. 38.1 depicting shear stress, and FIG. 41.2 is a graph of the simulated hydraulic fracture of FIG. 41.1 modified based on the DFN;

FIG. 42 is a schematic diagram depicting predicted proppant placement;

FIG. 43 is a graph depicting predicted cumulative production of a well;

FIG. 44 is a graph depicting predicted reservoir pressure of a well;

FIGS. 45.1-45.2 are flow charts depicting various method of performing a fracture operation involving seismic moment;

FIGS. 46.1-46.4 are graphs depicting various stages of calibration of a discrete fracture network; and

FIG. 47 is a flow chart depicting a method of calibrating a discrete fracture network.

DETAILED DESCRIPTION

The description that follows includes apparatuses, methods, techniques, and instruction sequences that embody techniques of the inventive subject matter. However, it is understood that the described embodiments may be practiced without these specific details.

I. Fracture Modeling

Models have been developed to understand subsurface fracture networks. The models may consider various factors and/or data, but may not be constrained by accounting for either the amount of pumped fluid or mechanical interactions between fractures and injected fluid and among the fractures. Constrained models may be provided to give a fundamental understanding of involved mechanisms, and may be complex in mathematical description and/or involve computer processing resources and time in order to provide accurate simulations of hydraulic fracture propagation. A constrained model may be configured to perform simulations to consider factors, such as interaction between fractures, over time and under desired conditions.

An unconventional fracture model (UFM) (or complex model) may be used to simulate complex fracture network propagation in a formation with pre-existing natural fractures. Multiple fracture branches can propagate simultaneously and intersect/cross each other. Each open fracture may exert additional stresses on the surrounding rock and adjacent fractures, which may be referred to as “stress shadow” effect. The stress shadow can cause a restriction of fracture parameters (e.g., width), which may lead to, for example, a greater risk of proppant screenout. The stress shadow can also alter the fracture propagation path and affect fracture network patterns. The stress shadow may affect the modeling of the fracture interaction in a complex fracture model.

A method for computing the stress shadow in a complex hydraulic fracture network is presented. The method may be performed based on an enhanced 2D Displacement Discontinuity Method (2D DDM) with correction for finite fracture height or 3D Displacement Discontinuity Method (3D DDM). The computed stress field from 2D DDM may be compared to 3D numerical simulation (3D DDM or flac3D) to determine an approximation for the 3D fracture problem. This stress shadow calculation may be incorporated in the UFM. The results for simple cases of two fractures shows the fractures can either attract or expel each other depending, for example, on their initial relative positions, and may be compared with an independent 2D non-planar hydraulic fracture model.

Additional examples of both planar and complex fractures propagating from multiple perforation clusters are presented, showing that fracture interaction may control the fracture dimension and propagation pattern. In a formation with small stress anisotropy, fracture interaction can lead to dramatic divergence of the fractures as they may tend to repel each other. However, even when stress anisotropy is large and fracture turning due to fracture interaction is limited, stress shadowing may have a strong effect on fracture width, which may affect the injection rate distribution into multiple perforation clusters, and hence overall fracture network geometry and proppant placement.

FIGS. 1.1 and 1.2 depict fracture propagation about a wellsite 100. The wellsite has a wellbore 104 extending from a wellhead 108 at a surface location and through a subterranean formation 102 therebelow. A fracture network 106 extends about the wellbore 104. A pump system 129 is positioned about the wellhead 108 for passing fluid through tubing 142.

The pump system 129 is depicted as being operated by a field operator 127 for recording maintenance and operational data and/or performing the operation in accordance with a prescribed pumping schedule. The pumping system 129 pumps fluid from the surface to the wellbore 104 during the fracture operation.

The pump system 129 may include a water source, such as a plurality of water tanks 131, which feed water to a gel hydration unit 133. The gel hydration unit 133 combines water from the tanks 131 with a gelling agent to form a gel. The gel is then sent to a blender 135 where it is mixed with a proppant from a proppant transport 137 to form a fracturing fluid. The gelling agent may be used to increase the viscosity of the fracturing fluid, and to allow the proppant to be suspended in the fracturing fluid. It may also act as a friction reducing agent to allow higher pump rates with less frictional pressure.

The fracturing fluid is then pumped from the blender 135 to the treatment trucks 120 with plunger pumps as shown by solid lines 143. Each treatment truck 120 receives the fracturing fluid at a low pressure and discharges it to a common manifold 139 (sometimes called a missile trailer or missile) at a high pressure as shown by dashed lines 141. The missile 139 then directs the fracturing fluid from the treatment trucks 120 to the wellbore 104 as shown by solid line 115. One or more treatment trucks 120 may be used to supply fracturing fluid at a desired rate.

Each treatment truck 120 may be normally operated at any rate, such as well under its maximum operating capacity. Operating the treatment trucks 120 under their operating capacity may allow for one to fail and the remaining to be run at a higher speed in order to make up for the absence of the failed pump. A computerized control system may be employed to direct the entire pump system 129 during the fracturing operation.

Various fluids, such as conventional stimulation fluids with proppants, may be used to create fractures. Other fluids, such as viscous gels, “slick water” (which may have a friction reducer (polymer) and water) may also be used to hydraulically fracture shale gas wells. Such “slick water” may be in the form of a thin fluid (e.g., nearly the same viscosity as water) and may be used to create more complex fractures, such as multiple micro-seismic fractures detectable by monitoring.

As also shown in FIGS. 1.1 and 1.2, the fracture network includes fractures located at various positions around the wellbore 104. The various fractures may be natural fractures 144 present before injection of the fluids, or hydraulic fractures 146 generated about the formation 102 during injection. FIG. 1.2 shows a depiction of the fracture network 106 based on microseismic events 148 gathered using conventional means.

Multi-stage stimulation may be the norm for unconventional reservoir development. However, an obstacle to optimizing completions in shale reservoirs may involve a lack of hydraulic fracture models that can properly simulate complex fracture propagation often observed in these formations. A complex fracture network model (or UFM), has been developed (see, e.g., Weng, X., Kresse, O., Wu, R., and Gu, H., Modeling of Hydraulic Fracture Propagation in a Naturally Fractured Formation. Paper SPE 140253 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Tex., USA, January 24-26 (2011) (hereafter “Weng 2011”); Kresse, O., Cohen, C., Weng, X., Wu, R., and Gu, H. 2011 (hereafter “Kresse 2011”). Numerical Modeling of Hydraulic Fracturing in Naturally Fractured Formations. 45th US Rock Mechanics/Geomechanics Symposium, San Francisco, Calif., June 26-29, the entire contents of which are hereby incorporated herein).

Existing models may be used to simulate fracture propagation, rock deformation, and fluid flow in the complex fracture network created during a treatment. The model may also be used to solve the fully coupled problem of fluid flow in the fracture network and the elastic deformation of the fractures, which may have similar assumptions and governing equations as conventional pseudo-3D fracture models. Transport equations may be solved for each component of the fluids and proppants pumped.

Conventional planar fracture models may model various aspects of the fracture network. The provided UFM may also involve the ability to simulate the interaction of hydraulic fractures with pre-existing natural fractures, i.e. determine whether a hydraulic fracture propagates through or is arrested by a natural fracture when they intersect and subsequently propagates along the natural fracture. The branching of the hydraulic fracture at the intersection with the natural fracture may give rise to the development of a complex fracture network.

A crossing model may be extended from Renshaw and Pollard (see, e.g., Renshaw, C. E. and Pollard, D. D. 1995, An Experimentally Verified Criterion for Propagation across Unbounded Frictional Interfaces in Brittle, Linear Elastic Materials. Int. J. Rock Mech. Min. Sci. & Geomech. Abstr., 32: 237-249 (1995) the entire contents of which is hereby incorporated herein) interface crossing criterion, to apply to any intersection angle, and may be developed (see, e.g., Gu, H. and Weng, X. Criterion for Fractures Crossing Frictional Interfaces at Non-orthogonal Angles. 44th US Rock symposium, Salt Lake City, Utah, Jun. 27-30, 2010 (hereafter “Gu and Weng 2010”), the entire contents of which are hereby incorporated by reference herein) and validated against experimental data (see, e.g., Gu, H., Weng, X., Lund, J., Mack, M., Ganguly, U. and Suarez-Rivera R. 2011. Hydraulic Fracture Crossing Natural Fracture at Non-Orthogonal Angles, A Criterion, Its Validation and Applications. Paper SPE 139984 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Tex., January 24-26 (2011) (hereafter “Gu et al. 2011”), the entire contents of which are hereby incorporated by reference herein), and integrated in the UFM.

To properly simulate the propagation of multiple or complex fractures, the fracture model may take into account an interaction among adjacent hydraulic fracture branches, often referred to as the “stress shadow” effect. When a single planar hydraulic fracture is opened under a finite fluid net pressure, it may exert a stress field on the surrounding rock that is proportional to the net pressure.

In the limiting case of an infinitely long vertical fracture of a constant finite height, an analytical expression of the stress field exerted by the open fracture may be provided. See, e.g., Warpinski, N. F. and Teufel, L. W, Influence of Geologic Discontinuities on Hydraulic Fracture Propagation, JPT, February, 209-220 (1987) (hereafter “Warpinski and Teufel”) and Warpinski, N. R., and Branagan, P. T., Altered-Stress Fracturing. SPE JPT, September, 1989, 990-997 (1989), the entire contents of which are hereby incorporated by reference herein. The net pressure (or more precisely, the pressure that produces the given fracture opening) may exert a compressive stress in the direction normal to the fracture on top of the minimum in-situ stress, which may equal the net pressure at the fracture face, but quickly falls off with the distance from the fracture.

At a distance beyond one fracture height, the induced stress may be only a small fraction of the net pressure. Thus, the term “stress shadow” may be used to describe this increase of stress in the region surrounding the fracture. If a second hydraulic fracture is created parallel to an existing open fracture, and if it falls within the “stress shadow” (i.e. the distance to the existing fracture is less than the fracture height), the second fracture may, in effect, see a closure stress greater than the original in-situ stress. As a result, a higher pressure may be used to propagate the fracture, and/or the fracture may have a narrower width, as compared to the corresponding single fracture.

One application of the stress shadow study may involve the design and optimization of the fracture spacing between multiple fractures propagating simultaneously from a horizontal wellbore. In ultra low permeability shale formations, fractures may be closely spaced for effective reservoir drainage. However, the stress shadow effect may prevent a fracture propagating in close vicinity of other fractures (see, e.g., Fisher, M. K., J. R. Heinze, C. D. Harris, B. M. Davidson, C. A. Wright, and K. P. Dunn, Optimizing horizontal completion techniques in the Barnett Shale using microseismic fracture mapping. SPE 90051 presented at the SPE Annual Technical Conference and Exhibition, Houston, 26-29 Sep. 2004, the entire contents of which are hereby incorporated by reference herein in its entirety).

The interference between parallel fractures has been studied in the past (see, e.g., Warpinski and Teufel; Britt, L. K. and Smith, M. B., Horizontal Well Completion, Stimulation Optimization, and Risk Mitigation. Paper SPE 125526 presented at the 2009 SPE Eastern Regional Meeting, Charleston, Sep. 23-25, 2009; Cheng, Y. 2009. Boundary Element Analysis of the Stress Distribution around Multiple Fractures: Implications for the Spacing of Perforation Clusters of Hydraulically Fractured Horizontal Wells. Paper SPE 125769 presented at the 2009 SPE Eastern Regional Meeting, Charleston, Sep. 23-25, 2009; Meyer, B. R. and Bazan, L. W., A Discrete Fracture Network Model for Hydraulically Induced Fractures: Theory, Parametric and Case Studies. Paper SPE 140514 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Tex., USA, Jan. 24-26, 2011; Roussel, N. P. and Sharma, M. M, Optimizing Fracture Spacing and Sequencing in Horizontal-Well Fracturing, SPEPE, May, 2011, pp. 173-184, the entire contents of which are hereby incorporated by reference herein). The studies may involve parallel fractures under static conditions.

An effect of stress shadow may be that the fractures in the middle region of multiple parallel fractures may have smaller width because of the increased compressive stresses from neighboring fractures (see, e.g., Germanovich, L. N., and Astakhov D., Fracture Closure in Extension and Mechanical Interaction of Parallel Joints. J. Geophys. Res., 109, B02208, doi: 10.1029/2002 JB002131 (2004); Olson, J. E., Multi-Fracture Propagation Modeling: Applications to Hydraulic Fracturing in Shales and Tight Sands. 42nd US Rock Mechanics Symposium and 2nd US-Canada Rock Mechanics Symposium, San Francisco, Calif., Jun. 29-Jul. 2, 2008, the entire contents of which are hereby incorporated by reference herein). When multiple fractures are propagating simultaneously, the flow rate distribution into the fractures may be a dynamic process and may be affected by the net pressure of the fractures. The net pressure may be strongly dependent on fracture width, and hence, the stress shadow effect on flow rate distribution and fracture dimensions warrants further study.

The dynamics of simultaneously propagating multiple fractures may also depend on the relative positions of the initial fractures. If the fractures are parallel, e.g. in the case of multiple fractures that are orthogonal to a horizontal wellbore, the fractures may repel each other, resulting in the fractures curving outward. However, if the multiple fractures are arranged in an en echlon pattern, e.g. for fractures initiated from a horizontal wellbore that is not orthogonal to the fracture plane, the interaction between the adjacent fractures may be such that their tips attract each other and even connect (see, e.g., Olson, J. E. Fracture Mechanics Analysis of Joints and Veins. PhD dissertation, Stanford University, San Francisco, Calif. (1990); Yew, C. H., Mear, M. E., Chang, C. C., and Zhang, X. C. On Perforating and Fracturing of Deviated Cased Wellbores. Paper SPE 26514 presented at SPE 68th Annual Technical Conference and Exhibition, Houston, Tex., October 3-6 (1993); Weng, X., Fracture Initiation and Propagation from Deviated Wellbores. Paper SPE 26597 presented at SPE 68th Annual Technical Conference and Exhibition, Houston, Tex., October 3-6 (1993), the entire contents of which are hereby incorporated by reference herein).

When a hydraulic fracture intersects a secondary fracture oriented in a different direction, it may exert an additional closure stress on the secondary fracture that is proportional to the net pressure. This stress may be derived and be taken into account in the fissure opening pressure calculation in the analysis of pressure-dependent leakoff in fissured formation (see, e.g., Nolte, K., Fracturing Pressure Analysis for nonideal behavior. JPT, February 1991, 210-218 (SPE 20704) (1991) (hereafter “Nolte 1991”), the entire contents of which are hereby incorporated by reference herein).

For more complex fractures, a combination of various fracture interactions as discussed above may be present. To properly account for these interactions and remain computationally efficient so it can be incorporated in the complex fracture network model, a proper modeling framework may be constructed. A method based on an enhanced 2D Displacement Discontinuity Method (2D DDM) may be used for computing the induced stresses on a given fracture and in the rock from the rest of the complex fracture network (see, e.g., Olson, J. E., Predicting Fracture Swarms—The Influence of Sub critical Crack Growth and the Crack-Tip Process Zone on Joints Spacing in Rock. In The Initiation, Propagation and Arrest of Joints and Other Fractures, ed. J. W. Cosgrove and T. Engelder, Geological Soc. Special Publications, London, 231, 73-87 (2004) (hereafter “Olson 2004”), the entire contents of which are hereby incorporated by reference herein). Fracture turning may also be modeled based on the altered local stress direction ahead of the propagating fracture tip due to the stress shadow effect. The simulation results from the UFM model that incorporates the fracture interaction modeling are presented.

UFM Model Description

To simulate the propagation of a complex fracture network that includes of many intersecting fractures, equations governing the underlying physics of the fracturing process may be used. The basic governing equations may include, for example, equations governing fluid flow in the fracture network, the equation governing the fracture deformation, and the fracture propagation/interaction criterion.

Continuity equation assumes that fluid flow propagates along a fracture network with the following mass conservation:

$\begin{array}{cc}\frac{\partial q}{\partial s}+\frac{\partial \left(H_{\mathrm{fl}}\stackrel{\_}{w}\right)}{\partial t}+q_L=0& \left(1\right)\end{array}$ where q is the local flow rate inside the hydraulic fracture along the length, w is an average width or opening at the cross-section of the fracture at position s=s(x,y), H_fl is the height of the fluid in the fracture, and q_L is the leak-off volume rate through the wall of the hydraulic fracture into the matrix per unit height (velocity at which fracturing fluid infiltrates into surrounding permeable medium) which is expressed through Carter's leak-off model. The fracture tips propagate as a sharp front, and the length of the hydraulic fracture at any given time t is defined as l(t).

The properties of driving fluid may be defined by power-law exponent n′ (fluid behavior index) and consistency index K′. The fluid flow could be laminar, turbulent or Darcy flow through a proppant pack, and may be described correspondingly by different laws. For the general case of 1D laminar flow of power-law fluid in any given fracture branch, the Poiseuille law (see, e.g., Nolte, 1991) may be used:

$\begin{array}{cc}\frac{\partial p}{\partial s}=-{\alpha }_0\frac{1}{{\stackrel{\_}{w}}^{2n^{\prime }+1}}\frac{q}{H_{\mathrm{fl}}}{\frac{q}{H_{\mathrm{fl}}}}^{n^{\prime }-1}& \left(2\right)\\ \mathrm{where}& \\ {\alpha }_0=\frac{2K^{\prime }}{{\varphi \left(n^{\prime }\right)}^{n^{\prime }}}·{\left(\frac{4n^{\prime }+2}{n^{\prime }}\right)}^{n^{\prime }};\varphi \left(n^{\prime }\right)=\frac{1}{H_{\mathrm{fl}}}{\int }_{H_{\mathrm{fl}}}{\left(\frac{w\left(z\right)}{\stackrel{\_}{w}}\right)}^{\frac{2n^{\prime }+1}{n^{\prime }}}dz& \left(3\right)\end{array}$ Here w(z) represents fracture width as a function of depth at current position s, α is coefficient, n′ is power law exponent (fluid consistency index), ϕ is shape function, and dz is the integration increment along the height of the fracture in the formula.

Fracture width may be related to fluid pressure through the elasticity equation. The elastic properties of the rock (which may be considered as mostly homogeneous, isotropic, linear elastic material) may be defined by Young's modulus E and Poisson's ratio v. For a vertical fracture in a layered medium with variable minimum horizontal stress σ_h(x, y, z) and fluid pressure p, the width profile (w) can be determined from an analytical solution given as:w(x,y,z)=w(p(x,y),H,z)  (4)where W is the fracture width at a point with spatial coordinates x, y, z (coordinates of the center of fracture element); p(x,y) is the fluid pressure, H is the fracture element height, and z is the vertical coordinate along fracture element at point (x,y).

Because the height of the fractures may vary, the set of governing equations may also include the height growth calculation as described, for example, in Kresse 2011.

In addition to equations described above, the global volume balance condition may be satisfied:

$\begin{array}{cc}{\int }_0^{t}Q\left(t\right)dt={\int }_0^{L\left(t\right)}H\left(s,t\right)\stackrel{\_}{w}\left(s,t\right)ds+{\int }_{H_L}{\int }_0^t{\int }_0^{L\left(t\right)}2g_{L}dsdtd{h}_l& \left(5\right)\end{array}$ where g_L is fluid leakoff velocity, Q(t) is time dependent injection rate, H(s,t) height of the fracture at spacial point s(x,y) and at the time t, ds is length increment for integration along fracture length, d_t is time increment, dh_l is increment of leakoff height, H_L is leakoff height, an s₀ is a spurt loss coefficient. Equation (5) provides that the total volume of fluid pumped during time t is equal to the volume of fluid in the fracture network and the volume leaked from the fracture up to time t. Here L(t) represents the total length of the HFN at the time t and S₀ is the spurt loss coefficient. The boundary conditions may use the flow rate, net pressure and fracture width to be zero at all fracture tips.

The system of Eq. 1-5, together with initial and boundary conditions, may be used to represent a set of governing equations. Combining these equations and discretizing the fracture network into small elements may lead to a nonlinear system of equations in terms of fluid pressure p in each element, simplified as f(p)=0, which may be solved by using a damped Newton-Raphson method.

Fracture interaction may be taken into account to model hydraulic fracture propagation in naturally fractured reservoirs. This includes, for example, the interaction between hydraulic fractures and natural fractures, as well as interaction between hydraulic fractures. For the interaction between hydraulic and natural fractures a semi-analytical crossing criterion may be implemented in the UFM using, for example, the approach described in Gu and Weng 2010, and Gu et al. 2011.

Modeling of Stress Shadow

For parallel fractures, the stress shadow can be represented by the superposition of stresses from neighboring fractures. FIG. 2 is a schematic depiction of a 2D fracture 200 about a coordinate system having an x-axis and a y-axis. Various points along the 2D fractures, such as a first end at h/2, a second end at −h/2 and a midpoint are extended to an observation point (x,y). Each line L extends at angles θ₁, θ₂ from the points along the 2D fracture to the observation point.

The stress field around a 2D fracture with internal pressure p can be calculated using, for example, the techniques as described in Warpinski and Teufel. The stress that affects fracture width is σ_x and can be calculated from:

$\begin{array}{cc}{\sigma }_x=p\left[1-\frac{\stackrel{\_}{L}}{\sqrt{{\stackrel{\_}{L}}_1{\stackrel{\_}{L}}_2}}\cos \left(\theta -\frac{{\theta }_1+{\theta }_2}{2}\right)-\frac{\stackrel{\_}{L}}{{\left({\stackrel{\_}{L}}_1{\stackrel{\_}{L}}_2\right)}^{\frac{3}{2}}}\sin \theta \sin \left(\frac{3}{2}\left({\theta }_1+{\theta }_2\right)\right)\right]& \left(6\right)\\ \mathrm{where}& \\ \theta =\arctan \left(-\frac{\stackrel{\_}{x}}{\stackrel{\_}{y}}\right)& \left(7.1\right)\\ {\theta }_1=\arctan \left(-\frac{\stackrel{\_}{x}}{\stackrel{\_}{1+\stackrel{\_}{y}}}\right)& \left(7.2\right)\\ {\theta }_2=\arctan \left(\frac{\stackrel{\_}{x}}{1-\stackrel{\_}{y}}\right)& \left(7.3\right)\end{array}$ and where σ_x is stress in the x direction, p is internal pressure, and x, y, L, L₁, L₂ are the coordinates and distances in FIG. 2 normalized by the fracture half-height h/2. Since σ_x varies in the y-direction as well as in the x-direction, an averaged stress over the fracture height may be used in the stress shadow calculation.

The analytical equation given above can be used to compute the average effective stress of one fracture on an adjacent parallel fracture and can be included in the effective closure stress on that fracture.

For more complex fracture networks, the fractures may orient in different directions and intersect each other. FIG. 3 shows a complex fracture network 300 depicting stress shadow effects. The fracture network 300 includes hydraulic fractures 303 extending from a wellbore 304 and interacting with other fractures 305 in the fracture network 300.

A more general approach may be used to compute the effective stress on any given fracture branch from the rest of the fracture network. In UFM, the mechanical interactions between fractures may be modeled based on an enhanced 2D Displacement Discontinuity Method (DDM) (Olson 2004) for computing the induced stresses (see, e.g., FIG. 3).

In a 2D, plane-strain, displacement discontinuity solution, (see, e.g., Crouch, S. L. and Starfield, A. M., Boundary Element Methods in Solid Mechanics, George Allen & Unwin Ltd, London. Fisher, M. K. (1983) (hereafter Crouch and Starfield 1983), the entire contents of which are hereby incorporated by reference) may be used to describe the normal and shear stresses (σ_n and σ_s) acting on one fracture element induced by the opening and shearing displacement discontinuities (D_n and D_s) from all fracture elements. To account for the 3D effect due to finite fracture height, Olson 2004 may be used to provide a 3D correction factor to the influence coefficients C^ij in combination with the modified elasticity equations of 2D DDM as follows:

$\begin{array}{cc}{\sigma }_n^i=\sum _{j=1}^N{A}^{\mathrm{ij}}{C}_{ns}^{\mathrm{ij}}{D}_s^j+\sum _{j=1}^N{A}^{\mathrm{ij}}{C}_{\mathrm{nn}}^{\mathrm{ij}}{D}_n^j& \left(8.1\right)\\ {\sigma }_s^i=\sum _{j=1}^N{A}^{\mathrm{ij}}{C}_{\mathrm{ss}}^{\mathrm{ij}}{D}_s^j+\sum _{j=1}^N{A}^{\mathrm{ij}}{C}_{\mathrm{sn}}^{\mathrm{ij}}{D}_n^j& \left(8.2\right)\end{array}$ where A is a matrix of influence coefficients described in eq. (9), N is a total number of elements in the network whose interaction is considered, i is the element considered, and j=1, N are other elements in the network whose influence on the stresses on element i are calculated; and where C^ij are the 2D, plane-strain elastic influence coefficients. These expressions can be found in Crouch and Starfield 1983.

Elem i and j of FIG. 3 schematically depict the variables i and j in equations (8.1, 8.2). Discontinuities D_s and D_n applied to Elem j are also depicted in FIG. 3. Dn may be the same as the fracture width, and the shear stress s may be 0 as depicted. Displacement discontinuity from Elem j creates a stress on Elem i as depicted by σ_s and σ_n.

The 3D correction factor suggested by Olson 2004 may be presented as follows:

$\begin{array}{cc}{A}^{\mathrm{ij}}=1-\frac{d_{\mathrm{ij}}^{\beta }}{{\left[d_{\mathrm{ij}}^2+{\left(h/\alpha \right)}^2\right]}^{\beta /2}}& \left(8\right)\end{array}$ where h is the fracture height, d_ij is the distance between elements i and j, α and β are fitting parameters. Eq. 9 shows that the 3D correction factor may lead to decaying of interaction between any two fracture elements when the distance increases.

In the UFM model, at each time step, the additional induced stresses due to the stress shadow effects may be computed. It may be assumed that at any time, fracture width equals the normal displacement discontinuities (D_n) and shear stress at the fracture surface is zero, i.e., D^j. Substituting these two conditions into Eqs. 8.1 and 8.2, the shear displacement discontinuities (D_s) and normal stress induced on each fracture element (σ_n) may be found.

The effects of the stress shadow induced stresses on the fracture network propagation pattern may be described in two folds. First, during pressure and width iteration, the original in-situ stresses at each fracture element may be modified by adding the additional normal stress due to the stress shadow effect. This may directly affect the fracture pressure and width distribution which may result in a change on the fracture growth. Second, by including the stress shadow induced stresses (normal and shear stresses), the local stress fields ahead of the propagating tips may also be altered which may cause the local principal stress direction to deviate from the original in-situ stress direction. This altered local principal stress direction may result in the fracture turning from its original propagation plane and may further affect the fracture network propagation pattern.

Validation of Stress Shadow Model

Validation of the UFM model for the cases of bi-wing fractures may be performed using, for example, Weng 2011 or Kresse 2011. Validation may also be performed using the stress shadow modeling approach. By way of example, the results may be compared using 2D DDM to Flac 3D as provided in Itasca Consulting Group Inc., 2002, FLAC3D (Fast Lagrangian Analysis of Continua in 3 Dimensions), Version 2.1, Minneapolis: ICG (2002) (hereafter “Itasca, 2002”).

Comparison of Enhanced 2D DDM to Flac3D

The 3D correction factors suggested by Olson 2004 contain two empirical constants, α and β. The values of α and β may be calibrated by comparing stresses obtained from numerical solutions (enhanced 2D DDM) to the analytical solution for a plane-strain fracture with infinite length and finite height. The model may further be validated by comparing the 2D DDM results to a full three dimensional numerical solutions, utilizing, for example, FLAC3D, for two parallel straight fractures with finite lengths and heights.

The validation problem is shown in FIG. 4. FIG. 4 a schematic diagram 400 comparing enhanced 2D DDM to Flac3D for two parallel straight fractures. As shown in FIG. 400, two parallel fractures 407.1, 407.2 are subject to stresses σ_x, σ_y along an x, y coordinate axis. The fractures have length 2L_xf, and pressure of the fracture p₁, p₂, respectively. The fractures are a distance s apart.

The fracture in Flac3D may be simulated as two surfaces at the same location but with un-attached grid points. Constant internal fluid pressure may be applied as the normal stress on the grids. Fractures may also be subject to remote stresses, σ_x and σ_y. Two fractures may have the same length and height with the ratio of height/half-length=0.3.

Stresses along x-axis (y=0) and y-axis (x=0) may be compared. Two closely spaced fractures (s/h=0.5) may be simulated as shown in the comparison of FIGS. 5.1-5.3. These figures provide a comparison of extended 2D DDM to Flac3D: Stresses along x-axis (y=0) and y-axis (x=0).

These figures include graphs 500.1, 500.2, 500.3, respectively, illustrating 2D DDM and Flac3D of extended fractures for σy along the y-axis, σx along the y-axis, and σy along the x-axis, respectively. FIG. 5.1 plots σy/p (y-axis) versus normalized distance from fracture (x-axis) using 2D DDM and Flac3D. FIG. 5.2 plots σx/p (y-axis) versus normalized distance from fracture (x-axis) using 2D DDM and Flac3D. FIG. 5.3 plots σy/p (y-axis) versus normalized distance from fracture (x-axis) using 2D DDM and Flac3D. The location L_f of the fracture tip is depicted along line x/h.

As shown in FIGS. 5.1-5.3, the stresses simulated from enhanced 2D DDM approach with 3D correction factor match pretty well to those from the full 3D simulator results, which indicates that the correction factor allows capture the 3D effect from the fracture height on the stress field.

Comparison to CSIRO model

The UFM model that incorporates the enhanced 2DDM approach may be validated against full 2D DDM simulator by CSIRO (see, e.g., Zhang, X., Jeffrey, R. G., and Thiercelin, M. 2007, Deflection and Propagation of Fluid-Driven Fractures at Frictional Bedding Interfaces: A Numerical Investigation. Journal of Structural Geology, 29: 396-410, (hereafter “Zhang 2007”) the entire contents of which is hereby incorporated by reference in its entirety). This approach may be used, for example, in the limiting case of very large fracture height where 2D DDM approaches do not consider 3D effects of the fractures height.

The comparison of influence of two closely propagating fractures on each other's propagation paths may be employed. The propagation of two hydraulic fractures initiated parallel to each other (propagating along local max stress direction) may be simulated for configurations, such as: 1) initiation points on top of each other and offset from each other for isotropic, and 2) anisotropic far field stresses. The fracture propagation path and pressure inside of each fracture may be compared for UFM and CSIRO code for the input data given in Table 1.

TABLE 1
Input data for validation against CSIRO model
Injection rate	0.106	m3/s	40	bbl/min
Stress anisotropy	0.9	MPa	130	psi
Young's modulus	3 × 10{circumflex over ( )}10	Pa	4.35e+6	psi
Poisson's ratio	0.35	0.35
Fluid viscosity	0.001	pa-s	1	cp
Fluid Specific Gravity	1.0	1.0
Min horizontal stress	46.7	MPa	6773	psi
Max horizontal stress	47.6	MPa	6903	psi
Fracture toughness	1	MPa-m0.5	1000	psi/in0.5
Fracture height	120	m	394	ft

When two fractures are initiated parallel to each other with initiation points separated by dx=0, dy=33 ft (10.1 m) (max horizontal stress field is oriented in x-direction), they may turn away from each other due to the stress shadow effect.

The propagation paths for isotropic and anisotropic stress fields are shown in FIGS. 6.1 and 6.2. These figures are graphs 600.1, 600.2 depicting propagation paths for two initially parallel fractures 609.1, 609.2 in isotropic and anisotropic stress fields, respectively. The fractures 609.1 and 609.2 are initially parallel near the injection points 615.1, 615.2, but diverge as they extend away therefrom. Comparing with isotropic case, the curvatures of the fractures in the case of stress anisotropy are depicted as being smaller. This may be due to the competition between the stress shadow effect which tends to turn fractures away from each other, and far-field stresses which pushes fractures to propagate in the direction of maximum horizontal stress (x-direction). The influence of far-field stress becomes dominant as the distance between the fractures increases, in which case the fractures may tend to propagate parallel to maximum horizontal stress direction.

FIGS. 7.1 and 7.2 depict graphs 700.1, 7002 showing a pair of fractures initiated from two different injection points 711.1, 711.2, respectively. These figures show a comparison for the case when fractures are initiated from points separated by a distance dx=dy=(10.1 m) for an isotropic and anisotropic stress field, respectively. In these figures, the fractures 709.1, 709.2 tend to propagate towards each other. Examples of similar type of behavior have been observed in lab experiments (see, e.g., Zhang 2007).

As indicated above, the enhanced 2D DDM approach implemented in UFM model may be able to capture the 3D effects of finite fracture height on fracture interaction and propagation pattern, while being computationally efficient. A good estimation of the stress field for a network of vertical hydraulic fractures and fracture propagation direction (pattern) may be provided.

EXAMPLE CASES

Case #1—Parallel Fractures in Horizontal Wells

FIG. 8 is a schematic plot 800 of parallel transverse fractures 811.1, 811.2, 811.3 propagating simultaneously from multiple perforation clusters 815.1, 815.2, 815.3, respectively, about a horizontal wellbore 804. Each of the fractures 811.1, 811.2, 811.3 provides a different flow rate q₁, q₂, q₃ that is part of the total flow q_t at a pressure p₀.

When the formation condition and the perforations are the same for all the fractures, the fractures may have about the same dimensions if the friction pressure in the wellbore between the perforation clusters is proportionally small. This may be assumed where the fractures are separated far enough and the stress shadow effects are negligible. When the spacing between the fractures is within the region of stress shadow influence, the fractures may be affected not only in width, but also in other fracture dimension. To illustrate this, a simple example of five parallel fractures may be considered.

In this example, the fractures are assumed to have a constant height of 100 ft (30.5 m). The spacing between the fractures is 65 ft (19.8 m). Other input parameters are given in Table 2 below:

TABLE 2
Input parameters for Case #1
Young's modulus       6.6 × 106 psi = 4.55e+10 Pa
Poisson's ratio       0.35
Rate                  12.2 bbl/min = 0.032 m3/s
Viscosity             300 cp = 0.3 Pa-s
Height                100 ft = 30.5 m
Leakoff coefficient   3.9 × 10−2 m/s1/2
Stress anisotropy     200 psi = 1.4 Mpa
Fracture spacing      65 ft = 19.8 m
No. of perfs per frac 100

For this simple case, a conventional Perkins-Kern-Nordgren (PKN) model (see, e.g., Mack, M. G. and Warpinski, N. R., Mechanics of Hydraulic Fracturing. Chapter 6, Reservoir Stimulation, 3rd Ed., eds. Economides, M. J. and Nolte, K. G. John Wiley & Sons (2000)) for multiple fractures may be modified by incorporating the stress shadow calculation as given from Eq. 6. The increase in closure stress may be approximated by averaging the computed stress from Eq. 6 over the entire fracture. Note that this simplistic PKN model may not simulate the fracture turning due to the stress shadow effect. The results from this simple model may be compared to the results from the UFM model that incorporates point-by-point stress shadow calculation along the entire fracture paths as well as fracture turning.

FIG. 9 shows the simulation results of fracture lengths of the five fractures, computed from both models. FIG. 9 is a graph 900 depicting length (y-axis) versus time (t) of five parallel fractures during injection. Lines 917.1-917.5 are generated from the UFM model. Lines 919.1-919.5 are generated from the simplistic PKN model.

The fracture geometry and width contour from the UFM model for the five fractures of FIG. 9 are shown in FIG. 10. FIG. 10 is a schematic diagram 1000 depicting fractures 1021.1-1021.5 about a wellbore 1004.

Fracture 1021.3 is the middle one of the five fractures, and fractures 1021.1 and 1021.5 are the outmost ones. Since fractures 1021.2, 1021.3, and 1021.4 have smaller width than that of the outer ones due to the stress shadow effect, they may have larger flow resistance, receive less flow rate, and have shorter length. Therefore, the stress shadow effects may not only be fracture width but also fracture length under dynamic conditions.

The effect of stress shadow on fracture geometry may be influenced by many parameters. To illustrate the effect of some of these parameters, the computed fracture lengths for the cases with varying fracture spacing, perforation friction, and stress anisotropy are shown in Table 3 below.

FIGS. 11.1 and 11.2 shows the fracture geometry predicted by the UFM for the case of large perforation friction and the case of large fracture spacing (e.g., about 120 ft (36.6 m)). FIGS. 11.1 and 11.2 are schematic diagrams 1100.1 and 1100.2 depicting five fractures 1123.1-1123.5 about a wellbore 1104. When the perforation friction is large, a large diversion force that uniformly distributes the flow rate into all perforation clusters may be provided. Consequently, the stress shadow may be overcome and the resulting fracture lengths may become approximately equal as shown in FIG. 11.1. When fracture spacing is large, the effect of the stress shadow may dissipate, and fractures may have approximately the same dimensions as shown in FIG. 11.2.

TABLE 3
Influence of various parameters on fracture geometry
	Base	120 ft spacing	No. of	Anisotropy = 50 psi
Frac	case	(36.6 m)	perfs = 2	(345000 Pa)
1	133	113	105	111
2	93	104	104	95
3	83	96	104	99
4	93	104	100	95
5	123	113	109	102

Case #2—Complex Fractures

In an example of FIG. 12, the UFM model may be used to simulate a 4-stage hydraulic fracture treatment in a horizontal well in a shale formation. See, e.g., Cipolla, C., Weng, X., Mack, M., Ganguly, U., Kresse, O., Gu, H., Cohen, C. and Wu, R., Integrating Microseismic Mapping and Complex Fracture Modeling to Characterize Fracture Complexity. Paper SPE 140185 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Tex., USA, Jan. 24-26, 2011, (hereinafter “Cipolla 2011”) the entire contents of which are hereby incorporated by reference in their entirety. The well may be cased and cemented, and each stage pumped through three or four perforation clusters. Each of the four stages may include of approximately 25,000 bbls (4000 m³) of fluid and 440,000 lbs (2e+6 kg) of proppant. Extensive data may be available on the well, including advanced sonic logs that provide an estimate of minimum and maximum horizontal stress. Microseismic mapping data may be available for all stages. See, e.g., Daniels, J., Waters, G., LeCalvez, J., Lassek, J., and Bentley, D., Contacting More of the Barnett Shale Through an Integration of Real-Time Microseismic Monitoring, Petrophysics, and Hydraulic Fracture Design. Paper SPE 110562 presented at the 2007 SPE Annual Technical Conference and Exhibition, Anaheim, Calif., USA, Oct. 12-14, 2007. This example is shown in FIG. 12. FIG. 12 is a graph depicting microseismic mapping of microseismic events 1223 at various stages about a wellbore 1204.

The stress anisotropy from the advanced sonic log, indicates a higher stress anisotropy in the toe section of the well compared to the heel. An advanced 3D seismic interpretation may indicate that the dominant natural fracture trend changes from NE-SW in the toe section to NW-SE in heel portion of the lateral. See, e.g., Rich, J. P. and Ammerman, M., Unconventional Geophysics for Unconventional Plays. Paper SPE 131779 presented at the Unconventional Gas Conference, Pittsburgh, Pa., USA, Feb. 23-25, 2010, the entire contents of which is hereby incorporated by reference herein in its entirety.

Simulation results may be based on the UFM model without incorporating the full stress shadow calculation (see, e.g., Cipolla 2011), including shear stress and fracture turning (see, e.g., Weng 2011). The simulation may be updated with the full stress model as provided herein. FIGS. 13.1-13.4 show a plan view of a simulated fracture network 1306 about a wellbore 1304 for all four stages, respectively, and their comparison to the microseismic measurements 1323.1-1323.4, respectively.

From simulation results in FIGS. 13.1-13.4, it can be seen that for Stages 1 and 2, the closely spaced fractures did not diverge significantly. This may be because of the high stress anisotropy in the toe section of the wellbore. For Stage 3 and 4, where stress anisotropy is lower, more fracture divergence can be seen as a result of the stress shadow effect.

Case #3—Multi-Stage Example

Case #3 is an example showing how stress shadow from previous stages can influence the propagation pattern of hydraulic fracture networks for next treatment stages, resulting in changing of total picture of generated hydraulic fracture network for the four stage treatment case.

This case includes four hydraulic fracture treatment stages. The well is cased and cemented. Stages 1 and 2 are pumped through three perforated clusters, and Stages 3 and 4 are pumped through four perforated clusters. The rock fabric is isotropic. The input parameters are listed in Table 4 below. The top view of total hydraulic fracture network without and with accounting for stress shadow from previous stages is shown in FIGS. 13.1-13.4.

TABLE 4
Input parameters for Case #3
Young's modulus 4.5 × 106 psi = 3.1e+10 Pa
Poisson's ratio 0.35
Rate            30.9 bpm = 0.082 m3/s
Viscosity       0.5 cp = 0.0005 pa-s
Height          330 ft = 101 m
Pumping time    70 min

FIGS. 14.1-14.4 are schematic diagrams 1400.1-1400-4 depicting a fracture network 1429 at various stages during a fracture operation. FIG. 14.1 shows a discrete fracture network (DFN) 1429 before treatment. FIG. 14.2 depicts a simulated DFN 1429 after a first treatment stage. The DFN 1429 has propagated hydraulic fractures (HFN) 1431 extending therefrom due to the first treatment stage. FIG. 14.3 shows the DFN depicting a simulated HFN 1431.1-1431.4 propagated during four stages, respectively, but without accounting for previous stage effects. FIG. 14.4 shows the DFN depicting HFN 1431.1, 1431.2′-1431.4′ propagated during four stages, but with accounting for the fractures, stress shadows and HFN from previous stages.

When stages are generated separately, they may not see each other as indicated in FIG. 14.3. When stress shadow and HFN from previous stages are taken into account as in FIG. 14.4 the propagation pattern may change. The hydraulic fractures 1431.1 generated for the first stage is the same for both case scenarios as shown in FIGS. 14.3 and 14.4. The second stage 1431.2 propagation pattern may be influenced by the first stage through stress shadow, as well as through new DFN (including HFN 1431.1 from Stage 1), resulting in the changing of propagation patterns to HFN 1431.2′. The HFN 1431.1′ may start to follow HFN 1431.1 created at stage 1 while intercounting it. The third stage 1431.3 may follow a hydraulic fracture created during second stage treatment 1431.2, 1431.2′, and may not propagate too far due to stress shadow effect from Stage 2 as indicated by 1431.3 versus 1431.3′. Stage 4 (1431.4) may tend to turn away from stage three when it could, but may follow HFN 1431.3′ from previous stages when encounters it and be depicted as HFN 1431.4′ in FIG. 14.4.

A method for computing the stress shadow in a complex hydraulic fracture network is presented. The method may involve an enhanced 2D or 3D Displacement Discontinuity Method with correction for finite fracture height. The method may be used to approximate the interaction between different fracture branches in a complex fracture network for the fundamentally 3D fracture problem. This stress shadow calculation may be incorporated in the UFM, a complex fracture network model. The results for simple cases of two fractures show the fractures can either attract or expel each other depending on their initial relative positions, and compare favorably with an independent 2D non-planar hydraulic fracture model.

Simulations of multiple parallel fractures from a horizontal well may be used to confirm the behavior of the two outmost fractures that may be more dominant, while the inner fractures have reduced fracture length and width due to the stress shadow effect. This behavior may also depend on other parameters, such as perforation friction and fracture spacing. When fracture spacing is greater than fracture height, the stress shadow effect may diminish and there may be insignificant differences among the multiple fractures. When perforation friction is large, sufficient diversion to distribute the flow equally among the perforation clusters may be provided, and the fracture dimensions may become approximately equal despite the stress shadow effect.

When complex fractures are created, if the formation has a small stress anisotropy, fracture interaction can lead to dramatic divergence of the fractures where they tend to repel each other. On the other hand, for large stress anisotropy, there may be limited fracture divergence where the stress anisotropy offsets the effect of fracture turning due to the stress shadow, and the fracture may be forced to go in the direction of maximum stress. Regardless of the amount of fracture divergence, the stress shadowing may have an effect on fracture width, which may affect the injection rate distribution into multiple perforation clusters, and overall fracture network footprint and proppant placement.

FIG. 15 is a flow chart depicting a method 1500 of performing a fracture operation at a wellsite, such as the wellsite 100 of FIG. 1.1. The wellsite is positioned about a subterranean formation having a wellbore therethrough and a fracture network therein. The fracture network has natural fractures as shown in FIGS. 1.1 and 1.2. The method (1500) may involve (1580) performing a stimulation operation by stimulating the wellsite by injection of an injection fluid with proppant into the fracture network to form a hydraulic fracture network. In some cases, the stimulation may be performed at the wellsite or by simulation.

The method involves (1582) obtaining wellsite data and a mechanical earth model of the subterranean formation. The wellsite data may include any data about the wellsite that may be useful to the simulation, such as natural fracture parameters of the natural fractures, images of the fracture network, etc. The natural fracture parameters may include, for example, density orientation, distribution, and mechanical properties (e.g., coefficients of friction, cohesion, fracture toughness, etc.) The fracture parameters may be obtained from direct observations of borehole imaging logs, estimated from 3D seismic, ant tracking, sonic wave anisotropy, geological layer curvature, microseismic events or images, etc. Examples of techniques for obtaining fracture parameters are provided in PCT/US2012/059774, the entire contents of which are hereby incorporated by reference herein in their entirety.

Images may be obtained by, for example, observing borehole imaging logs, estimating fracture dimensions from wellbore measurements, obtaining microseismic images, and/or the like. The fracture dimensions may be estimated by evaluating seismic measurements, ant tracking, sonic measurements, geological measurements, and/or the like. Other wellsite data may also be generated from various sources, such as wellsite measurements, historical data, assumptions, etc. Such data may involve, for example, completion, geological structure, petrophysical, geomechanical, log measurement and other forms of data. The mechanical earth model may be obtained using conventional techniques.

The method (1500) also involves (1584) generating a hydraulic fracture growth pattern over time, such as during the stimulation operation. FIGS. 16.1-16.4 depict an example of (1584) generating a hydraulic fracture growth pattern. As shown in FIG. 16.1, in its initial state, a fracture network 1606.1 with natural fractures 1623 is positioned about a subterranean formation 1602 with a wellbore 1604 therethrough. As proppant is injected into the subterranean formation 1602 from the wellbore 1604, pressure from the proppant creates hydraulic fractures 1691 about the wellbore 1604. The hydraulic fractures 1691 extend into the subterranean formation along L₁ and L₂ (FIG. 16.2), and encounter other fractures in the fracture network 1606.1 over time as indicated in FIGS. 16.2-16.3. The points of contact with the other fractures are intersections 1625.

The generating (1584) may involve (1586) extending hydraulic fractures from the wellbore and into the fracture network of the subterranean formation to form a hydraulic fracture network including the natural fractures and the hydraulic fractures as shown in FIG. 16.2. The fracture growth pattern is based on the natural fracture parameters and a minimum stress and a maximum stress on the subterranean formation. The generating may also involve (1588) determining hydraulic fracture parameters (e.g., pressure p, width w, flow rate q, etc.) of the hydraulic fractures, (1590) determining transport parameters for the proppant passing through the hydraulic fracture network, and (1592) determining fracture dimensions (e.g., height) of the hydraulic fractures from, for example, the determined hydraulic fracture parameters, the determined transport parameters and the mechanical earth model. The hydraulic fracture parameters may be determined after the extending. The determining (1592) may also be performed by from the proppant transport parameters, wellsite parameters and other items.

The generating (1584) may involve modeling rock properties based on a mechanical earth model as described, for example, in Koutsabeloulis and Zhang, 3D Reservoir Geomechanics Modeling in Oil/Gas Field Production, SPE Paper 126095, 2009 SPE Saudi Arabia Section Technical Symposium and Exhibition held in Al Khobar, Saudi Arabia, 9-11 May, 2009 (“Koutsabeloulis 2009”), the entire contents of which are hereby incorporated by reference herein. The generating may also involve modeling the fracture operation by using the wellsite data, fracture parameters and/or images as inputs modeling software, such as UFM™ and PETREL™ commercially available from SCHLUMBERGER TECHNOLOGY CORPORATION™ (see: www.slb.com), to generate successive images of induced hydraulic fractures in the fracture network.

The method (1500) also involves (1594) performing stress shadowing on the hydraulic fractures to determine stress interference between the hydraulic fractures (or with other fractures), and (1598) repeating the generating (1584) based on the stress shadowing and/or the determined stress interference between the hydraulic fractures. The repeating may be performed to account for fracture interference that may affect fracture growth. Stress shadowing may involve performing, for example, a 2D or 3D DDM for each of the hydraulic fractures and updating the fracture growth pattern over time. The fracture growth pattern may propagate normal to a local principal stress direction according to stress shadowing. The fracture growth pattern may involve influences of the natural and hydraulic fractures over the fracture network (see FIG. 16.3).

Stress shadowing may be performed for multiple wellbores of the wellsite. The stress shadowing from the various wellbores may be combined to determine the interaction of fractures as determined from each of the wellbores. The generating may be repeated for each of the stress shadowings performed for one or more of the multiple wellbores. The generating may also be repeated for stress shadowing performed where stimulation is provided from multiple wellbores. Multiple simulations may also be performed on the same wellbore with various combinations of data, and compared as desired. Historical or other data may also be input into the generating to provide multiple sources of information for consideration in the ultimate results.

The method also involves (1596) determining crossing behavior between the hydraulic fractures and an encountered fracture if the hydraulic fracture encounters another fracture, and (1598) repeating the generating (1584) based on the crossing behavior if the hydraulic fracture encounters a fracture (see, e.g., FIG. 16.3). Crossing behavior may be determined using, for example, the techniques of PCT/US2012/059774, the entire contents of which is hereby incorporated herein in its entirety.

The determining crossing behavior may involve performing stress shadowing. Depending on downhole conditions, the fracture growth pattern may be unaltered or altered when the hydraulic fracture encounters the fracture. When a fracture pressure is greater than a stress acting on the encountered fracture, the fracture growth pattern may propagate along the encountered fracture. The fracture growth pattern may continue propagation along the encountered fracture until the end of the natural fracture is reached. The fracture growth pattern may change direction at the end of the natural fracture, with the fracture growth pattern extending in a direction normal to a minimum stress at the end of the natural fracture as shown in FIG. 16.4. As shown in FIG. 16.4, the hydraulic fracture extends on a new path 1627 according to the local stresses σ₁ and σ₂.

Optionally, the method (1500) may also involve (1599) validating the fracture growth pattern. The validation may be performed by comparing the resulting growth pattern with other data, such as microseismic images as shown, for example, in FIGS. 7.1 and 7.2.

The method may be performed in any order and repeated as desired. For example, the generating (1584)-(1599) may be repeated over time, for example, by iteration as the fracture network changes. The generating (1584) may be performed to update the iterated simulation performed during the generating to account for the interaction and effects of multiple fractures as the fracture network is stimulated over time.

II. Interpretation of Microseismicity

In an aspect of the present disclosure, at least one embodiment relates to techniques for performing oilfield operations, such as fracture and/or stimulation operations. More particularly, at least one embodiment of the present disclosure relates to a method for microseismic data interpretation using a geomechanics model to compute the stress field surrounding the created fracture network and potential shear failure in the natural fractures. This may lead to a means for calibrating and a more accurate determination of the fracture network geometry.

The disclosure also relates to interpretation of hydraulic fracturing based on microseismicity and stress analysis. A method is provided to consider microseismicity triggered as a result of interaction between hydraulic and natural fractures. Geomechanic models may be used to determine stress fields surrounding a fracture network and potential shear failure in natural fractures of the fracture network. Hydraulic fracture geometry may be determined based on the geomechanic models.

Hydraulic fracture interpretation may be performed using the 2D and 3D DDM methods to describe induced stress on a given fracture by other fractures as described above. Hydraulic fracture interpretation may also be performed using 2D DDM and 3D DDM stress field methods to compute the stress field for a collection of fractures with known interfacial displacements. In the stress field methods, a microseismicity prediction employs the DDM to compute the stress in the rock and/or on closed natural fractures located away from the hydraulic fractures. DDM may be used to generate induced stresses on a fracture by other fractures using 2D, 3D DDM and/or to generate stresses on distant fractures using extended DDM.

Current hydraulic fracture monitoring methods and systems may map where the fractures occur and the extent of the fractures. Some methods and systems of microseismic monitoring may process seismic event locations by mapping seismic arrival times and polarization information into three-dimensional (3D) space through the use of modeled travel times and/or ray paths. These methods and systems can be used to infer hydraulic fracture propagation over time.

Understanding the nature and degree of hydraulic fracture complexity may be useful to the economic development of unconventional resources. Examples of hydraulic fracture techniques are described in the following papers: Mayerhofer et al., Integrating of Microseismic Fracture Mapping Results with Numerical Fracture Network Production Modeling in the Barnett Shale, Society of Petroleum Engineers (SPE) 102103, presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Tex., 24-24 Sep. 2006; Mayerhofer et al., What is Stimulated Reservoir Volume (SRV)?, SPE 119890 presented at the SPE Shale Gas Production Conference, Fort Worth, Tex., 16-18 Nov. 2008; Warpinski et al., Stimulating Unconventional Reservoirs: Maximizing Network Growth while Optimizing Fracture Conductivity, SPE 114173 presented at the SPE Unconventional Reservoirs Conference, Keystone, Colo., 10-12 Feb. 2008; and Cipolla et al., The Relationship between Fracture Complexity, Reservoir Properties, and Fracture Treatment Design, SPE 115769 presented at the SPE Annual Technical Conference and Exhibition, Denver, Colo., 21-24 Sep. 2008, the entire contents of which are hereby incorporated by reference herein.

Complex hydraulic fracture propagation may be interpreted from microseismic measurements, for example, from unconventional reservoirs and tight gas reservoirs. Examples of complex hydraulic fracture techniques are provided in the following articles: Maxwell et al., Microseismic Imaging of Hydraulic Fracture Complexity in the Barnett Shale, SPE 77440 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Tex., Sep. 29-Oct. 2, 2002; Fisher et al., Integrating Fracture Mapping Technologies to Optimize Stimulations in the Barnett Shale, 77411 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Tex., Sep. 29-Oct. 2, 2002; Cipolla et al., Effect of Well Placement on Production and Frac Design in a Mature Tight Gas Field, 95337 presented at the SPE Annual Technical Conference and Exhibition, Dallas, Tex., 9-12 Oct. 2005; and Warpinski et al., Stimulating Unconventional Reservoirs: Maximizing Network Growth while Optimizing Fracture Conductivity, SPE 114173 presented at the SPE Unconventional Reservoirs Conference, Keystone, Colo., 10-12 Feb. 2008, the entire contents of which are hereby incorporated by reference herein.

Additional techniques relating to fracturing are provided in Zhao, X. P. and Young, R. P. 2009, Numerical Simulation of Seismicity Induced by Hydraulic Fracturing in Naturally Fractured Reservoirs, Paper SPE 124690 presented at the Annual Technical Conference and Exhibition, New Orleans, La., USA, October 4-7; Meyer, B. R. and Bazan, L. W. (2011) “A Discrete Fracture Network Model for Hydraulically-Induced Fractures: Theory, Parametric and Case Studies,” Paper SPE 140514 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Tex., January 24-26; Jeffery, R. G., Zhang, X., and Thiercelin, M. 2009, Hydraulic Fracture Offsetting in Naturally Fractured Reservoirs: Quantifying a Long-Recognized Process, Paper SPE 119351 presented at 2009 SPE Hydraulic Fracturing Technology Conference, Woodlands, Tex., 19-21 January; and Wu, R., Kresse, O., Weng, X., Cohen, C., and Gu, H. 2012, Modeling of Interaction of Hydraulic Fractures in Complex Fracture Networks, Paper SPE 152052 presented at the SPE Hydraulic Fracturing Technology Conference and Exhibition held in The Woodlands, Tex., USA, 6-8 February (“Wu 2012”), the entire contents of which are hereby incorporated by reference herein.

FIGS. 17-19.2 depict stresses applied to hydraulic fractures and natural fractures of a rock medium, such as the formation around a wellbore as shown, for example, in FIGS. 1.1 and 1.2. As demonstrated by these figures, microseismic events may be triggered by interaction between fracture geometry and stress properties relating to the fractures. Microseismic events recorded during hydraulic fracturing operations may be utilized to interpret induced fracture geometry. Each microseismic event may be a result of a sudden release of local elastic energy stored in the rock associated with crack propagation, for example, under shear deformation.

Examples of microseismic event techniques are provided in Warpinski, N. R., Branagan, P. T., Peterson, R. E., Wolhart, S. L., and Uhl, J. E. 1998, Mapping Hydraulic Fracture Growth and Geometry Using Microseismic Events Detected by a Wireline Retrievable Accelerometer Array, Paper SPE 40014 presented at the 1998 Gas Technology Symposium, Calgary, Alberta, Canada, March 15-18; Cipolla, C. L., Peterman, F., Creegan, T., McCarley, D. and Nevels, H. 2005, Effect of Well Placement on Production and Frac Design in a Mature Tight Gas Field, Paper SPE 95337 presented at the 2005 SPE Annual Technical Conference and Exhibition, Dallas, Tex., October 9-12; Maxwell, S. C., Urbancic, T. I., Steinsberger, N. P., and Zinno, R. 2002, Microseismic Imaging of Hydraulic Fracture Complexity in the Barnett Shale, Paper SPE 77440 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Tex., September 29-October 2; and Fisher, M. K., Davidson, B. M., Goodwin, A. K., Fielder, E. O., Buckler, W. S., and Steinberger, N. P. 2002, Integrating Fracture Mapping Technologies to Optimize Stimulations in the Barnett Shale, Paper SPE 77411 presented at the 2002 SPE Annual Technical Conference and Exhibition, San Antonio, Tex., USA, September 29-October 2, the entire contents of which are hereby incorporated by reference herein.

FIG. 17 is a schematic diagram 1700 depicting a simple planar hydraulic fracture 1701 propagating in a rock medium 1704 containing pre-existing natural fractures 1702. The depicted hydraulic fracture 1701 may be a fracture generated, for example, in the formation 102 of FIG. 1.1. The area 1706 surrounding the hydraulic fracture 1701 indicates fluid infiltration into rock matrix of the rock medium 1704.

The homogeneous rock matrix of the rock medium 1704 may be initially subjected to in-situ stresses (e.g., minimum horizontal stress σ_min, maximum horizontal stress σ_max) in the earth. The faces of the natural fractures 1702 may be in contact with each other since the rock medium 1704 is subjected to compressive in-situ stresses σ_min, σ_max as indicated by the arrows. If the natural fractures 1702 are not aligned with directions of the principal stresses σ_min, σ_max the faces of the natural fractures 1702 may be subjected to shear forces, in addition to compressional normal forces. If the shear stress at the interface exceeds a limiting value, which may be defined as the sum of the cohesion and the normal stress multiplied by a Coulomb friction coefficient (COF), a rock interface may slip, triggering propagation of the fracture and a microseismic event that may be detected from a geophone (not shown) at some distance.

Shear failure may be interpreted based on failure parameters, such as a failure envelope (e.g., a Mohr-Coulomb failure envelope) and a stress state (e.g., a Mohr circle). FIG. 18 is a graph 1800 depicting a Mohr-Coulomb failure envelope 1808 and a Mohr circle 1810. The Mohr-Coulomb failure envelope 1808 may be applicable for a natural fracture interface for the rock medium 1704 of FIG. 17. This failure envelope 1808 may be used as a model describing a response of the rock medium to shear stresses. The Mohr-Coulomb failure envelope 1808 is a plot of shear strength of the rock medium (y-axis) versus the applied normal stress (x-axis). The y-axis denotes σ_shear.

The horizontal axis (x-axis) of the graph 1800 depicts effective stress, defined as the total stress σ_total in the rock minus the pore pressure P_p. The failure envelope 1808 extends from a point alone the negative x-axis to σ_normal on the positive x-axis a distance thereabove. A tensile line 1812 of the failure envelope 1808 extending from the x-axis to the y-axis provides tensile failure of the rock medium. A shear line 1814 extending from the y-axis along a top side of the failure envelope 1808 may indicate shear failure. A compaction line 1816 extending from the shear failure to the x-axis may indicate compaction.

The Mohr circle 1810 of a natural fracture may be used to indicate an initial stress state in the rock medium 1704. The Mohr circle 1810 extends between σ′_min and σ′_max a distance above the x-axis. The Mohr circle 1810 represents normal and shear stresses on a rock face at any orientation θ. The Mohr circle 1810 may be used to determine graphically a stress component acting on a rotated coordinate system. In other words, the Mohr circle 1810 may be used to determine the stress components acting on a differently oriented plane passing through a certain material point. As the pore pressure increases, the Mohr circle 1810 may shift to the left, and may cause the natural fracture 1701 to slip even when the total stress remains constant.

The failure envelope 1808 may be different from the failure envelope for the rock matrix of the rock medium 1704 which may have a different cohesion 1811 (cohesion is the intersection of shear failure slope with the y-axis) and different slope. If the initial stress state in the rock medium 1704 is such that the corresponding Mohr circle 1810 touches the shear failure envelope 1808, a natural fracture oriented at the angle θ that corresponds to the point touching the failure envelop may fail under shear. While a Mohr-Coulomb failure envelope and a Mohr circle are depicted, other failure envelopes or stress states may be used for failure analysis.

Referring to FIGS. 17 and 18, during a hydraulic fracturing treatment (e.g., as shown in FIG. 1.1), fluid can invade into the rock matrix surrounding the hydraulic fracture 1701. As a result, the pore pressure in the rock matrix may increase, and cause the Mohr circle 1810 to shift to the left as explained above. This shifting may be a primary mechanism of microseismicity during hydraulic fracturing in a permeable rock. Another mechanism which may be a dominant mechanism for ultra-low permeability rocks may be stress disturbance surrounding the hydraulic fracture 1701 as schematically depicted in FIG. 19.

FIGS. 19.1 and 19.2 schematically illustrate stress disturbance 1900 of stresses σ_min, σ_vertical applied to the hydraulic fracture 1701. These stress disturbances may trigger an existing natural fracture 1702 to slide if its properties and the initial stress state are such that the natural fracture 1702 is close to a shear failure condition. A small disturbance of the stress, as that induced in the rock surrounding the hydraulic fracture 1701, can push the Mohr circle 1810 to reach the shear failure and creates a microseismic event.

As shown by the cross-sectional view of FIG. 19.1, a stress disturbed region 1918 proportional to fracture height of the fracture 1701 may be generated. Shear deformation 1920 may be generated about the stress distributed region as indicated by the double arrows. As shown by the map view of FIG. 19.1, tensile deformation T may be applied to the hydraulic fracture as indicated by the opposing arrows.

Similar to the natural fractures 1702, if the stress state is such that the shear envelope 1808 of the rock matrix is reached, a shear crack may be created in the rock matrix, which may also trigger a microseismic event. It may be easier to reach the failure condition for at least some of the existing natural fractures 1702 than the rock matrix.

Hydraulic fracturing may be used for hydrocarbon recovery, for example, in ultra-tight unconventional reservoirs, such as shale gas. As in conventional reservoirs, microseismic monitoring may be used to help determine created fracture geometry. Microseismic monitoring may show widespread events cloud, which may indicate complex fracture patterns, or networks, are created during the hydraulic fracturing. When a complex fracture pattern is created, the ability to use a microseismic cloud to delineate the detailed fracture network's structure may be difficult, for example, due to the fact that the microseismic events may not be located on the hydraulic fracture planes and/or may be at natural fractures surrounding the hydraulic fractures, and/or due to uncertainty associated with microseismic event locations.

Examples of microseismic location uncertainty are provided in Maxwell, S. C. 2009, Microseismic Location Uncertainty, CSEG RECORDER, April 2009, pp. 41-46; and Maxwell, S. C., Underhill, B., Bennett, L., Woerpel, C. and Martinez, A. 2010, Key Criteria for a Successful Microseismic Project, Paper SPE 134695 presented at the SPE Annual Technical Conference and Exhibition, Florence, Italy, 19-22 Sep. 2010, the entire contents of which are hereby incorporated by reference herein.

FIG. 20 is a schematic diagram 2000 illustrating how microseismicity may be triggered as a result of interaction between hydraulic fracture 2001 and a natural fracture 2002. A timeline 2022 is provided depicting microseismic events 2028 occurring along the hydraulic fracture 2001 and the natural fracture 2002. Examples of microseismicity are provided in Maxwell, S. C. and Cipolla, C. 2011, What Does Microseismicity Tell Us About Hydraulic Fracturing. Paper SPE 146932 presented at the 2011 SPE Annual Technical Conference and Exhibition, Denver, Colo., October 30-November 2, the entire contents of which are hereby incorporated by reference herein.

At time t1, the hydraulic fracture 2001 is far enough away from the natural fracture 2002 such that stress disturbances surrounding the hydraulic fracture 2001 is insufficient to trigger slippage of interfaces of the natural fracture 2002. In this case, no microseismicity may be emitted from the natural fracture. At time t2, the hydraulic fracture 2001 is sufficiently close to the natural fracture 2002 such that the stress disturbance causes shear slippage to occur at the natural fracture 2002, leading to a microseismic event 2028.

At time t3 the hydraulic fracture 2001 intersects the natural fracture 2002 and can propagate along the natural fracture 2002 or branch off from the natural fracture 2002. In some cases, the natural fracture 2002 that is already in communication with the hydraulic fracture 2001 may still have its interfaces “stick” again as a result of rock deformation or pressure fluctuation. At a later time t4, the interface may slip again and emit a new microseismic event 2028.

Hydraulic fracture planes/surfaces may be directly extracted from microseismic data. Examples of methods for extracting microseismic data are provided in Fisher et al., Integrating Fracture Mapping Technologies to Optimize Stimulations in the Barnett Shale, Paper SPE 77411 presented at the 2002 SPE Annual Technical Conference and Exhibition, San Antonio, Tex., USA, Sep. 29-Oct. 2, 2002; Craig, D. P. and Burkhart, R., Using Maps of Microseismic Events to Define Reservoir Discontinuities, Paper SPE 135290 presented at SPE Annual Technical Conference and Exhibition, Florence, Italy, 19-22 September, 2010; Williams et al., Quantitative Interpretation of Major Planes From Microseismic Event Locations With Application in Production Prediction, submitted to SEG Annual Meeting (2010), and US Patent Application No. 2011/0029291, the entire contents of which are hereby incorporated by reference herein.

In at least some cases, the fracture surfaces extracted directly from the microseismic events cloud using certain methods may have large uncertainties, for example, since the events may not be at the actual hydraulic fracture surfaces as discussed above. These methods may not utilize other information, such as formation properties and pumped fluid volume. The interpretation of microseismic acoustic signals may yield information, such as the moment tensor of the microseismic source, the stress drop, and the effective area corresponding to the slip. Such information may not fully be utilized to correlate to the hydraulic fracture geometry.

To incorporate formation characterization and pumping information, hydraulic fracture models for simulating complex fracture propagation in natural fractured formations have been developed. Examples of hydraulic fracture models are provided in Weng et al., Modeling of Hydraulic Fracture Network Propagation in a Naturally Fractured Formation, Paper SPE 140253 presented at the SPE Hydraulic Fracturing Technology Conference and Exhibition held in The Woodlands, Tex., USA, 24-26 Jan. 2011 (“Weng 2011”); Cipolla et al., Integrating Microseismic Mapping and Complex Fracture Modeling to Characterize Hydraulic Fracture Complexity, Paper SPE 140185 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Tex., USA, Jan. 24-26, 2011; and Gu et al., “Hydraulic Fracture Crossing Natural Fracture at Non-Orthogonal Angles, A Criterion, Its Validation and Applications,” Paper SPE 139984 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Tex., Jan. 24-26, 2011, the entire contents of which are hereby incorporated by reference herein.

The models may consider the interaction of the hydraulic fracture with natural fractures and/or fissures, and predict detailed structure of the generated fracture networks. The models may use a simulator, such as UFM™, that may involve, a priori, a pre-defined population of natural fractures in the formation. These natural fractures may be generated based on information obtained from 3D seismic data, borehole imaging logs, and/or core characterization. The generated natural fractures may have large uncertainties that can lead to inaccurate prediction from the complex fracture simulator. Microseismic data may provide a means to validate and/or calibrate the simulation results.

Since the microseismic data may not provide precise fracture plane as discussed above, the fracture model's predicted “footprint” of the overall fracture network may be compared against an overall microseismic cloud. The model parameters may be adjusted until the model results approximately agree with the observed microseismic cloud. This calibration approach may have some inherent uncertainty, for example, where a footprint of the fracture network may not be the same as an area delineated by the microseismic cloud. This may occur, for example, where the shear failure events can be triggered at some distance from the actual fractures.

FIG. 21 is a schematic diagram 2100 depicting an example of progressive propagation of hydraulic fractures 2101a-f and natural fractures 2102a-f. Detailed complex hydraulic fracture models may be used to predict progressive propagation of multiple fracture branches in a complex fracture network. The formation initially may include many natural fractures 2102a-f.

As shown in FIG. 21, various interactions 2130a-f may occur between hydraulic fractures 2101a-f and natural fractures 2102a-f. Interaction 2130a shows no intersection between the hydraulic fracture 2101a and the natural fracture 2102a. Interaction 2130b shows arrest and/or slippage between the hydraulic fracture 2101a and the natural fracture 2102a. Interaction 2130c shows the hydraulic fracture 2101c propagating along the natural fracture 2102c and the natural fracture 2102c dilating. Interaction 2130d shows the hydraulic fracture 2101d crossing the natural fracture 2102c. Interaction 2130e shows an intersection between the hydraulic fracture 2101e and the natural fracture 2102e, with the natural fracture 2102e remaining closed. Interaction 2130f shows an intersection between the hydraulic fracture 2101f and the natural fracture 2102f, with the natural fracture 2102e having a fissure opening 2103 developing after crossing between the hydraulic fracture 2101f and the natural fracture 2102f.

In some cases, such as interactions 2130b-2130f, the hydraulic fractures 2101a-f and natural fractures 2102a-f may intersect. Interaction of the hydraulic fractures 2101a-f and the natural fracture 2102a-f may result in fracture branching where the hydraulic fractures 2101a-f and the natural fracture 2102a-f intersect. The intersections 2130a-f may result in hydraulic fractures 2101a-f opening up and propagating along the natural fractures 2102a-f and lead to fracture branching and complexity.

Characterization of natural fractures underground may be difficult, if not impossible, in some cases. Initial population of natural fractures of a discrete fracture network (DFN) may be stochastically created. The stochastic population of the DFN may be constrained by information obtained from seismic data and borehole imaging measurements, and/or utilizing geological and geostatistical models.

FIG. 22.1 shows a schematic diagram 2200.1 depicting a DFN 2232 about a wellbore 2236. Traces of statistically generated DFN are depicted near the wellbore 2236, with statistically created DFN traces uniformly distributed in a formation 2234. The traces depict natural fractures 2202 positioned about the formation 2234.

FIG. 22.2 is a schematic diagram 2200.2 showing a predicted hydraulic fracture network (HFN) 2236 simulated from the uniformly distributed DFN 2232. Hydraulic fractures 2201 are generated from a complex fracture model for the corresponding DFN 2232. FIG. 22.2 also shows microseismic events 2238 (shown as balls in the graph 2200.2) collected during the fracture treatment.

In the case depicted in FIG. 22.2, the predicted HFN 2236 footprint does not match with a microseismic cloud 2240 of the microseismic events 2238. Attempts to provide a match may be made by changing rock properties and/or initial natural fracture distribution to try to match the microseismic events 2238. It is not certain that the microseismic events 2238 represent actual hydraulic fracture planes, as they may be shear induced slippage of natural fractures 2202 away from the hydraulic fractures 2203 as already discussed above.

Forcing the complex fracture model to match the microseismic cloud 2240 may introduce error. Another approach may be to predict an induced stress field surrounding the created HFN 2236, and to determine the shear failure condition in the natural fractures and the rock matrix so the failure “footprint” approximately matches the microseismic. Additionally, from the computed stress field, the natural fractures that undergo slippage and their orientation can be determined, which can be compared to the slip orientation determined from the microseismic moment tensor to obtain more reliable interpretation.

FIGS. 23.1 and 23.2 depict methods 2300.1, 2300.2 of performing a fracture operation at a wellsite. In at least one embodiment of the present disclosure, the methods 2300.1, 2300.2 are presented for interpretation of microseismicity and its use for calibration of complex fracture simulation by coupling the stress and rock failure analysis. Each of the methods 2300.1, 2300.2 may involve 2350 performing a stimulation operation comprising stimulating the wellsite by injecting an injection fluid with proppant into the fracture network and/or 2352 generating wellsite data (e.g. natural fracture parameters of the natural fractures, pump data, and microseismic measurements) The methods 2300.1, 2300.2 may be performed with all or part of the method 1500 of FIG. 15.

The method 2300.1 involves 2354 predicting fracture geometry, 2356 determining a three dimensional (3D) stress field, and 2358 performing failure assessment and calibration against microseismic events.

Fracture Geometry Prediction

Predicting fracture geometry 2354 may be performed, for example, by 2360 modeling fractures, such as natural, hydraulic, and/or complex fractures, based on the wellsite data, and 2362 generating a discrete fracture network from wellsite data. The hydraulic fracture geometry may first be computed using a hydraulic fracture model based on known geological, geomechanical and fracture treatment data. In the case of complex fractures in a naturally fractured formation, the model can be used to predict the complex fracture planes, as well as the fracture width, fluid pressure and other parameters associated with the fracture system. Examples of modeling are provided in US Patent Application No. 2008/0183451. Predictions may be performed by simulating using, for example, UFM as described above.

3D Stress Field Computation

A three dimensional (3D) stress field may be determined 2356 by modeling. For any given hydraulic fracture geometry computed by the fracture model, the 3D stress field (or region) surrounding the hydraulic fractures (see, e.g., FIG. 19) can be computed by modeling 2364 using, for example, a numerical geomechanics model. For example, a finite element numerical geomechanics code, and/or a finite difference code can be used. Such numerical simulation may be time consuming since it involves building complex 3D fine grids surrounding each of the fractures, and may be computationally intensive. Examples of modeling are provided in Koutsabeloulis 2009 and Zhang 2007, and may employ Itasca 2002 and/or FLAC3D™ commercially available from ITASCA™ (see: http://www.itascacg.com/).

The 3D dimensional stress field may also be determined 2356 using computationally efficient methods based on Displacement Discontinuity Method (DDM). The DDM may be performed using, for example, enhanced two dimensional (2D) DDM and/or 3D DDM. Examples of

1. Enhanced 2D DDM

The method may be based on an enhanced 2D DDM 2366, such as those described herein. 2D DDM has been used in complex fracture modeling to compute the interaction among complex hydraulic fractures (also called “stress shadow” effect), and discussed herein and in PCT/US2012/063340. Examples of 2D DDM are provided in Olson 2004, and complex fracture models are provided in Weng 2011 and Wu 2012.

FIG. 3 shows a schematic diagram 300 showing a plan view of a complex fracture network 300. The fracture network 300 is discretized into many connected small elements ELEM i,j. In each element ELEM i,j, fluid pressure and width may be determined by solving a system of coupled elasticity and fluid flow equations. Examples of fluid flow in fractures are provided in Weng 2011. To account for the interaction among adjacent fractures, 2D DDM may be utilized. Examples of 2D techniques are provided in Crouch and Starfield 1983.

The 2D DDM equations relate the normal and shear stresses (σ_n and σ_s) acting on one fracture element Elem i to the contributions of the opening and shearing displacement discontinuities (D_n and D_s) from all fracture elements Elem i,j, as shown in equations below. To account for the 3D effect due to finite fracture height there is introduced a 3D correction factor 2368 to the influence coefficients C^ij and the modified elasticity equations (8.1) and (8.2) of 2D DDM as described herein. Techniques involving 3D effects are provided in Olson 2004.

The 3D correction factor may be presented as set forth in equation (12). The introduced 3D correction factor may lead to decaying of interaction between any two fracture elements when the distance increases, properly reflecting the 3D effect of finite fracture height. The enhanced 2D DDM method may be validated 2370 against 3D Finite Difference solutions in simple cases to confirm good approximations. Correction techniques are described in Wu 2012.

In the above method for stress shadow computation, the stresses may be computed 2372 at the center of each element of the hydraulic fracture network. Similar equations can be applied for by computing the stress field in the rock away from the hydraulic fracture elements. By computing the normal and shear stresses acting on portions of the discrete fracture network, such as the pre-existing natural fractures and/or any points in the rock matrix, the shear failure condition can be evaluated.

2. 3D DDM

In some cases, the enhanced 2D DDM method may be limited to an evaluation of average stresses in the horizontal plane (assuming the fractures are vertical). The method may also be based on 3D DDM 2374.

For a given hydraulic fracture network, the network may be discretized into connected small rectangular (or polygonal) elements. For any given rectangular element subjected on displacement discontinuity between its two faces represented by D_x, D_y, and D_z, the induced stresses in the rock at any point (x, y, z) can be computed using the 3D DDM solution.

FIG. 24 shows a diagram 2400 of a local x,y,z coordinate system for one of the rectangular elements 2470 positioned along an x-y plane. The induced displacement and stress field can be expressed as:u_x=[2(1−v)f_,z−zf_,xx]D_x−zf_,xyD_y−[(1−2v)f_,x+zf_,xz]D_z  (10)u_y=−zf_,xyD_x+[2(1−v)f_,z−zf_,yy]D_y−[(1−2v)f_,y+zf_,yz]D_z  (11)u_z=[(1−2v)f_,x−zf_,xz]D_x+[(1−2v)f_,y−zf_,yz]D_y+[2(1−v)f_,z−zf_,zz]D_z  (12)σ_xx=2G{[2f_,xz−zf_,xxx]D_x+[2vf_,yz−zf_,xxy]D_y+[f_,zz+(1−2v)f_,yy−zf_,xxz]D_z}  (13)σ_yy=2G{[2vf_,xz−zf_,xyy]D_x+[2f_,yz−zf_,yyy]D_y+[f_,zz+(1−2v)f_,xx−zf_,yyz]D_z}  (14)σ_zz=2G{−zf_,xzzD_x−zf_,yzz]D_y+[f_,zz−zf_,xxz]D_z}  (15)τ_xy=2G{[(1−v)f_,yz−zf_,xxy]D_x+[(1−v)f_,xz−zf_,xyy]D_y−[(1−2v)f_,xy+zf_,xyz]D_z}  (16)τ_yz=2G{−[vf_,xy+zf_,xyz]D_x+[f_,zz+vf_,xx−zf_,yyz]D_y−zf_,yzzD_z}  (17)τ_xz=2G{[(f_,zz+vf_,yy−zf_,xxz]D_x−[vf_,xy+zf_,xyz]D_y−zf_,xzzD_z}  (18)where a and b are the half lengths of the edges of the rectangle, and

$\begin{array}{cc}f\left(x,y,z\right)=\frac{1}{8\pi \left(1-v\right)}{\underset{A}{\int \int }\left[{\left(x-\xi \right)}^2+{\left(y-\eta \right)}^2+z^2\right]}^{-1/2}d\xi d\eta ,& \left(19\right)\\ \xi \le a,\eta \le b& \end{array}$ For any given observation point P (x,y,z) in the 3D space, by superposing the stresses from all fracture elements and by applying proper coordinate transform, the induced stress at the point P may be computed 2376. Techniques involving 3D DDM are provided in Crouch, S. L. and Starfield, A. M. (1990), Boundary Element Methods in Solid Mechanics, Unwin Hyman, London, the entire contents of which are hereby incorporated by reference herein.Failure Assessment and Calibration Against Microseismic Events

Failure assessment and calibration may be performed 2358 against microseismic events. The stresses can be computed in different locations in 3D space for different analysis purposes. The stresses may be generated by applying the stress field to fixed points in 3D space, to generate plots 2378 of stress components, and/or to generate stresses 2380 along observed microseismic locations. The following lists a few such applications but the method is not limited to these applications.

1. 3D stress Contour

The stress computation can be applied to fixed points in 3D space to generate contour plots 2378 of various stress components or plots of derived failure parameters from the stresses. The 3D contour plots give indication of where stress concentrations are or where the rock are most likely induce shear failure that may be correlated to the microseismic event locations or event density.

2. Stresses at Given Natural Fractures

The stresses can be computed 2380 at the natural fractures or along the natural fractures. The shear stress or other relevant indicators pertaining to failure conditions can be computed. Again, this can be compared 2382 to the microseismic locations and moment tensor attributes to determine if the assumed natural fracture parameters are consistent with the microseismic observations and if any adjustments are provided to the fracture parameters.

3. Stresses at Microseismic Event Locations

The stresses can be computed 2384 at the observed microseismic event locations. Based on the computed stresses, the likelihood of shear slippage or the boundary condition can be assessed. Since the shear slip takes place at the microseismic event location, agreement or disagreement of the model prediction with the reality may provide a measure of correctness of the model results.

Regardless where in space the stresses are computed, the comparison 2386 of the predicted propensity for shear slippage or failure can be made against the microseismic observation. If the model prediction does not agree well with the microseismic observations, modifications in natural fracture system or other rock parameters may be used and the simulation rerun until adequate match is obtained. After the adjusting 2388, the wellsite data may be modified at 2352 and the method repeated. Once the calibrations are complete, the fracture parameters may be adjusted 2388 based on the comparing. The stimulation operation 2390 may also be adjusted based on the fracture parameters.

The method provides a direct tie of the observed microseismicity and the stress field anticipated from the induced hydraulic fractures. By doing so, many effects due to initial heterogeneous stress distribution in the rock formation, variation of natural fractures and their attributes and their distribution in the reservoir, major faults with different properties, etc., can be taken into consideration. This may reduce uncertainties in the analysis and interpretation of the microseismic events and may provide more deterministic validation/calibration of the fracture geometry from the fracture model.

The calibration process may also provide better understanding of the microseismic source mechanisms and parameters, which provides the basis for improved microseismic measurement installation or design considerations in subsequent treatments in the same well, or in future treatments in the adjacent wells.

FIG. 23.2 provides another method 2300.2 of performing a fracture operation. In this version, the method involves 2350 performing a stimulation operation comprising stimulating the wellsite by injecting an injection fluid with proppant into the fracture network and 2352 generating wellsite data (e.g. natural fracture parameters of the natural fractures, pump data, and microseismic measurements) as in FIG. 23.1. The method 2300.2 also involves 2375 modeling hydraulic fractures of the fracture network based on the wellsite data and defining a hydraulic fracture geometry of the hydraulic fractures, 2377 generating a stress field of the hydraulic fractures using a geomechanical model (e.g., 2D or 3D DDM), 2379 determining shear failure parameters comprising failure envelope and a stress state about the fracture network (e.g., along the natural fractures, hydraulic fractures, and/or rock medium), 2381 determining a location of shear failure of the fracture network from the failure envelope and the stress state, 2383 calibrating the hydraulic fracture geometry by comparing the microseismic measurements with the simulated hydraulic fracture network and/or the activated discrete fracture network, 2385 adjusting the discrete fracture network based on the comparing, and 2387 adjusting the stimulation operation based on the comparing.

Part or all of the methods may be performed in any order and repeated as desired.

III. Interpretation of Microseismicity with Seismic Moment

This disclosure also relates to techniques for performing fracture operations involving modeling hydraulic and discrete fracture networks, defining shear and tensile components of the hydraulic fracture network, and determining a simulated moment density from the shear and tensile components. The discrete fracture network may be calibrated by comparing the simulated moment density with an actual moment density determined from wellsite data. This information may be used to predict proppant placement, production, and reservoir pressure.

The techniques herein may be used, for example, to extract and estimate the attributes or properties of a hydraulic (induced) fracture network from microseismic activity created during stimulation treatments in unconventional reservoirs. The techniques may not be restricted to a particular formation, well type, and/or type of array used to acquire the microseismic signal.

Microseismic evidence of fracture complexity has led to the recent development of modeling tools to simulate the growth of fracture networks. These complex fracture models may rely on calibration from microseismic location information, although microseismic source mechanics can also provide additional model verification. Modeled geomechanical deformation associated with hydraulic fracture stimulation of a complex hydraulic fracture provides information that can be compared with observed microseismic deformation. Partitioning of modeled strains into shear and dilatational components may allow relative comparison of the appropriate displacement mode with observed cumulative microseismic moments.

A number of simple fracture geometries are investigated to illustrate the deformation modes of the modeled fracture displacements. A workflow is also described where the input parameters of the simulation are varied to match both the footprint and deformation of the microseismicity, which then results in an estimate of the complete fracture network volume and proppant placement. In this way, the effective stimulated volume can be assessed and used as an input to a reservoir simulation to investigate well performance and reservoir drainage. Embodiments of the present disclosure may include one or more methods, computing devices, non-transitory computer-readable medium, and systems for microseismic fracture network (MFN) modeling.

Understanding the nature and degree of hydraulic fracture complexity may be useful to the economic development of unconventional resources. During hydraulic fracturing treatments, geomechanical interactions between hydraulic fractures and natural fractures may have an impact on the degree of complexity of the resulting fracture network. Examples of hydraulic fracture techniques are described in the following papers: Mayerhofer et al., Integrating of Microseismic Fracture Mapping Results with Numerical Fracture Network Production Modeling in the Barnett Shale, Society of Petroleum Engineers (SPE) 102103, presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Tex., 24-24 Sep. 2006; Mayerhofer et al., What is Stimulated Reservoir Volume (SRV)?, SPE 119890 presented at the SPE Shale Gas Production Conference, Fort Worth, Tex., 16-18 November 2008; Warpinski et al., Stimulating Unconventional Reservoirs: Maximizing Network Growth while Optimizing Fracture Conductivity, SPE 114173 presented at the SPE Unconventional Reservoirs Conference, Keystone, Colo., 10-12 Feb. 2008; and Cipolla et al., The Relationship between Fracture Complexity, Reservoir Properties, and Fracture Treatment Design, SPE 115769 presented at the SPE Annual Technical Conference and Exhibition, Denver, Colo., 21-24 Sep. 2008.

Complex hydraulic fracture propagation may be interpreted from microseismic measurements, for example, from unconventional reservoirs and tight gas reservoirs. Examples of complex hydraulic fracture techniques are provided in the following articles: Maxwell et al., Microseismic Imaging of Hydraulic Fracture Complexity in the Barnett Shale, SPE 77440 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Tex., Sep. 29-Oct. 2, 2002; Fisher et al., Integrating Fracture Mapping Technologies to Optimize Stimulations in the Barnett Shale, 77411 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Tex., Sep. 29-Oct. 2, 2002; Cipolla et al., Effect of Well Placement on Production and Frac Design in a Mature Tight Gas Field, 95337 presented at the SPE Annual Technical Conference and Exhibition, Dallas, Tex., 9-12 Oct. 2005; and Warpinski et al., Stimulating Unconventional Reservoirs: Maximizing Network Growth while Optimizing Fracture Conductivity, SPE 114173 presented at the SPE Unconventional Reservoirs Conference, Keystone, Colo., 10-12 Feb. 2008.

Stimulation and completion design decisions may be made based on the anticipated fracture complexity, which may be a factor for the ultimate well performance. Geomechanical analysis tools may be used to simulate the fracture network resulting from hydraulic fracture stimulation of a pre-existing discrete fracture network (DFN). In some cases, challenges may exist in distinguishing between small scale fracture complexity and simple planar fracture growth. A factor that may influence the creation of complex fracture systems is the presence and distribution of natural fractures. An example of complex fractures is shown in Cipolla et al., Integrating Microseismic Mapping and Complex Fracture Modeling to Characterize Fracture Complexity, SPE 140185 presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Tex., 24-26 Feb. 2011. DFN models have been used to simulate production in naturally fractured reservoirs as shown, for example, in the following papers: Dershowitz et al., A Workflow for Integrated Barnett Shale Reservoir Modeling and Simulation, SPE 122934 presented at the SPE Latin American and Caribbean Petroleum Engineering Conference, Cartagena, Columbia, 31 May-3 Jun. 2009; Qui et al., Applying Curvature and Fracture Analysis to the Placement of Horizontal Wells: Example from the Mabee (San Adres) Reservoir, Tex., SPE 70010 presented at the SPE Permian Basin Oil and Gas Recovery Conference, Midland, Tex. 15-17 May 2001; and Will et al., Integration of Seismic Anisotropy and Reservoir-Performance Data for Characterization of Naturally Fractured Reservoirs Using Discrete-Feature-Network Models, SPE 84412 presented at the SPE Annual Technical Conference and Exhibition, Denver, Colo., 5-8 Oct. 2003. These methods, along with log-based approaches (see, e.g., Bratton et al., Rock Strength Parameters from Annular Pressure While Drilling and Dipole Sonic Dispersion Analysis, Presented at the SPWLA 45th Annual Logging Symposium, Noordwijk, The Netherlands, 6-9 Jun. 2004) may be descriptive. Some such methods may be used to characterize a structure of the natural fracture network by using seismic information to extend observations at the wellbore across the reservoir.

Some models have also been developed to quantify the propagation of complex hydraulic fracture networks in, for example, formations embedded with predefined, deterministic or stochastic natural fractures. Examples of complex fracture models are described in the following: Sahimi, M., New Models For Natural And Hydraulic Fracturing On Heterogeneous Rock, SPE 29648 presented at the SPE Western Regional Meeting, Bakersfield, Calif. (1995); Fomin et al., Advances In Mathematical Modeling Of Hydraulic Stimulation Of A Subterranean Fractured Reservoir, Proc. SPIE 5831: 148-154 (2005); Napier et al., Comparison Of Numerical And Physical Models For Understanding Shear Fracture Process, Pure Appl. Geophys, 163: 1153-1174 (2006); Tezuka et al., Fractured Reservoir Characterization Incorporating Microseismic Monitoring And Pressure Analysis During Massive Hydraulic Injection, IPTC 12391 presented at the International Petroleum Technology Conference, Kuala Lumpur, Malaysia (2008); Olsen et al., Modeling Simultaneous Growth Of Multiple Hydraulic Fractures And Their Interaction With Natural Fractures, SPE 119739 presented at the Hydraulic Fracturing Technology Conference, The Woodlands, Tex. (2009); and Xu et al., Characterization of Hydraulically Induced Shale Fracture Network Using an Analytical/Semi-Analytical Model, SPE 124697 presented at the SPE Annual Technical Conference and Exhibition, New Orleans, 4-7 Oct. 2009; and Weng et al., Modeling of Hydraulic Fracture Propagation in a Naturally Fractured Formation, SPE 140253 presented at the SPE Hydraulic Fracturing Technology Conference, Woodlands, Tex., USA, 24-26 Jan. 2011. In some models, microseismic activity may be used to constrain the fracturing process.

INTRODUCTION

FIGS. 25.1-25.4 illustrate simplified, schematic views of oilfield 2500 having subterranean formation 2502 containing reservoir 2504 therein in accordance with implementations of various technologies and techniques described herein. FIG. 25.1 illustrates a survey operation being performed by a survey tool, such as seismic truck 2506.1, to measure properties of the subterranean formation. The survey operation is a seismic survey operation for producing sound vibrations. In FIG. 25.1, one such sound vibration, sound vibration 2512 generated by source 2510, reflects off horizons 2514 in earth formation 2516. A set of sound vibrations is received by sensors, such as geophone-receivers 2518, situated on the earth's surface. The data received 2520 is provided as input data to a computer 2522.1 of a seismic truck 2506.1, and responsive to the input data, computer 2522.1 generates seismic data output 2524. This seismic data output may be stored, transmitted or further processed as desired, for example, by data reduction. The surface unit 2534 is also depicted as having a microseismic fracture operation system 2550 as will be described further herein.

FIG. 25.2 illustrates a drilling operation being performed by drilling tools 2506.2 suspended by rig 2528 and advanced into subterranean formations 2502 to form wellbore 2536. Mud pit 2530 is used to draw drilling mud into the drilling tools via flow line 2532 for circulating drilling mud down through the drilling tools, then up wellbore 2536 and back to the surface. The drilling mud may be filtered and returned to the mud pit. A circulating system may be used for storing, controlling, or filtering the flowing drilling muds. The drilling tools are advanced into subterranean formations 2502 to reach reservoir 2504. Each well may target one or more reservoirs. The drilling tools are adapted for measuring downhole properties using logging while drilling tools. The logging while drilling tools may also be adapted for taking core sample 2533 as shown.

Computer facilities may be positioned at various locations about the oilfield 2500 (e.g., the surface unit 2534) and/or at remote locations. Surface unit 2534 may be used to communicate with the drilling tools and/or offsite operations, as well as with other surface or downhole sensors. Surface unit 2534 is capable of communicating with the drilling tools to send commands to the drilling tools, and to receive data therefrom. Surface unit 2534 may also collect data generated during the drilling operation and produces data output 2535, which may then be stored or transmitted.

Sensors (S), such as gauges, may be positioned about oilfield 2500 to collect data relating to various oilfield operations as described previously. As shown, sensor (S) is positioned in one or more locations in the drilling tools and/or at rig 2528 to measure drilling parameters, such as weight on bit, torque on bit, pressures, temperatures, flow rates, compositions, rotary speed, and/or other parameters of the field operation. Sensors (S) may also be positioned in one or more locations in the circulating system.

Drilling tools 2506.2 may include a bottom hole assembly (BHA) (not shown) near the drill bit (e.g., within several drill collar lengths from the drill bit). The bottom hole assembly includes capabilities for measuring, processing, and storing information, as well as communicating with surface unit 2534. The bottom hole assembly further includes drill collars for performing various other measurement functions.

The bottom hole assembly may include a communication subassembly that communicates with surface unit 2534. The communication subassembly is adapted to send signals to and receive signals from the surface using a communications channel such as mud pulse telemetry, electro-magnetic telemetry, or wired drill pipe communications. The communication subassembly may include, for example, a transmitter that generates a signal, such as an acoustic or electromagnetic signal, which is representative of the measured drilling parameters. It will be appreciated by one of skill in the art that a variety of telemetry systems may be employed, such as wired drill pipe, electromagnetic or other known telemetry systems.

The wellbore may be drilled according to a drilling plan that is established prior to drilling. The drilling plan may set forth equipment, pressures, trajectories and/or other parameters that define the drilling process for the wellsite. The drilling operation may then be performed according to the drilling plan. However, as information is gathered, the drilling operation may to deviate from the drilling plan. Additionally, as drilling or other operations are performed, the subsurface conditions may change. The earth model may also provide adjustment as new information is collected.

The data gathered by sensors (S) may be collected by surface unit 2534 and/or other data collection sources for analysis or other processing. The data collected by sensors (S) may be used alone or in combination with other data. The data may be collected in one or more databases and/or transmitted on or offsite. The data may be historical data, real time data, or combinations thereof. The real time data may be used in real time, or stored for later use. The data may also be combined with historical data or other inputs for further analysis. The data may be stored in separate databases, or combined into a single database.

Surface unit 2534 may include transceiver 2537 to allow communications between surface unit 2534 and various portions of the oilfield 2500 or other locations. Surface unit 2534 may also be provided with or functionally connected to one or more controllers (not shown) for actuating mechanisms at oilfield 2500. Surface unit 2534 may then send command signals to oilfield 2500 in response to data received. Surface unit 2534 may receive commands via transceiver 2537 or may itself execute commands to the controller. A processor may be provided to analyze the data (locally or remotely), make the decisions and/or actuate the controller. In this manner, oilfield 2500 may be selectively adjusted based on the data collected. This technique may be used to optimize portions of the field operation, such as controlling drilling, weight on bit, pump rates, or other parameters. These adjustments may be made automatically based on computer protocol, and/or manually by an operator. In some cases, well plans may be adjusted to select optimum operating conditions, or to avoid problems. The surface unit 2534 is also depicted as having a microseismic fracture operation system 2550 as will be described further herein.

FIG. 25.3 illustrates a wireline operation being performed by wireline tool 2506.3 suspended by rig 2528 and into wellbore 2536 of FIG. 25.2. Wireline tool 2506.3 is adapted for deployment into wellbore 2536 for generating well logs, performing downhole tests and/or collecting samples. Wireline tool 2506.3 may be used to provide another method and apparatus for performing a seismic survey operation. Wireline tool 2506.3 may, for example, have an explosive, radioactive, electrical, or acoustic energy source 2544 that sends and/or receives electrical signals to surrounding subterranean formations 2502 and fluids therein.

Wireline tool 2506.3 may be operatively connected to, for example, geophones 2518 and a computer 2522.1 of a seismic truck 2506.1 of FIG. 25.1. Wireline tool 2506.3 may also provide data to surface unit 2534. Surface unit 2534 may collect data generated during the wireline operation and may produce data output 2535 that may be stored or transmitted. Wireline tool 2506.3 may be positioned at various depths in the wellbore 2536 to provide a surveyor other information relating to the subterranean formation 2502.

Sensors (S), such as gauges, may be positioned about oilfield 2500 to collect data relating to various field operations as described previously. As shown, sensor S is positioned in wireline tool 2506.3 to measure downhole parameters which relate to, for example porosity, permeability, fluid composition and/or other parameters of the field operation.

FIG. 25.4 illustrates a production operation being performed by production tool 2506.4 deployed from a production unit or Christmas tree 2529 and into completed wellbore 2536 for drawing fluid from the downhole reservoirs into surface facilities 2542. The fluid flows from reservoir 2504 through perforations in the casing (not shown) and into production tool 2506.4 in wellbore 2536 and to surface facilities 2542 via gathering network 2546.

Sensors (S), such as gauges, may be positioned about oilfield 2500 to collect data relating to various field operations as described previously. As shown, the sensor (S) may be positioned in production tool 2506.4 or associated equipment, such as Christmas tree 2529, gathering network 2546, surface facility 2542, and/or the production facility, to measure fluid parameters, such as fluid composition, flow rates, pressures, temperatures, and/or other parameters of the production operation.

Production may also include injection wells for added recovery. One or more gathering facilities may be operatively connected to one or more of the wellsites for selectively collecting downhole fluids from the wellsite(s).

While FIGS. 25.2-25.4 illustrate tools used to measure properties of an oilfield, it will be appreciated that the tools may be used in connection with non-oilfield operations, such as gas fields, mines, aquifers, storage, or other subterranean facilities. Also, while certain data acquisition tools are depicted, it will be appreciated that various measurement tools capable of sensing parameters, such as seismic two-way travel time, density, resistivity, production rate, etc., of the subterranean formation and/or its geological formations may be used. Various sensors (S) may be located at various positions along the wellbore and/or the monitoring tools to collect and/or monitor the desired data. Other sources of data may also be provided from offsite locations.

The field configurations of FIGS. 25.1-25.4 are intended to provide a brief description of an example of a field usable with oilfield application frameworks. Part, or all, of oilfield 2500 may be on land, water, and/or sea. Also, while a single field measured at a single location is depicted, oilfield applications may be utilized with any combination of one or more oilfields, one or more processing facilities and one or more wellsites.

FIG. 25.5 depicts the microseismic fracture operation system 2550. As shown, the microseismic fracture operation system 2550 includes a microseismic tool 2552, a fracture tool 2554, a wellsite tool 2556, an optimizer 2558 and an oilfield tool 2560. The microseismic tool 2552 may be used to perform Ant-tracking. The fracture tool 2554 may be used to perform fracture extraction. The wellsite tool 2556 may be used to generate fracture attributes, such as permeabilities. The optimizer 2558 may be used to perform dynamic modeling and adjust the fracture attributes based on the dynamic modeling. The oilfield tool 2560 may be used to obtain wellsite data from, for example, the sensors S from FIGS. 25.1-25.4 and manipulate the data for use by the other tools of the microseismic fracture operation system 2550. Each of these functions is described further herein.

FIG. 26 illustrates a schematic view, partially in cross section of oilfield 2600 having data acquisition tools 2602.1, 2602.2, 2602.3 and 2602.4 positioned at various locations along oilfield 2600 for collecting data of subterranean formation 2604 in accordance with implementations of various technologies and techniques described herein. Data acquisition tools 2602.1-2602.4 may be the same as data acquisition tools 2506.1-2506.4 of FIGS. 25.1-25.4, respectively, or others not depicted. As shown, data acquisition tools 2602.1-2602.4 generate data plots or measurements 2608.1-2608.4, respectively. These data plots are depicted along oilfield 2600 to demonstrate the data generated by the various operations.

Data plots 2608.1-2608.3 are examples of static data plots that may be generated by data acquisition tools 2602.1-2602.3, respectively, however, it should be understood that data plots 2608.1-2608.3 may also be data plots that are updated in real time. These measurements may be analyzed to better define the properties of the formation(s) and/or determine the accuracy of the measurements and/or for checking for errors. The plots of each of the respective measurements may be aligned and scaled for comparison and verification of the properties.

Static data plot 2608.1 is a seismic two-way response over a period of time. Static plot 2608.2 is core sample data measured from a core sample of the formation 2604. The core sample may be used to provide data, such as a graph of the density, porosity, permeability, or some other physical property of the core sample over the length of the core. Tests for density and viscosity may be performed on the fluids in the core at varying pressures and temperatures. Static data plot 2608.3 is a logging trace that may provide a resistivity or other measurement of the formation at various depths.

A production decline curve or graph 2608.4 is a dynamic data plot of the fluid flow rate over time. The production decline curve may provide the production rate as a function of time. As the fluid flows through the wellbore, measurements are taken of fluid properties, such as flow rates, pressures, composition, etc.

Other data may also be collected, such as historical data, user inputs, economic information, and/or other measurement data and other parameters of interest. As described below, the static and dynamic measurements may be analyzed and used to generate models of the subterranean formation to determine characteristics thereof. Similar measurements may also be used to measure changes in formation aspects over time.

The subterranean structure 2604 has a plurality of geological formations 2606.1-2606.4. As shown, this structure has several formations or layers, including a shale layer 2606.1, a carbonate layer 2606.2, a shale layer 2606.3 and a sand layer 2606.4. A fault 2607 extends through the shale layer 2606.1 and the carbonate layer 2606.2. The static data acquisition tools are adapted to take measurements and detect characteristics of the formations.

While a specific subterranean formation with specific geological structures is depicted, it will be appreciated that oilfield 2600 may contain a variety of geological structures and/or formations, sometimes having extreme complexity. In some locations, for example below the water line, fluid may occupy pore spaces of the formations. Each of the measurement devices may be used to measure properties of the formations and/or its geological features. While each acquisition tool is shown as being in specific locations in oilfield 2600, it will be appreciated that one or more types of measurement may be taken at one or more locations across one or more fields or other locations for comparison and/or analysis.

The data collected from various sources, such as the data acquisition tools of FIG. 26, may then be processed and/or evaluated. The seismic data displayed in static data plot 2608.1 from data acquisition tool 2602.1 is used by a geophysicist to determine characteristics of the subterranean formations and features. The core data shown in static plot 2608.2 and/or log data from well log 2608.3 may be used by a geologist to determine various characteristics of the subterranean formation. The production data from graph 2608.4 may be used by the reservoir engineer to determine fluid flow reservoir characteristics. The data analyzed by the geologist, geophysicist and the reservoir engineer may be analyzed using modeling techniques.

FIG. 27 illustrates an oilfield 2700 for performing production operations in accordance with implementations of various technologies and techniques described herein. As shown, the oilfield has a plurality of wellsites 2702 operatively connected to central processing facility 2754. The oilfield configuration of FIG. 27 is not intended to limit the scope of the oilfield application system. Part or all of the oilfield may be on land and/or sea. Also, while a single oilfield with a single processing facility and a plurality of wellsites is depicted, any combination of one or more oilfields, one or more processing facilities and one or more wellsites may be present.

Each wellsite 2702 has equipment that forms wellbore 2736 into the earth. The wellbores extend through subterranean formations 2706 including reservoirs 2704. These reservoirs 2704 contain fluids, such as hydrocarbons. The wellsites draw fluid from the reservoirs and pass them to the processing facilities via surface networks 2744. The surface networks 2744 have tubing and control mechanisms for controlling the flow of fluids from the wellsite to processing facility 2754.

Microseismic Source Characterization

Beyond hypocentral location and the temporal relationship with the injection program, there are two aspects of the microseismic source deformation that may be relevant in providing insight into the geomechanical deformations of the hydraulic fracture network. The first is the scalar seismic moment (M₀), which relates the microseismic source strength to the coseismic strain measure via the product of the slip area (A) and displacement (d):M₀=μAd  (20)where μ is the shear modulus.

The magnitude measure of the microseismic source strength can be estimated by the moment magnitude (Mw) (see, e.g., Hanks and Kanamori, A Moment Magnitude Scale, Journal of Geophysical Research, Vol. 84, Issue B5, pp. 2348-50, 1979 (referred to herein as “Hanks and Kanamori”)):M_w=⅔ log(M₀)−6.  (21)The slip displacement or strain is an attribute that can be directly estimated with a numeric geomechanical simulation, such that equivalent moments or moment magnitude can be estimated from the simulation.

The second aspect of the microseismic source is the source focal mechanism. The focal mechanism refers to the orientation of a fault plan that has slipped, and can be derived from a solution of the moment tensor which may be estimated by an analysis of observed seismic waveforms. Focal mechanisms can be used to estimate the fracture orientation of the microseismic source using a variety of methods. In particular, moment tensor inversion methods can also be used to estimate the mode of the microseismic source slip and whether shear, tensile opening or a combination has occurred (see, e.g., Hanks and Kanamori, 1979). For a given fracture segment orientation within a DFN, geomechanical simulations can also predict the comparable mode of slip.

Microseismic source characterization can therefore provide deformation characteristics consistent with the aspects of geomechanical simulations of fracture network strains. The recorded microseismicity represents a component of the total fracture network deformation, although aseismic deformation may also occur and may represents a component of the fracture strains. Once the mode of the microseismicity is determined, the corresponding geomechanical mode of failure can be quantitatively compared with the numerical simulations.

Fracture Network Deformation Modes

FIGS. 28 and 29.1-35.2 show various cases of fracture geometry depicting shear and tensile deformation of a hydraulic fracture. In each figure, the fracture 2923, 2923′ is depicted in lighter gray, and shear and tensile stresses applied thereto are depicted in darker shading. In order to illustrate the relative deformation modes that result from a hydraulic fracture treatment, a number of simple fracture geometries may be simulated. For each of the geometries, the subsequent fracture strains may be estimated and projected into shear and tensile components. Strains may be estimated from a fracture mechanics model that honors the mass balance of the injection in order to generate sufficient fracture volume to contain the injected fluid volume via the generation of hydraulic fractures that interact with preexisting fractures.

During the fracture dilation, associated geomechanical strains are computed which may include both tensile and shear displacements depending on the dilatational characteristics of the fracture network. Through the remainder of the discussion the deformation may focus on inelastic displacements of the hydraulic fracture network itself. The elastic changes in the rock around the dilating fracture network and any associated induced displacement of preexisting fractures that may be disconnected from the hydraulic fracture may or may not be considered.

FIG. 28 depicts conceptualized growth of a hydraulic fracture 2823 over time. Stage 1) depicts earliest time the hydraulic fracture 2823 grows outwards from an injection point 2817 towards a preexisting fracture 2819. Stage 2) shows the hydraulic fracture 2823′ as it grows into the preexisting fracture 2819, filling with fluid and starting to dilate. Stage 3) shows the hydraulic fracture 2823″ as it continues to grow, creating a new fracture 2823.1 at the end of the preexisting hydraulic fracture 2823″.

Opening modes of the fractures 2823, 2823′, 2823″, 2823.1 at the various stages results in tensile opening 2825, and also induces localized shearing 2827. These fracture segments have a potential combination of tension and shear displacements as shown in Table 5 below:

TABLE 5
Tension and Shear Displacement of Hydraulic Fracture Over Time
Stage	σ	τ
1)	X	—
2)	X	X
3)	X	X

FIGS. 29.1-35.2 depict various examples of hydraulic fracture creation for the cases as set forth in Table 6 below:

TABLE 6
Shear And Tensile Deformation For The Various Cases
	total shear	total tensile	Percent	max shear	max tensile	Percent
Case #	(m)	(m)	Total Shear	(m)	(m)	Max Shear
1 - No	0	0.5483	0.0%	0	3.11E−03	0.0%
fractures
2 - Single	0.078	0.5342	14.6%	2.13E−03	3.39E−03	62.9%
Asymmetric
3 - Single	0.1051	0.4469	23.5%	1.52E−03	3.77E−03	40.4%
Symmetric
4 - Multiple	0.1152	0.4643	24.8%	2.09E−03	3.82E−03	54.7%
Symmetric
5 - Long	0.1286	0.573	22.4%	1.35E−03	3.98E−03	33.9%
Symmetric
6 - Long	0.1119	0.5412	20.7%	1.61E−03	3.64E−03	44.3%
Asymmetric
7 - Short	0.0806	0.5411	14.9%	1.19E−03	3.52E−03	33.8%
Symmetric

Table 6 summarizes the total deformation types as well as the localized maximum deformation (which can be thought of as the localized shearing).

Each case 1-7 in Table 6 above is depicted in a pair of figures including both a shear τ plot and a tensile σ plot. FIGS. 29.1-35.1 depict total shear τ for a fracture 2923 plotted along X(m) (x-axis) versus Y(m) (y-axis). FIGS. 29.2-35.2 depicts total tensile σ for a fracture 2923′ plotted along X(m) (x-axis) versus total Y(m) (y-axis). Each of the cases 1-7 is described in further detail below:

Case 1: No Preexisting Fractures

In this simplest scenario, a planar hydraulic fracture 2923 is created (FIGS. 29.1) which deforms to hydraulic fracture 2923′ through tensile opening with no associated shear strain (FIG. 29.2). In each of these scenarios, the injection point is at x=0 from which an east-west tensile hydraulic fracture 2923 grows. Although a tensile fracture creates shear stress lobes in the rock near the tensile fracture tip, the opposing fracture faces experience relative opening displacements, unless a fracture encounters a preexisting fracture in a desired slip direction. In this example with no preexisting fractures, a tensile hydraulic fracture 2923′ is generated with purely tensile opening and no shear deformations.

In the following Cases 2-4, east-west preexisting hydraulic fractures 2923, 2923′ are used in addition to the north-south pre-existing fracture 2919.1-.4, 2919.1-.4′ to generate the specific geometries. In this case, shear deformations are created along the hydraulic fracture branch 2923 and more localized along a short ‘dogleg’ portion of the preexisting fracture 2919.1-.4, 2919.1-.4′. A small amount of shear is caused along the initial hydraulic fracture 2923 due to the asymmetry of the cross-cutting pre-existing fracture 2919 and associated dilation of the segment leading to branching in the form of a dog-leg. As indicated in Table 6 above, the shear deformation along the dog-leg may be the largest localized shear found for the scenarios, and the total shear may be relatively small.

Case 2: Single, Asymmetric Preexisting Fracture

FIG. 30.1 depicts simulated shear of the hydraulic fracture 2923. FIG. 30.2 depicts simulated tensile deformation associated with the tensile hydraulic fracture 2923′. The horizontal portion defined by the original hydraulic fracture 2923 represents the stimulated fracture network. Portions 2919.1 extending beyond hydraulic fracture 2923 are represented as shear displacements in FIG. 30.1 and portions 2919.1′ extending beyond hydraulic fracture 2923′ are tensile displacements in FIG. 30.2.

In this scenario, the single, planar, tensile hydraulic fracture 2923 of FIG. 29.1 is initially created which eventually grows into a preexisting north-south fracture 2919.1 of FIG. 30.1. Here, the preexisting fracture is asymmetric about the injection point and hydraulic fracture, resulting in a single branched fracture network (FIGS. 30.1 and 30.2).

FIG. 30.1 depicts shear and FIG. 30.2 depicts tensile displacements associated with an asymmetric cross-cutting fracture 2919.1, 2919.1′ about hydraulic fracture 2923, 2923′. Note the preexisting hydraulic fractures 2923, 2923′ are purposely arranged to create a symmetric fracture 2923.2 about the injection point x=0.

Case 3: Single, Symmetric Preexisting Fracture

FIG. 31.1 depicts shear and FIG. 31.2 tensile displacements associated with a symmetric cross-cutting fracture. Here a single, planar hydraulic fracture 2923 intersects a symmetric fracture 2919.2 generating a double branching hydraulic fracture (FIG. 31.1). Shear is developed along the cross-cutting fracture 2923′ and along each of the branching fractures 2919.2′.

In contrast to the asymmetric case of FIG. 30.1, shear is developed along the entire length resulting in a more extensive shearing structure (see Table 6). Another difference from the asymmetric case is there is no shear along the initial hydraulic fracture 2923′.

Case 4: Multiple, Symmetric Preexisting Fractures

FIG. 32.1 depicts shear and FIG. 32.2 depicts tensile displacements associated with a multiple, symmetric cross-cutting fractures 2923, 2923′. A variation of the single, symmetric, double-branching fracture 2919.2, 2919.2′ (case 3) includes an additional preexisting fracture 2919.2.1, 2919.2.1′ parallel with the first which creates additional branching (FIGS. 32.1 and 32.2). A similar pattern of deformation to case 2 is found with additional shearing along the dogleg structure. The increased fracture branches of the fracture network results in further increased total shearing (Table 5).

For the following Cases 5-7 no additional east-west fractures are included in addition to the north-south. Shear deformation occurs over the preexisting north-south fracture.

Case 5: Long, Symmetric Fracture

FIG. 33.1 depicts shear displacements 2919.3 and FIG. 33.2 depicts tensile displacements 2919.3′ associated with a long, symmetric cross-cutting fracture 2923,2923′. In this scenario, a relatively long, cross-cutting fracture is simulated (FIGS. 33.1 and 33.2).

Case 6: Long, Asymmetric Fracture

FIG. 34.1 depicts shear displacements 2919.4 and FIG. 34.2 depicts tensile displacements 2919.4′ associated with a long, asymmetric cross-cutting fracture 2923. With the preexisting fracture 2923 asymmetric about the initial hydraulic fracture, an additional fracture branch 2937 grows off the closest end of the fracture (FIG. 34.1).

Shear is created on both the cross-cutting 2923 and branching fracture 2919.4. A small amount of shear is created on the central hydraulic fracture 2923′, similar to case 2. Note that the branching fracture 2919.4′, 2937′ about hydraulic fracture 2923′ is at an angle due to stress shadowing associated with the shear along the cross cutting fracture.

Case 7: Short, Symmetric Fracture

FIG. 35.1 depicts shear displacements 2919.5, 2937.1 and FIG. 35.2 depicts tensile displacements 2919.5′, 2937.1′ associated with a short, symmetric cross-cutting fracture 2923, 2923′. In this scenario, a short, symmetric fracture 2923 is simulated (FIGS. 35.1 and 35.2). Again two branching fractures 2923.5, 2937.1 are generated off the orthogonal cross-cutting fracture 2923 and two branching fractures 2923.5′, 2937.1′ are generated off the orthogonal cross-cutting fracture 2923′, with shearing components along both. As indicated in Table 6, less total shearing is generated compared to the longer fracture scenario (#5) similar to the comparison between cases 2 and 3.

Based on the above cases, it may be determined that: (1) the more complex the fracture network and the greater the preexisting fracture density may be, the more the shear deformation; (2) longer cross-cutting fractures may produce more shearing; (3) asymmetric fractures may produce less total shear, and larger localized shear; (4) asymmetric fractures may produce a small amount of shear on the original tensile hydraulic fracture; and (5) shear deformation itself may not be a good proxy for the amount of tensile deformation. In view of these and other considerations, methods may be provided for performing fracture operations that take into consideration fracture geometry, and shear and tensile deformation of the fracture network.

Microseismic Validation

Modeled geomechanical deformation associated with hydraulic fracture stimulation of a complex hydraulic fracture provides context for interpretation of microseismic deformation. Partitioning of modeled strains into shear and tensile (dilatational) components allows relative comparison of the appropriate displacement mode with observed cumulative microseismic moments. Input parameters of a simulation may be varied to match both the footprint and deformation of the microseismicity, which then results in an estimate of the complete fracture network volume and proppant placement. In this way, the effective stimulated volume can be assessed and used as an input to a reservoir simulation to investigate well performance and reservoir drainage.

Microseismic monitoring may be used to image hydraulic fracture stimulation of unconventional reservoirs. The timing and location of microseismicity may be used to interpret the geometry and hydraulic fracture growth. Microseismic waveforms also contain information about inelastic deformation that can also be used to characterize the hydraulic fracture. The detected microseismic activity represents a portion of the geomechanical deformations associated with the hydraulic fracturing. See, e.g., Maxwell, S. C., “What Does Microseismic Tell Us About Hydraulic Fracture Deformation,” Recorder, 29-43, October, 2011 (referred herein to as “Maxwell 2011”). The detected movements may be restricted to time scales corresponding to bandwidth of the monitoring equipment. In at least some scenarios, the microseismicity corresponds to shear deformations, and the hydraulic fracture may be considered a tensile parting of the rock. Therefore, aseismic deformation may include an aspect of the fracture movements beyond what is observed through microseismicity (see, e.g., Maxwell, 2011). These deformations may be taken into consideration in analyzing fracture networks.

Locations of microseismicity may be used to constrain the fracture network. For a specific stress state, complex hydraulic fracture networks can be modeled for a given discrete fracture network (DFN) of preexisting fractures. The DFN may be adjusted to match the observed extent of the microseismicity. A DFN can be constructed using formation image logs and seismically derived fractures. In some case, there may be uncertainties in various aspects of the DFN some of which can be constrained using microseismic data.

Various techniques exist to use microseismic locations to directly image discrete fractures, particularly if high resolution locations have been computed (e.g. double difference, relative picking). In another example, location uncertainty can be minimized using clustering or collapsing algorithms. Microseismic trends can also be identified from sets of locations using various methods. Microseismic can also be used to statistically define various attributes of the DFN. For example, microseismic source mechanisms can be used to define fracture orientations. Microseismic source radius of slip (derived from the frequency content) can help constrain the length distributions.

One aspect is the fracture density, which can potentially be determined from the density of the microseismicity. Although the microseismic event count density could potentially be used, the shear displacement distribution is also related to the fracture density. Indeed, the modeled fracture displacements could be directly quantified as a seismic moment density and compared to the observed seismic moment density. Seismic moment density may be expressed as:M_pq=u_iv_jc_ijpq  (22)where u_i is displacement discontinuity across the fault zone, v_j is fault normal direction, c_ijpq is the elastic tensor of a source region and holds for arbitrary anisotropy. Others have compared the observed microseismic deformation in context of the entire deformation that occurs during the hydraulic fracture, based on either mass or energy balance considerations.

At least some deformation is found to occur aseismically, either too low amplitude to be measured or at characteristics time scales beyond what can be detected with conventional seismic instrumentation. In particular, a portion of the tensile deformation related to opening of fractures may be expected to be aseismic. Therefore, an accounting of the aseismic deformation may be used in a comparison between the modeled and observed seismic moment density. A relative comparison can be made between the modeled and observed seismic moments which can potentially assist in constraining the relative spatial heterogeneity of the fracture density. In the following case study, an example will be given of comparing the microseismic deformation to modeled deformation of the fracture network.

The ability to simulate hydraulic fracture growth may be used in frac engineering design. Hydraulic fracture stimulations can be modeled through fracture mechanics models that simulate the fracture dilation/strain, leak-off, hydraulic conductivity and associated pressure profile for a given injection volume. Models exist for simple scenarios of relatively planar, 2D fractures. In cases of complex fracture networks, modeling capabilities may be used to address creation of new hydraulic fractures and/or activation of pre-existing natural fractures which result in coupled geomechanical and hydraulic interaction between individual components of the fracture network. See Weng, X., Kresse, O., Cohen, C., Wu, R. and Gu, H., “Modeling of Hydraulic Fracture Network Propagation in a Naturally Fractured Formation,” SPE140253, (2011).

The complexity of a hydraulic fracture network may depend in part on differential stress and strength of the various fractures in the DFN: with planar fractures favored in scenarios of large differential stress and strong fractures and fracture networks in scenarios of low differential stress and weak fractures. The fracture complexity may be difficult to predict a priori, due to reservoir heterogeneity and interactions of treatment and completion designs. Before the hydraulic fracture treatment, geomechanical simulations can provide deterministic predictions of the fracture networks for specific scenarios. After the treatment, the geomechnical prediction can be calibrated with corresponding measurements, including microseismic.

Microseismic provides observations to validate these geomechanical predicted networks either simply by comparison with the extent of the observed microseismically active region or through quantification of the observed deformation using microseismic source characterization. The observed microseismic deformation may represent just a portion of the complete deformation, such as the relatively rapid fracture movements and/or the shearing components. Therefore, geomechanical model validation or calibration may involve portioning of the fracture network strains into components consistent with both the microseismic and aseismic elements.

In one aspect presented herein, a comparison of microseismic deformation and modeled geomechanical strains, and a workflow to calibrate a fracture network model are discussed. Aspects of microseismic source characterization and how it can be used to supplement a DFN and mechanical earth model will be provided, quantifying fracture deformation. Interaction of hydraulic fractures and simplistic fracture geometries are shown herein in part to illustrate certain factors controlling the deformation modes. Non-limiting examples are presented herein to describe the partitioning of modeled strains into shear and dilatational components followed by a relative comparison of the appropriate displacement mode with observed cumulative microseismic moments.

Based on the example(s) disclosed herein, a workflow is presented where the input parameters of the simulation may be varied to match both the footprint and deformation of the microseismicity, which then results in an estimate of the complete fracture network volume and proppant placement. In this way, effective stimulated volume can be assessed and used as input to a reservoir simulation to investigate well performance and reservoir drainage.

Example—Four Stage Hydraulic Fracture Simulation

1. Geomechanical Fracture Network Simulation

In the example depicted in FIG. 36, a four stage hydraulic fracture stimulation of the wellbore 1204 of FIG. 12 is depicted. FIG. 36 is the same as FIG. 12, except that a fracture network 3645 is shown about the treatment well 1204 and the monitor well 1205. As shown in FIG. 36, microseismic events 1223 are mapped in stages 1-4 and depicted in microseismic clusters 1223.1-1223.4, respectively, about a wellbore 1204.

Stress variation through the reservoir is believed to have resulted in a change in fracture geometry from relatively narrow, planar fractures for the first two toe stages to a wider, complex fracture network for the final heel stages. A fracture network simulation may be created and calibrated to approximate the spatial extent of the microseismicity as demonstrated by the microseismic events 1223 shown in FIG. 36. The clusters of microseismic events 1223.1-.4 in each of the stages 1-4 depict fracture network segments approximate the extent of the microseismicity.

A fracture network simulation was created and calibrated to approximate the spatial extent of the microseismicity as shown in FIG. 37. FIG. 37 is a plot 3700 illustrating a simulated hydraulic fracture network 3723 corresponding to the microseismic events 1223 of FIG. 36. The plot 3700 is depicted along direction Y (north) (m) (y-axis) versus X (east) (m) (x-axis). The simulation may be performed using the same techniques as set forth in FIGS. 14.1-14.4 above. In this case, the simulated hydraulic fracture network 3723 includes four fracture network segments (or portions) 3723.1-.4 corresponding to the microseismic event clusters 1223.1-.4 of FIG. 36. These fracture network segments 3723.1-.4 approximate the extent of microseismicity in FIG. 36. A closer match may be created by modifying the geometry of the input pre-existing fractures of the fracture network 3645 of FIG. 36.

2. Shear and Tensile Deformations

FIGS. 38.1-39.2 depict modeled and observed deformations of the fracture network 3723 of FIG. 37. The modeled and observed deformations may be compared, and the modeled deformations converted to an effective seismic moment. An implicit assumption of the fracture network model may be used to create sufficient fracture volume to accommodate the total injected volume, through fracture dilation. Dilations within a fracture network may also induce shear movements on other fractures, such that the resulting fracture strains may be a combination of shear and tensile dilation. For the fracture model, displacements can be projected as either normal (i.e. tensile or dilatational opening) or parallel (i.e. shear) components relative to the fracture orientation.

FIGS. 38.1 and 38.2 illustrate modeled shear and tensile deformation modes in the fracture network shown in FIG. 37. FIGS. 39.1 and 39.2 show corresponding contours of density the cumulative strains of the hydraulic fracture network 3723 of FIG. 37 broken into shear and tensile components, respectively.

In FIGS. 38.1 and 38.2, each plot 3800.1 and 3800.2 are depicted along direction Y (north) (m) (y-axis) versus X (east) (m) (x-axis). FIG. 38.1 shows modeled shear deformations proportional to shear displacement with the arrows indicating regions with additional shearing T. FIG. 38.2 shows modeled tensile deformation with the arrows indicating regions with significant tensile σ dilation. The deformation is in some places has predominantly tensile segments 3723.1, and other shear segments 3723.2 (mostly shear).

The tensile and shear segments 3723.1, 3723.2 may be generated by breaking down rock fracture deformation of the hydraulic fracture network 3723. The depicted modeled shear deformations proportional to shear displacement (e.g., a maximum of about 0.02 m), and modeled tensile deformation (e.g., a maximum of about 0.03 m). The arrows of FIG. 38.1 indicate two regions with significant shearing. The arrows of FIG. 38.2 indicate two regions with significant dilation.

As shown in FIG. 38.2, the shear segments 3723.2, 3723.3 (the single planar fractures in the middle of the hydraulic fracture network 3723) dilates as a mostly tensile deformation mode. The maximum tensile deformation within the network 3723 is found to be approximately 3 cm, while the maximum shear deformation is approximately 2 cm.

Equations (20) and (21) may be used to convert shear and tensile components of the simulated hydraulic fracture network of FIGS. 38.1 and 38.2 to a simulated moment density in FIGS. 39.1 and 39.2. FIGS. 39.1 and 39.2 show corresponding contours of logs of density of total modeled shear and tensile cumulative deformations. FIG. 39.1 is a plot 3900.1 depicting contours of log of total modeled deformation for shear of FIG. 38.1. FIG. 39.2 is a plot 3900.2 depicting contours of log of total modeled deformation for tension of FIG. 38.2. FIGS. 39.1 and 39.2 each plot 3900.1 and 3900.2 are depicted along direction Y (north) (m) (y-axis) versus X (east) (m) (x-axis).

FIGS. 39.1 and 39.2 may provide another view of plots 3800.1 and 3800.2 of FIGS. 38.1 and 38.2 are simulated using a log of total modeled deformation for shear. Shear stresses and tensile stresses are depicted in FIGS. 39.1 and 39.2 by the stress τ arrows and the tensile σ arrows, respectively. Arrow M indicates an area with high shearing.

To compare with the observed microseismicity, a consistent 25 m grid was used to compute the seismic moment density of both the total modeled and observed deformations. The grid spacing was selected to match the average location uncertainty. In this example, the observed microseismic amplitudes are consistent with the shear radiation patterns of NE-SW or NW-SE strike slip displacements. More generally, seismic moment tensor inversion could also be used to estimate the mode of the microseismic deformation.

Based on shear microseismic slippage assumption, contours of the cumulative microseismic moments can be compared with corresponding modeled shear deformations. Beyond the observed microseismic deformations, aseismic deformations are also expected to contribute to the total expected deformation. The seismic efficiency, defined as the ratio of the radiated seismic energy to the total energy release is also a factor leading to the expectation that microseismic represents only a portion of the total strains. Assuming that these factors are constant through the fracture network, a relative comparison can be made with the microseismicity.

FIG. 40 is a plot 4000 depicting cumulative seismic moment from the observed microseismicity. The plot 4000 shows another view of the microseismic events 1223 depicted in FIG. 36 calculated based on a magnitude of the measured microseismic events of FIG. 36. The plot 4000 also shows contours of the area of high shear M′ in segment 3723.1. These contours may be approximately consistent with the modeled shear deformation (FIG. 38.1).

FIG. 40 may be used to provide the actual moment density taken from the microseismic events of FIG. 36. The contours of FIG. 40 are approximately consistent with the modeled shear deformation of FIG. 39.1. The modeled deformation depicted by plot 4000 may be more constant through the fracture network than shown of FIG. 38.1 where the observed microseismic moment Mo is largest near the treatment well (e.g., about treatment well 1204 of FIG. 12), namely at a relative planar fracture at a middle of the plot 4000 (shown at arrow in FIGS. 39.1 and 40). The model depicted by FIG. 40 indicates this fracture segment 3723.1 adjacent the arrow M is predominantly a shear deformation and segment 3723.2 refers to a tensile opening, which if true would imply a more effective fracture in segment 3723.1 despite the relatively weak microseismicity.

3. Seismic Moment

The modeled deformations can also be converted to a modeled (or an effective) seismic moment (Mo′) by multiplying the displacements by the shear modulus and area of each fracture segment (see, e.g., Equation (20)). Table 7 compares the total modeled moments (Mo) for the tensile (σ) and shear (τ) components and the observed microseismicity for each stage of the four stages of FIG. 36. As shown below, the modeled tensile σ component is larger than the modeled shear τ component (e.g., by about 50×) from the model, and the microseismicity Mo is about 0.1% of the modeled shear τ component.

TABLE 7
OBSERVED CUMULATIVE SEISMIC MOMENTS
AND EFFECTIVE MODELED RESULTS
	Stage 1	Stage 2	Stage 3	Stage 4
Observed Moment	1.29	0.24	0.16	0.62
(Mw) (GNm)
Modeled Shear	19,300	21,800	12,400	19,200
(τ) (GNm)
Modeled Tensile	36,400	35,000	34,400	33,800
(σ) (GNm)

The observed moment (Mw) may be determined from the modeled shear (τ) of FIG. 38.1 and the modeled tensile (σ) of FIG. 38.2 calculated using equation (21).

In this example, the modeled strains are predominantly tensile, consistent with the simple geometry scenarios. The model deformation is also larger than observed, again pointing to aseismic deformation. The observed deformation Mo is relatively large in stage 1 and low in stage 2 compared to, for example, stage 4. The modeled shear is highest for stage 2, suggesting too much complexity in the simulated fracture network for this stage. Stage 1 simulation is found to be more consistent in deformation with the other stages in contrast with the large observed deformation for this particular stage. Further investigation of the observed distributions of magnitudes indicates a localized region in a north-east part of the fracture segment 3723.1 accounting for almost half the seismic moment (at arrows M, M′ in FIGS. 39.1 and 40).

3. Calibration

A second model was run for stage 1, including an adjusted DFN to simulate this region of localized shearing in the observed data. FIGS. 41.1 and 41.2 show plots 4100.1 and 4100.1′, respectively, of a comparison of the shear displacements in the original stage 1 model and an updated model where the DFN was manually adjusted to match the localized shear deformation (arrows). FIG. 41.1 depicts a portion 41.1 of FIGS. 38.1, and provides a more detailed view of the stage 1 fracture portion 3723.1 with shear τ applied thereto as indicated by the arrow. FIG. 41.1 depicts a modeled shear strain τ associated with stage 1 for the original model.

FIG. 41.2 depicts a modified fracture segment 3723.1′ adjusted for DFN. The adjusted model of the modified fracture segment 3723.1′ resulting from an adjusted DFN. The modified fracture segment 3723.1′ as an effective shear moment increase of 46% over the original fracture segment 3723.1, with localized shearing similar to observed microseismic (e.g., microseismic events of FIG. 37). The modified fracture segment 3723.1′ may be generated using a process as depicted in FIGS. 46.1-46.4 as is described in more detail herein.

4. Predictions

The modifications depicted in FIG. 41.2 may be used to provide a microseismically calibrated fracture model as shown by the map 4200 in FIG. 42. The map 4200 is rotated and viewed from a different angle to depict the hydraulic fractures in greater detail. The microseismically calibrated fracture model may be used to provide a prediction of proppant distribution through the fracture network 3645.

FIG. 42 shows the map 4200 of the fracture width/proppant distribution for stage 1 of the case study. The map 4200 depicts propped regions 4255 and unpropped regions 4253 in the hydraulic fracture network 3739 depicted based on the adjusted model of the fracture segment 3723.1′ of FIG. 41.2.

This map 4200 also shows that the proppant is predicted to be concentrated close to the wellbore 1204, with relatively little of the total volume propped. Hydraulic conductivity can then be assigned based on permeability enhancement associated with shear displacements, in addition to proppant distribution. In this particular example, although the fracture network 3739 may be largely unpropped, the conductivity may still be enhanced through shearing and dilation from mismatched surface topographies.

The proppant map 4200 and corresponding relative conductivity can then be incorporated into a reservoir simulator to predict the well production and the associated reservoir drainage. History matching to the pressure decline may be used to estimate the hydraulic conductivity of the propped and unpropped regions 4255, 4253. A reservoir simulator can then be used to predict the well performance (e.g., production) and estimated reservoir drainage over time as shown in FIG. 43.

FIG. 43 is a plot 4300 depicting forecasted cumulative production from the well for the calibrated conductivity of 0.03 md-ft (0.91 md-cm), and also a sensitivity test for scenarios of more and less conductivity. The plot 4400 shows gas production volume (V) (MSCF) (y-axis) versus time (t) (yr) (x-axis) at un-propped conductivity levels of 4359, 4361, 4363 of 0.0003 md-ft (0.0091 md-cm), 0.03 md-ft (0.91 md-cm), and 1.0 md-ft (30.33 md-cm), respectively. The change in unpropped conductivity level results in a 90% increase from line 4361 to 4359, and a 40% decrease from line 4361 to 4363. The predicted reservoir drainage can be used to investigate well spacing requirements.

The reservoir pressure (P) may also be predicted for the fracture network 3723 of FIG. 38.1 based on the proppant placement of FIG. 42. FIG. 44 is a plot 4400 depicting a map view of reservoir pressure (P) about the fracture network 3723 simulated after 20 years of production of a wellbore 1204.

Fracture dilation is one factor for fracture effectiveness providing sufficient fracture volume to accommodate proppant placement, thereby ensuring continued fracture permeability after the stimulation. The model discussed herein may be used to honor a mass balance of the injected fluid, and can therefore be used to predict the proppant placement within the fracture network as shown in FIG. 42. The resulting proppant map can then be used to populate permeability within the fracture network for reservoir simulation of the well performance and reservoir drainage as shown in FIGS. 43 and 44, leading to an optimized estimate of effective stimulated volume and reservoir recovery.

Fracture Operation

In one aspect, the present disclosure describes methodologies for performing a microseismic fracture operation. These methods may involve the use of complex fracture models that can be used to investigate the extent and amount of deformation for comparison with observed microseismicity (e.g., microseismic events of FIG. 36). Improving the match of the appropriate mode of the fracture simulation with the microseismicity may provide confidence in the overall simulation result. In the example(s) presented herein, validating the shear deformation with the microseismicity implies that the dilatational deformation is valid regardless of whether the observed microseismicity directly represents tensile opening modes. A geomechanical simulation of the hydraulic fracture may be used to distinguish estimated deformation between shear and tensile modes of strain.

FIG. 45.1 provides a method 4500.1 of performing a fracture operation that may use either shear or moment densities to define fracture networks. The method 4500.1 involves 2350 performing a stimulation operation comprising stimulating the wellsite by injecting an injection fluid with proppant into the fracture network, 2352 generating wellsite data (e.g. natural fracture parameters of the natural fractures, pump data, and microseismic measurements), and 2375 modeling hydraulic fractures of the fracture network based on the wellsite data and defining a hydraulic fracture geometry of the hydraulic fractures as described with respect to FIG. 23.1.

In this version, after the modeling 2375, a decision 4551 may be made to continue with a shear failure operation 4553.1 and/or to perform a seismic moment operation 4553.2. At 4551, the method 4500.1 may continue by performing the shear failure analysis operation 4553.1. The shear failure analysis operation 4553.1 includes 2377 generating a stress field of the hydraulic fractures using a geomechanical model (e.g., 2D or 3D DDM), 2379 determining shear failure parameters comprising failure envelope and a stress state about the fracture network (e.g., along the natural fractures, hydraulic fractures, and/or rock medium), 2381 determining a location of shear failure of the fracture network from the failure envelope and the stress state, and 2383 calibrating the hydraulic fracture geometry by comparing the microseismic measurements with the simulated hydraulic fracture network and/or the activated discrete fracture network as performed in method 2300.2 of FIG. 23.2.

The method 4500.1 may also involve performing the seismic moment operation 4553.2. The performing an aseismic operation 4553.2 may involve 4559—determining actual and modeled seismic moment densities, and 4561—calibrating a DFN of the fracture network by adjusting the DFN based on a comparison of modeled and actual seismic moment densities.

The seismic moment operation 4553.2 may be performed to take into consideration the effects of deformation on a fracture network demonstrated by, for example, FIGS. 28-35.2. The seismic moment portion 4553.2 may be performed in addition to, or as a replacement for, the failure portion 4553.1. In cases where both the shear failure portion 2351.1 and the seismic moment portion 2351.2 are both performed, the results of each portion may be compared and/or analyzed. The shear failure and/or seismic moment portions 4553.1, 4553.2 may be repeated and/or compared.

The shear failure operation 4553.1 and the seismic moment operation 4553.2 may be performed simultaneously or in series. The results of the shear failure operation 4553.1 and the seismic moment operation 4553.2 may be compared, analyzed, and/or combined. Upon completion of the seismic operation 4553.1 and/or the seismic moment operation 4553.2, the adjusting 2385 and 2387 may be performed as previously described with respect to method 2300.2 of FIG. 23.2. The adjusting 2385 and/or 2387 may be performed based on the individual failure operation 4553.2, the individual seismic moment operation 4553.2, and/or a combination of the failure operation 4553.2 and the seismic moment operation 4553.2.

FIG. 46.2 shows a method 4500.2 of performing a seismic moment operation that may be used as the performing 4553.2 of FIG. 45.1. The method 4500.2 involves 4555—modeling a hydraulic fracture network (see, e.g., FIG. 37) based on the wellsite data (e.g., log data 2352), 4559—determining actual and modeled seismic moment densities, and 4561—calibrating the DFN 2375 based on a comparison of predicted moment density (FIG. 39.1) and actual moment density (FIG. 40). The 4559—determining actual and modeled seismic moment densities may involve 4557—defining shear and tensile components of the simulated hydraulic fracture network (see, e.g., FIGS. 38.1, 38.2 and equations (20), (21)), 4558—converting the shear and tensile components of the simulated hydraulic fracture to a simulated moment density, 4560—generating an actual moment density (see, e.g., FIG. 40) based on the wellsite data (e.g., microseismic events FIG. 37).

The modeling 4555 may be the same as the modeling 2375 and/or as shown in FIG. 37. The defining 4557, converting 4558, generating 4560, and calibrating 4561 may be repeated to further refine the DFN. The method 4500.2 may also involve 4567—predicting proppant placement (FIG. 42), 4568—predicting production (FIG. 43), and/or 4569—predicting reservoir pressure for the fracture network (FIG. 44).

Part or all of the methods herein may be combined, performed in any order, and/or repeated as desired.

Calibration

In another aspect, the present disclosure relates to a method for using microseismic data to calibrate the initial discrete natural fracture network (DFN). The calibrated DFN model may be utilized as input for the complex hydraulic fracture network (HFN) model to simulate the fracture propagation during a fracture treatment. The calibrated DFN provides an accurate description of the reservoir and consequently a more accurate prediction of the created fracture geometry by the HFN simulator.

Detailed complex hydraulic fracture models predict the progressive propagation of multiple fracture branches in a fracture network. The formation initially may include of many natural fractures. The interaction of the hydraulic fracture and the natural fracture may result in fracture branching where they intersect.

Referring back to FIG. 21, different scenarios when a hydraulic fracture intersects a natural fracture are depicted. The scenarios that result in hydraulic fracture opening up and propagating along the natural fracture lead to fracture branching and complexity. FIG. 21 depicts a schematic of some of the possible outcomes when a hydraulic fracture intersects a natural fracture. Since an ideal characterization of natural fractures underground is not possible, the initial population of natural fractures is stochastically created, constrained by the information obtained from seismic data and borehole imaging measurements, utilizing geological and geostatistical models.

FIGS. 46.1-46.4 depict plots 4600.1-4600.4 of stages of simulation of fracture network 4647 about the wellbore 1204. FIG. 46.1 shows the top view of a statistically created DFN 4647 having traces uniformly distributed in the formation. FIG. 46.2 shows a predicted HFN 4661 generated from the complex fracture model for the corresponding DFN, along with the microseismic events 4663 collected during the fracture treatment. FIG. 46.1 shows traces of statistically generated DFN near the horizontal well 1204. FIG. 46.2 shows the simulated hydraulic fracture network 4661 generated from the uniformly distributed DFN 4647. In this case, the microseismic data shows distinctive clustering of the microseismic events. In comparison to the simulation results, a large area is present in between the event clusters where a lot of fracture surface areas are created according to the model, and with little microseismic activity.

Since the microseismic events correlate to the shear slippage of the natural fractures in the formation, induced by rock deformation and fluid flow into the formation surrounding the hydraulic fractures, the clustering of the microseismic events may be an indication of strong clustering of the natural fractures. In this case, the area between the clusters may be absent of many natural fractures and the model predicts incorrect fracture geometry due to the incorrect assumption of the initial natural fracture distribution.

In one aspect of the present disclosure, it is suggested that a more representative DFN can be generated by correcting the original DFN model utilizing the microseismic measurement. This calibration can be carried out by redistributing the natural fractures in proportion to the spatial distribution of the micro seismic events density, or using the moment density method described in FIG. 45.2.

FIG. 46.3 is a plot 4600.3 showing a calibrated DFN 4647′ with heterogeneous distribution of the natural fractures based on the microseismic measurements. The corresponding simulation of the hydraulic fracture 4661′ geometry is shown in plot 4600.4 of FIG. 46.4. FIG. 46.4 shows simulated hydraulic fractures 4661′ for the calibrated DFN. The results from the calibrated DFN should provide a description of the hydraulic fracture geometry with enhanced accuracy, and ultimate production performance of the well 1204.

FIG. 47 shows a method 4700 of calibrating a DFN. The method 4700 may be used, for example, to optimize complex fracture design utilizing microseismic measurements to calibrate the natural fracture distribution. The method 4500 involves (4571) generating the initial natural fracture distribution (DFN model) with their characteristics derived from wellsite data, such as seismic measurement, geological structure, borehole imaging log and core based description and measurements; (4573) generating initial hydraulic fracture design and carrying out simulation using a complex fracture model that incorporates the interaction of hydraulic fractures and natural fractures; (4575) pumping the fracturing treatment and collecting microseismic data in real-time; (4577) calibrating the initial DFN and redistributing the natural fractures according to the observed microseismic event distribution; (4579) calibrating additional natural fracture and formation parameters using the calibrated DFN distribution to match the predicted hydraulic fracture network coverage area against the overall microseismic area and the simulated treatment pressure against the measured pressure; and, (4581) revising the fracture design based on the calibrated model to optimize the next treatment stage in the same well or the next well in the same area.

The method of FIG. 47 may be used as the calibrating 4561 of FIG. 45.2 by replacing observed microseismic event distribution of 4577 with the observed seismic moment density.

Although the present disclosure has been described with reference to example embodiments and implementations thereof, the present disclosure is not to be limited by or to such embodiments and/or implementations. Rather, the systems and methods of the present disclosure are susceptible to various modifications, variations and/or enhancements without departing from the spirit or scope of the present disclosure. Accordingly, the present disclosure expressly encompasses all such modifications, variations and enhancements within its scope.

It should be noted that in the development of any such actual embodiment, or numerous implementation, specific decisions may be made to achieve the developer's specific goals, such as compliance with system related and business related constraints, which will vary from one implementation to another. Moreover, it will be appreciated that such a development effort might be complex and time consuming, yet may be a routine undertaking for those of ordinary skill in the art having the benefit of this disclosure. In addition, the embodiments used/disclosed herein can also include some components other than those cited.

In the description, each numerical value should be read once as modified by the term “about” (unless already expressly so modified), and then read again as not so modified unless otherwise indicated in context. Also, in the description, it should be understood that any range listed or described as being useful, suitable, or the like, is intended that any and every value within the range, including the end points, is to be considered as having been stated. For example, “a range of from 1 to 10” is to be read as indicating each and every possible number along the continuum between about 1 and about 10. Thus, even if specific data points within the range, or even no data points within the range, are explicitly identified or refer to a few specific ones, it is to be understood that inventors appreciate and understand that any and all data points within the range are to be considered to have been specified, and that inventors possessed knowledge of the entire range and all points within the range.

The statements made herein merely provide information related to the present disclosure and may not constitute prior art, and may describe some embodiments illustrating the invention. All references cited herein are incorporated by reference into the current application in their entirety.

The discussion herein is directed to certain specific implementations. It is to be understood that the discussion below is for the purpose of enabling a person with ordinary skill in the art to make and use any subject matter defined now or later by the patent “claims” found in any issued patent herein.

It should be understood that the various technologies described herein may be implemented in connection with hardware, software or a combination of both. Thus, various technologies, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium wherein, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the various technologies. In the case of program code execution on programmable computers, the computing device may include a processor, a storage medium readable by the processor (including volatile and non-volatile memory and/or storage elements), at least one input device, and at least one output device. One or more programs that may implement or utilize the various technologies described herein may use an application programming interface (API), reusable controls, and the like. Such programs may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) may be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language, and combined with hardware implementations.

While the foregoing is directed to implementations of various technologies described herein, other and further implementations may be devised without departing from the basic scope thereof, which may be determined by the claims that follow. Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims may not be limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Although a few example embodiments have been described in detail above, those skilled in the art will readily appreciate that many modifications are possible in the example embodiments without materially departing from the system and method for performing wellbore stimulation operations. Accordingly, all such modifications are intended to be included within the scope of this disclosure as defined in the following claims. In the claims, means-plus-function clauses are intended to cover the structures described herein as performing the recited function and not just structural equivalents, but also equivalent structures. Thus, although a nail and a screw may not be structural equivalents in that a nail employs a cylindrical surface to secure wooden parts together, whereas a screw employs a helical surface, in the environment of fastening wooden parts, a nail and a screw may be equivalent structures. It is the express intention of the applicant not to invoke 35 U.S.C. § 112, paragraph 6 for any limitations of any of the claims herein, except for those in which the claim expressly uses the words ‘means for’ together with an associated function.

Claims

1. A method of performing a fracture operation at a wellsite, the wellsite positioned about a subterranean formation having a wellbore (1204) therethrough and a complex fracture network therein, the fracture network comprising natural fractures, the wellsite stimulated by injection of an injection fluid with proppant into the fracture network, the method comprising: generating wellsite data (2352) comprising measurements of microseismic events of the subterranean formation;modeling (2375) a hydraulic fracture network and a discrete fracture network of the complex fracture network based on the wellsite data;characterized in that the method further comprises:performing (4553.2) a seismic moment operation, comprising: determining (4559) an actual seismic moment density based on the wellsite data and a predicted seismic moment density based on shear and tensile components of the simulated hydraulic fracture network, wherein determining the predicted seismic moment density comprises defining (4557) the shear and tensile components of the simulated hydraulic fracture network and converting (4558) the shear and tensile components of the simulated hydraulic fracture network to a simulated moment density; andcalibrating (4561) the discrete fracture network based on a comparison of the predicted moment density and the actual moment density; andadjusting (2387) the injection based on the calibrating.

2. The method of claim 1, further comprising predicting (4567) proppant placement based on the modeled discrete fracture network.

3. The method of claim 2, further comprising predicting (4568) production based on the predicted proppant placement.

4. The method of claim 2, further comprising predicting (4569) reservoir pressure based on the predicted proppant placement.

5. The method of claim 1, wherein modeling a discrete fracture network comprises generating an initial discrete fracture network from at least one of wellsite data comprising seismic measurement, geological structure, borehole imaging log, core, and combinations thereof based on description measurement.

6. The method of claim 1, wherein modeling hydraulic fracture network comprises generating an initial hydraulic fracture design and carrying out simulation using a complex fracture model that incorporates the interaction of hydraulic fractures and natural fractures.

7. The method of claim 1, wherein generating wellsite data comprises pumping fracturing treatment into a wellbore of the wellsite and collecting microseismic data in real-time.

8. The method of claim 1, wherein calibrating comprises: calibrating (4777) the discrete fracture network and redistributing the natural fractures according to an observed microseismic event distribution over an event area;calibrating (4779) additional natural fracture and formation parameters using the calibrated discrete fracture network to match a coverage area of the modeled hydraulic fracture network against the event area and the simulated treatment pressure against a measured pressure; andoptimizing (4781) the injection at the wellsite by revising the modeled hydraulic fracture based on the calibrated discrete fracture network.

9. The method of claim 1, further comprising performing a shear failure operation comprising: determining (2377) a stress field of the hydraulic fractures using a geomechanical model;determining (2379) shear failure parameters comprising a failure envelope and a stress state about the fracture network;determining (2381) a location of shear failure of the fracture network from the failure envelope and the stress state; andcomparing (2383) the modeled hydraulic fractures and the locations of shear failure against the measured microseismic events.

10. The method of claim 1, further comprising stimulating (2350) the wellsite by injecting the injection fluid with proppant into the fracture network.

11. The method of claim 10, wherein adjusting comprises adjusting (2387) the stimulation operation based on the calibrating.