US20130319657A1

US20130319657A1 - Method for controlling the trajectory of a hydraulic fracture in strata-containing natural fractures

Info

Publication number: US20130319657A1
Application number: US13/519,572
Authority: US
Inventors: Eduard Siebrits; Dmitry Chuprakov
Original assignee: Schlumberger Technology Corp
Current assignee: Schlumberger Technology Corp
Priority date: 2009-12-30
Filing date: 2009-12-30
Publication date: 2013-12-05
Also published as: MX2012007687A; RU2505670C1; CA2785793A1; WO2011081544A1

Abstract

The method of controlling the parameters of a hydraulic fracture comprises creating a matrix of relationship between initial formation, injection and fracture parameters of and a predicted increment of a hydraulic fracture path. The matrix is used for retrieving and deriving the predicted increment of the fracture path, depending on the actual initial parameters of a fracture being created. The actual increment of the hydraulic fracture path is measured and compared with the predicted increment of the fracture path. In case of a discrepancy between the actual and predicted increments, the actual initial parameters of the fracture are changed.

Description

FIELD OF THE DISCLOSURE

This invention relates to methods for controlling and optimizing parameters of a hydraulic fracture created during the hydraulic fracturing of oil-bearing and gas-bearing reservoirs containing an existing network of natural (geological) fractures, and can be used in appropriate oil and gas fields.

BACKGROUND OF THE DISCLOSURE

Hydraulic fracturing is a widely used method for stimulating hydrocarbon inflow from a formation into oil wells or gas wells. To ensure that the best economic result is achieved from the hydraulic fracturing treatment, a design model of the treatment is developed. This model is based on mechanical characteristics of the formation (such as formation stresses, moduli of elasticity and plasticity of the formation, cracking resistance, permeability, etc.), as well as on the selection of optimum parameters for injection of a fracturing fluid into the formation, including the selection of a proper fracturing fluid, proppants, injection conditions, etc. The design model of the fracture plays a rather important role which consists in ensuring that the actual parameters of the fracture geometry will be consistent with the predicted values, and that the selected fracturing fluid and proppant, as well as their quantities, injection rates and the proppant schedule are acceptable for successful implementation of the hydraulic fracturing process.
Most of the design models offered nowadays for commercial use (STIMPLAN, NSI Technologies; FracProPT, MFRAC, etc.), as well as the data contained, for example, in Thiercelin, M. J. 2009, “Hydraulic fracture propagation in discontinuous media,” Proc.: International Conference on Rock Joints and Jointed Rock masses, Tucson, 4-10 Jan 2009. pp. 12; Daneshy, A., 2003, “Off-balance growth: A new concept in hydraulic fracturing,” Journal of Petroleum Technology, 55, 4, April 2003: 78-85, etc. are based on the assumption that a single hydraulic fracture plane is created in the formation under treatment. The fracture initiates from the wellbore and increases in length and in height over time as the fracturing fluid and proppant are injected. The in-situ stress condition in the reservoir is such that that there is generally a minimum stress in one of the three stress components, and the created hydraulic fracture tends to propagate in the plane normal to the minimum stress. This assumption about a single planar fracture is usually acceptable for the hydraulic fracturing treatment in a formation consisting of horizontally homogeneous layers.
However, the number of hydraulic fracturing treatments in unconventional oil-bearing formations has recently begun to increase. For complex reservoirs (e.g., for shale-gas reservoirs containing a network of natural (geological) fractures), the assumption about the planar geometry of the hydraulic fracture becomes unjustified. In such formations, the fracturing fluid penetrates so-called “connection branches,” thus creating a complex network of crisscross fractures.
It has become obvious that conventional tools intended for development of optimum hydraulic fracturing treatment strategies and based on planar distribution do not produce the required effect when used in the producing layers having a complex three-dimensional configuration. In such cases, new development tools are required for determination of optimum hydraulic fracturing treatment strategies.
For example, based on the purpose which consists in conducting a hydraulic fracturing process in formations containing natural (geological) fractures, the closest analogue of the claimed invention is the method of fracturing a naturally fractured rock (WO 2008093264, E 21 B 43/26, published on Jul. 8, 2008). The method is used for conducting a hydraulic fracturing process in formations containing geological fractures. The method comprises: a) acquiring subterranean formation layer geomechanical properties, well completion and reservoir data for the subterranean formation, and a natural fracture network description for the subterranean formation; b) simulating a fracture treatment for the formation, the simulation comprising inputting data acquired into a model which simulates propagation of a network of fracture branches by dividing fracture segments into a plurality of elements to form a fracture grid; c) determining and preparing an optimum fracture fluid composition to achieve the fracturing objective; and, d) injecting the fracturing fluid into a wellbore at a pressure sufficient to fracture the subterranean formation.
However, the known method has disadvantages consisting in the fact that the proposed method is characterized by a great calculation resource intensity, which increases the capital investments required for implementation of this method. In addition, the method does not include any design model correction to be made by comparing the calculated measurements and the actually obtained measurements, which increases the experimental error and results in a low accuracy of the method on the whole.

SUMMARY OF THE DISCLOSURE

The claimed invention provides higher efficiency and accuracy of the hydraulic fracturing process control (namely, of the hydraulic fracture propagation path control in formations containing natural (geological) fractures), as well as reduced capital investments required for implementation of this invention.
The method comprises the following operations:

a) creating a matrix of relationship between initial formation, injection and fracture parameters and a predicted increment of a hydraulic fracture path;
b) starting a hydraulic fracturing process;
c) measuring actual initial parameters of a hydraulic fracture being created;
d) using the matrix for retrieving and deriving the predicted increment of the fracture path, depending on the actual initial parameters of the fracture being created;
e) measuring an actual increment of a path of the hydraulic fracture being created;
f) comparing the actual increment of path of the fracture being created with the predicted increment of the fracture path; and
g) in case of a discrepancy between the actual and predicted increments, changing the actual initial parameters of the fracture being created.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 illustrates a schematic representation of main parameters in case where a hydraulic fracture approaches an interface.

FIG. 2 is a schematic representation of possible cases of the hydraulic fracture propagation in case where a hydraulic fracture interacts with an interface.

FIG. 3 is a schematic representation of three possible base cases of the hydraulic fracture propagation in case where a hydraulic fracture interacts with a natural fracture.

FIG. 4 shows a fragment of a matrix of relationship between normalized parameters, and characteristics of the offset of the hydraulic fracture at the interface.

DETAILED DESCRIPTION

There are many factors that control the complexity of a hydraulic fracture geometry. In cases where the hydraulic fracture interacts with a bedding layer, a natural fracture, or a fractional fault (all of them being hereinafter referred to as the “interface”), the hydraulic fracture may expand right through the interface, may end at the interface, or may continue to propagate at a certain offset along the interface (FIG. 3).
The key parameters which control the selection of one of the possible fracture propagation paths listed above include, but are not limited to, the following: a fracturing fluid injection rate (Q); a fracturing fluid viscosity (μ); a remote stress difference (σ1-σ3); interface properties, such as a friction (σ), an adhesion (C), an angle of approach (β) of the hydraulic fracture to the interface, and a gap (d) between the nearest hydraulic fracture tip and the interface (FIG. 1).
FIG. 2 shows possible hydraulic fracture propagation paths in case where a hydraulic fracture intersects an interface. FIG. 1 and FIG. 2 show the interface as a thin horizontal line which cuts each drawing in half; “Mat 1” and “Mat 2” shows potentially different materials on each side of the interface, and the fracture approaches it from the bottom of the drawing. For schematic representation of a possible hydraulic fracture propagation path, these drawings are based on the assumption that the fracture has a planar geometry although the process can be expanded to a three-dimensional model.
By varying independently each of these control parameters through a series of physical experiments or through a numerical simulation, it is possible to determine the conditions under which different scenarios (cases) of the hydraulic fracture intersecting the interface (i.e., the behavior and the direction of the hydraulic fracture), shown in FIG. 2, are most likely to happen. This may require a great number of scenarios (series of experiments), depending on the required resolution. But with a numerical simulation made by using state-of-the-art computers, this is quite feasible and industrially applicable.
The obtained results can be summarized as a matrix (a database, a reference table, or other data files) containing key data which determine whether the hydraulic fracture will tend to be more planar during the fracturing fluid injection in one extreme case, or whether it will tend to branch out into many fractures covering a great reservoir volume in the other extreme case, or whether the propagating hydraulic fracture will stop growing at the interface without further propagation. This matrix can be developed through numerical calculations and/or physical experiments.
In addition, it is also necessary to obtain information on geometrical complexity of the hydraulic fracture propagation through a network of existing natural fractures. This information can be based on the result of the interaction between the hydraulic fracture and existing natural fractures at the interface. The known result of this interaction with a natural fracture can be superimposed to give the resulting path of the hydraulic fracture propagation through a sequence of natural fractures in a fractured formation. From now on, the hydraulic fracture path change which considers the hydraulic fracture characteristics (such as the behavior, the geometry and the direction of the fracture) will be regarded as the fracture “increment.”
So, the first step of the proposed method is to find a solution to the problem of the interaction between a hydraulic fracture and an existing natural fracture and to keep this solution for further use.
Since the result of the fracture propagation through the interface depends on such parameters as: 1) formation and natural fracture parameters, 2) fracturing fluid injection parameters, 3) initial fracture path parameters, we consider that, for the purposes of the claimed invention, the above-mentioned parameters form the “initial” parameters of formation, injection and fracture.
So, the “initial” parameters include, but are not limited to, the following:

- formation and natural fracture parameters (properties): formation mechanical stresses in this point, coefficients of interface adhesion and coefficients of interface friction, a relative angle of the formation and natural fracture slope relative to the hydraulic fracture, location and size of natural fractures,
- fracturing fluid injection parameters (properties): a viscosity of a fracturing fluid to be injected and the fracturing fluid injection rate, or an average fracturing fluid pressure in the hydraulic fracture,
- initial fracture path (geometry) parameters (properties): a fracture length, a gap (if any) between a hydraulic fracture tip and an interface, etc.

The results obtained for each parameter under investigation can be kept in a relevant data file (a reference table, a database or a spreadsheet), i.e., in a matrix of relationship, where a relevant predicted path (that is to say, a hydraulic fracture propagation increment) has been derived and kept for each specific set of initial parameters obtained in the course of numerical calculations or in the course of a physical experiment.
So, the matrix of relationship shall be developed for an independent set of parameters in order to reduce excessive data volume, database creation time and final retrieval time. After the set of parameters has been determined, it is necessary to specify the optimum resolution of parameter values for each independent parameter.
A hydraulic fracturing process is then implemented and actual measurements of initial parameters of a fracture being created (including formation and natural fracture parameters, fracturing fluid injection parameters, initial fracture path parameters) are taken.
For example, a number of hydraulic fracture and formation parameters are measured by using measuring instruments (e.g., sensors distributed over relevant zones of the formation, seismic and acoustic measuring instruments, tiltmeters, etc.). For example, the above-mentioned parameters are determined in the real-time mode, which allows to monitor continuously the hydraulic fracturing treatment and to increase the hydraulic fracturing process control efficiency.
To be able to use the said matrix of relationship during the implementation of the claimed method, it is necessary to use a programming tool capable of performing parameter change operations and logical data processing operations.
The design tool known from WO 2008093264, E 21 B 43/26 published on Jul. 8, 2008, can be used as such tool. Such a combined tool can be used for developing, monitoring and controlling the hydraulic fracturing treatment in the real-time mode, as well as for evaluating the results of the hydraulic fracturing treatment in a relevant formation containing natural fractures or faults.
The design tool shall be capable of accessing the matrix of relationship to determine the location of fracture re-initiation or different fracture propagation path (increment).
So, first of all, when conducting the hydraulic fracturing, it is necessary to provide a design tool with actual initial parameters which characterize a formation and natural fractures, a fracturing fluid injection and an initial hydraulic fracture path, and which are measured by using sensors or by taking dedicated measurements, as described above. The above-mentioned parameters measured during the hydraulic fracturing process shall also be provided in the real-time mode to the design tool.
These actually measured parameters shall be then converted into independent parameters (normalized values) because the parameters in the matrix of relationship are represented as normalized (non-dimensional) values. The said design tool is therefore capable of converting the parameters actually measured during the hydraulic fracturing process into normalized values (and vice versa).
In addition, the design tool is capable of accessing the matrix of relationship to retrieve a direct scenario (path) or a scenario (path) obtained by interpolation between two, or more than two, scenarios which represent the predicted increment of the hydraulic fracture propagation path, depending on the actually measured initial parameters of the hydraulic fracture being created.
The design tool then determines from the matrix of relationship the resulting fracture propagation path (the predicted fracture increment) along with relevant fracture characteristics (parameters), such as the offset of the fracture re-initiation along the interface on the opposite side of the fracture. And the normalized parameters are then converted into units of measurable physical quantities (for example, the actual distance of the offset).
While the fracture grows during the hydraulic fracturing process, the actual increment of the resulting fracture path is measured by using sensors or other measurement methods which allow to determine the required characteristics of the fracture. The said data are provided to the design tool.
The actual increment of the resulting fracture path is compared with the predicted increment of the fracture path. This comparison operation is performed by a control device.
In fact, the actually measured data on the hydraulic fracture propagation may differ from the results of the calculations made by the control device. To minimize prediction errors and to improve the final result of the hydraulic fracturing treatment, it is possible to use any discrepancies between the predicted data and the measured data to change the actual initial parameters of the resulting fracture in the near-real-time mode (which certainly allows you to optimize the entire method as a whole).
In such an operation, certain initial parameters are corrected (changed) during the hydraulic fracturing process. For example, it is possible to change the geometry or the pressure, the injection rate and the viscosity of the fracturing fluid in the near-real-time mode, considering the error resulting from the comparison operation.
Thus, we improve the consistency between the predicted parameters and the measured parameters during the implementation of the method, which, in turn (in combination with the essential features listed above), results in a higher efficiency of the hydraulic fracture path control.
So, first, it is necessary to create a matrix of relationship between initial parameters characterizing a formation and natural fractures, a fracturing fluid injection and an initial fracture path before the interaction with a natural fracture, on the one hand, and a predicted increment of a hydraulic fracture path after the interaction with the natural fracture, on the other hand. This matrix can be developed, for instance, by finding a numerical solution to the problem of the mechanical interaction between the hydraulic fracture at a constant internal pressure and the natural fracture at the moment of their contact, and can be represented in the form of a table as shown in FIG. 4.
To reduce the number of independent parameters of the problem, it is necessary to normalize all values in the numerical problem, so the solution represents the function of the coefficient of friction on the natural fracture λ, the angle of slope of the natural fracture β, the non-dimensional formation differential stress Δ=(σ₁−σ₃)/(σ₁+σ₃) and the non-dimensional hydraulic fracture excess pressure Π=2(p−σ₃)/(σ+σ₃). Considering that, having come across the natural fracture, the hydraulic fracture will continue to propagate if a new tensile crack initiates on the other side of the natural fracture, the offset of the tensile stress peak along the natural fracture and the stress peak itself are then selected as the characteristics of the hydraulic fracture path increment at the natural fracture. As a result of the normalization, the offset will be normalized to the hydraulic fracture length L_HFand the stress will be normalized to the average formation stress σ_m=(σ₁+σ₃)/2.
Then, hydraulic fracturing process is started and actual initial formation, injection and hydraulic fracture parameters are measured. At a certain stage, it is necessary to learn how a hydraulic fracture path will change after a hydraulic fracture has interacted with a natural fracture crossing its path. The measured formation stresses are equal to σ₁=6 MPa, σ₃=4 MPa, a hydraulic fracture pressure is estimated as p=5.5 MPa, and a current hydraulic fracture length is equal to L_HF=100 m. The natural fracture has a friction coefficient of 0.5. Let us then assume two possible orientations of the natural fracture with respect to the hydraulic fracture: a) 10° and b) 40°.
By using the normalization rules from the table, we obtain: λ=0.5, Δ=0.2, Π=0.3, and the angles for the first case (a): β=10°, for the second case (b): β=40°.
The matrix of relationship is then used for retrieving and deriving a predicted increment of the fracture path, in particular, an offset value and a stress peak value, depending on the given parameters. Based on the above-mentioned parameters, we retrieve the values of the hydraulic fracture path increment characteristics to find that the offset of the fracture is equal to 0.049 and the stress is equal to −0.0163 in the first case. In the second case, the offset of the fracture is equal to 0.025 and the stress is equal to −0.7929 (the relevant lines of the table shown in FIG. 4 are highlighted).
The normalized values are then converted into dimensional units of relevant characteristics. We obtain for case a): offset=4.9 m and stress=0.08 MPa, for case b): offset=2.5 m and stress=3.96 MPa.
Then, based on the comparison of the tensile stress peak with the tensile strength of the rock, a conclusion is made as to whether a secondary fracture will or will not be formed at the natural fracture. Assuming that the tensile strength is equal to 4 MPa, the hydraulic fracture will intersect the natural fracture at a distance of 2.5 m in case a). In case b), no intersection will take place and the hydraulic fracture will stop at this natural fracture.
Let us then assume that the hydraulic fracture path increment characteristics (namely, the intersection and the offset of the fracture in case a)) have been measured, and these measurements show that the offset is actually equal to 1.3 m.
The actual increment of the resulting fracture path is then compared with the predicted increment of the fracture path. As these values turned out to be different, the actual initial parameters of the resulting fracture shall be changed. In this case, the fluid pressure value shall be corrected. Having retrieved the required data from the matrix, we find that this value corresponds to a non-dimensional pressure of 0.2 for the given hydraulic fracture length. Conversion into dimensional units gives a new well pressure value of 5 MPa which will be used as a more accurate value in subsequent applications of the method.
The above embodiments of the invention are made for illustrative purposes only because the invention can be modified and can be practically used in a variety of similar ways which are obvious to persons who are skilled in this art and who can make use of the advantages offered by the concepts described in this application. Moreover, the applicant had no intention to impose any limitations on the details of the designs or developments shown herein, except for those described in the Claims below. It is therefore obvious that the specific embodiments of the invention, disclosed above, can be changed or modified, and all such modifications shall be considered to be protected in accordance with the scope and the nature of the invention.

Claims

1. Method for controlling a hydraulic fracture path in formations containing natural fractures comprising:

a) creating a matrix of relationship between initial formation, injection and fracture parameters and a predicted increment of a hydraulic fracture path,

b) starting a hydraulic fracturing process,

c) measuring actual initial parameters of a hydraulic fracture being created,

d) using the matrix for retrieving and deriving the predicted increment of the fracture path, depending on the actual initial parameters of the fracture being created,

e) measuring an actual increment of a path of the hydraulic fracture being created ,

f) comparing the actual increment of the path of the fracture being created with the predicted increment of the fracture path, and

g) in case of a discrepancy between the actual and the predicted increments changing the actual initial parameters of the fracture being created.

2. Method of claim 1, wherein the matrix of relationship between initial formation, injection and fracture parameters and the predicted increment of the hydraulic fracture path is created through numerical calculations or experiments or both.

3. Method of claim 1, wherein the initial formation, injection and fracture parameters include formation and natural fracture parameters, fracturing fluid injection parameters and initial fracture path parameters.

4. Method of claim 3, wherein formation mechanical stresses are used as the formation and natural fracture parameters.

5. Method of claim 3, wherein coefficients of interface adhesion are used as the formation and natural fracture parameters.

6. Method of claim 3, wherein coefficients of interface friction are used as the formation and natural fracture parameters.

7. Method of claim 3, wherein a relative angle between the hydraulic fracture and a natural fracture in a point of their contact is used as the formation and natural fracture parameters.

8. Method of claim 3, wherein a parameter characterizing the location of natural fractures is used as the formation and natural fracture parameters.

9. Method of claim 3, wherein a natural fracture size parameter is used as the formation and natural fracture parameters.

10. Method of claim 3, wherein a viscosity of a fracturing fluid to be injected is used as the fracturing fluid injection parameter.

11. Method of claim 3, wherein a fracturing fluid injection rate is used as the fracturing fluid injection parameter.

12. Method of claim 3, wherein an average fracturing fluid pressure in the hydraulic fracture is used as the fracturing fluid injection parameter.

13. Method of claim 3, wherein a fracture length is used as the initial fracture path parameter.

14. Method of claim 3, wherein a gap between a hydraulic fracture tip and an interface is used as the initial fracture path parameter.

15. Method of claim 1, wherein the initial formation, injection and fracture parameters and the predicted increment of the hydraulic fracture path are represented in the matrix as normalized values.

16. Method of claim 1, wherein operations c), d), e), f), g) are performed in the real-time mode of the hydraulic fracturing process.

17. Method of claim 1, wherein operations c), d), e), f), g) are performed by using a control device.

18. Method of claim 17, wherein the control device is capable of converting the actually measured values into normalized values, and vice versa.

19. Method of claim 1, wherein the actual initial parameters of the hydraulic fracture being created are measured by using measuring techniques.

20. Method of claim 1, wherein the actual increment of the hydraulic fracture being created is measured by using measuring techniques.