A real-time imaging method of artificial fracture network

Abstract

The application discloses a real-time imaging method of artificial fracture network, and relates to the technical field of fracture network imaging, and comprises the following steps: obtaining information of natural fractures, ground stress orientation and microseismic event position of a research area; when a new fracture is generated, generating and optimizing a new fracture network parameter combination randomly through a Monte Carlo method; determining a fracture to which each microseismic event belongs, projecting to determine the range thereof, and completing fracture network imaging in this stage. The real-time imaging method of artificial fracture network realizes real-time imaging of the fracture network based on microseismic events by obtaining information such as the microseismic event position of the corresponding area, can determine the size of the fracture network, directly depicts the reservoir fracture, and the imaging result of the fracture network is more complete and accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fracture network imaging, and particularly relates to a real-time imaging method of artificial fracture network. BACKGROUND

[0002] Deep underground energy represented by hot dry rock is expected to become a new energy source in the future due to its abundant reserves, renewability and environmental friendliness. However, the deep reservoir usually has very little fluid content and the formation rock permeability is very low, resulting in very low natural production. Therefore, the reservoir is usually subjected to hydraulic fracturing modification to generate a fracture network with strong permeability. The rock rupture will cause microseismic events with different strengths, so it is necessary to continuously adjust the fracturing scheme to avoid the occurrence of microseismic events, and the potential fracture network extension needs to be described. Since the hot dry rock reservoir is buried at a great depth, the monitoring data based on the microseismic events is a relatively effective means.

[0003] The existing fracture evaluation methods based on microseismic data can be roughly divided into two categories: 1) only relying on microseismic event position fitting to generate a fracture network; and 2) considering not only the microseismic event position but also the occurrence time, magnitude, focal mechanism and moment tensor attributes. However, the above methods cannot describe the process of the fracture network extending with the real-time extension of the microseismic events, and for the deep reservoir, the focal mechanism data often has large errors, so the evaluation results are not accurate enough using the above methods. SUMMARY

[0004] The present application aims to provide a real-time imaging method of artificial fracture network to solve the problems in the above background.

[0005] To achieve the above object, the present application provides the following technical scheme:

[0006] A real-time imaging method of artificial fracture network, comprising the following steps:

[0007] 1) obtaining information of natural fractures, in-situ stress orientation and microseismic event position of a study area, and randomly generating and optimizing new fracture network parameter combinations through a Monte Carlo method when new fractures are generated;

[0008] 2) determining the fracture to which each microseismic event belongs, projecting to determine the range thereof, and completing the fracture network imaging in this stage;

[0009] 3) adding new microseismic event groups, combining with the existing microseismic event groups, and repeating steps 1) and 2).

[0010] On the basis of the above technical scheme, the present application further provides the following optional technical scheme:

[0011] In an alternative, in step 1), microseismic location and fracturing operation data of the study area are obtained, microseismic events are divided into stages, and the microseismic event cloud of each stage is de-duplicated and DBSCAN denoised.

[0012] In an alternative, in step 1), a number of fractures is given and a combination of fracture network parameters is randomly generated using the Monte Carlo method, and a maximized objective function D is defined. targ As one of the preferred bases:

[0013]

[0014] where n f is the total number of fractures, p is the number of microseismic events fitted by the fractures at a given distance threshold, d i，j is the distance from the jth microseismic event to the ith fracture, and if it is less than the given threshold, the jth microseismic event is considered to be contained by the ith fracture.

[0015] In an alternative, another basis for optimizing the fracture network is the fitting rate, which includes the following steps:

[0016] 1) Calculate the distance between the existing fractures and the microseismic events, and exclude the fitted microseismic event points;

[0017] 2) When the remaining microseismic event points are greater than 5% of the total, select a fracture occurrence interval and a microseismic event point to define a fracture;

[0018] 3) Calculate the distance between the fracture and each microseismic event, and exclude the fitted microseismic event points;

[0019] 4) Repeat steps 2) and 3) until the set number of fractures is reached or the remaining microseismic event points are less than 5% of the total;

[0020] 5) Calculate two objective functions;

[0021] 6) Repeat steps 4) and 5) to obtain the optimal fracture network.

[0022] In an alternative, in step 2), in combination with the determined fracture parameters, the newly added fitted microseismic events are optimally assigned to the corresponding fracture network using the DBSCAN method. After the assignment is completed, the smaller cluster event points assigned to each fracture are removed, and the remaining corresponding new microseismic events are projected to the corresponding fracture, combined with the existing microseismic event projection points, and screened to delineate the fracture range.

[0023] In an alternative, the fractures generated by the fracture network imaging are parallelograms, and the range thereof adopts an equivalent radius r and a center point xc Characterization:

[0024]

[0025] x c = (x1+x2+x3+x4) / 4 (3)

[0026] Wherein S represents the area of the fracture, x1-x4 are the coordinates of the four vertices of the fracture.

[0027] Compared with the prior art, the present application has the following advantages:

[0028] The real-time imaging method of artificial fracture network realizes real-time imaging of the fracture network based on microseismic events by obtaining information such as the position of the microseismic events in the corresponding area, can determine the size of the fracture network, and realizes direct characterization of the reservoir fracture, and the imaging result of the fracture network is more complete and accurate. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 It is the implementation process diagram of the real-time imaging method of artificial fracture network.

[0030] Figure 2 It is the distribution diagram of the original microseismic events after processing in the first stage.

[0031] Figure 3 It is the fracture structure calculated in the first stage.

[0032] Figure 4 It is the distribution diagram of the original microseismic events after processing in the second stage.

[0033] Figure 5 It is the fracture structure calculated in the second stage.

[0034] Figure 6 It is the distribution diagram of the original microseismic events after processing in the fourth stage.

[0035] Figure 7 It is the fracture structure calculated in the fourth stage. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0037] The specific implementation of the present application will be described in detail below in combination with specific examples.

[0038] As Figure 1As shown, a real-time imaging method of artificial fracture network provided by one embodiment of the application comprises the following steps:

[0039] 1) Randomly generate fracture network parameter combinations by Monte Carlo method and optimize them, obtain the main occurrence quantity and distribution range of natural fractures in the study area, and the stress orientation and possible dip angle of the corresponding fractures, using the obtained natural fractures, stress orientation data, (effective) microseismic events, and the fractures already known in the previous stage; the obtained fracture orientation is mostly strike, which needs to be converted into dip before applying this method (one strike may correspond to two dips).

[0040] Obtain the microseismic position and fracturing construction data of the study area, divide the microseismic events into stages, and perform de-duplication and DBSCAN denoising processing (parameter selection: the minimum cluster point number is selected as 5, and the search radius is determined according to the maximum positioning error of the microseismic events) on the microseismic event cloud of each stage.

[0041] Randomly generate fracture network parameter combinations (such as the number of existing fractures is equal to the given number of fractures, this step can be skipped) under the given number of fractures (including the fractures already known in the previous stage) by Monte Carlo method: define the maximization objective function D targ As one of the optimization bases:

[0042]

[0043] Where n f is the total number of fractures, p is the number of microseismic events fitted by the fractures under the given distance threshold, d i，j is the distance from the jth microseismic event to the ith fracture, and if it is less than the given threshold, it is considered that the jth microseismic event can be contained by the ith fracture (which fracture it is specifically assigned to is determined in the subsequent process).

[0044] The second basis for optimizing the fracture network is the fitting rate, that is, the proportion of the number of microseismic events fitted by the fractures to the total number of effective microseismic events, and the calculation steps of the fitting rate are: (1) calculate the distance between the existing fractures and the microseismic events, and exclude the fitted microseismic event points; (2) when the remaining microseismic event points are greater than 5% of the total, select a fracture occurrence interval and a microseismic event point to define a fracture; (3) calculate the distance between the fracture and each microseismic event, and exclude the fitted microseismic event points; (4) repeat steps (2) and (3) until the set number of fractures is reached or the remaining microseismic event points are less than 5% of the total; (5) calculate two objective functions; (6) repeat steps (4) and (5), and optimize the fracture network with larger two objective functions as the optimal fracture network.

[0045] 2) Determine the fracture each microseismic event belongs to, project to determine its range, complete the fracture network imaging of this stage.

[0046] With the determined fracture parameters, the newly added microseismic events to be fitted are preferably assigned to the corresponding fracture network using the DBSCAN method. After the assignment is completed, remove the smaller cluster event points assigned to each fracture (these points will be reused in the subsequent stage), project the remaining corresponding new microseismic events to the corresponding fracture, merge and screen with the existing microseismic event projection points to delineate the fracture range, and the fracture generated by the fracture network imaging is a parallelogram, whose range uses the equivalent radius r and the center point x c Characterization:

[0047]

[0048] x c = (x1+x2+x3+x4) / 4 (3)

[0049] Where S represents the area of the fracture, x1 to x4 are the coordinates of the four vertices of the fracture.

[0050] 3) Add new microseismic event groups, combine with existing microseismic events, and repeat steps 1) and 2).

[0051] In step 1), a distance threshold is given to determine whether a point is contained in a fracture, which is the absolute positioning error of the microseismic event; in step 2), there may be some microseismic events that cannot find the fracture they belong to, which are considered in this invention as not participating in the fracture imaging of this stage, but can be used as new microseismic events in the subsequent stage of fracture imaging.

[0052] As shown in Figures 2-7 As a specific embodiment of the present application, microseismic position monitoring data of a certain hot dry rock site is selected for this research. According to the site-related literature research and well logging data, two groups of fracture occurrences are set: 1) natural fracture strikes 300°-330° (i.e. dip 30°-60° or 210°-240°), dip angle 0°-20°; 2) horizontal maximum principal stress direction 23°-45° (corresponding to fracture dip 113°-135° or 293°-315°), dip angle 70°-85°. Since the positioning error of microseismic monitoring is about 40m, the fitting distance threshold and the search radius in DBSCAN microseismic denoising clustering are both set to 40m. After multiple tests, the search radius in DBSCAN clustering of fracture size is determined to be 80m. A total of 2717 microseismic events were generated during fracturing. These events are divided into four stages in chronological order (the stage division described here is only to illustrate the implementation process of the method and has no actual basis, and in order to ensure representativeness, only the results of the first, second and fourth stages are shown):

[0053] Phase 1: The original microseismic events are processed to obtain... Figure 2 The distribution shown is obtained by fitting a set of possible fracture network parameters to these microseismic events using the Monte Carlo method with 10 fractures. Then, the corresponding fracture extent is calculated, and the final fracture structure is as follows. Figure 3 As shown.

[0054] Second stage: The original microseismic events are processed to obtain... Figure 4 The distribution shown uses Figure 3 The 10 fractures shown were used to fit these microseismic events, but this was insufficient for the desired fit. Therefore, a Monte Carlo random generation optimization method was used to calculate and fit a single fracture, and then the extents of all fractures were updated. The results are as follows. Figure 5 As shown.

[0055] Fourth stage: Results obtained after processing Figure 6 As shown in the microseismic events, the existing 11 fractures are sufficient for this stage, so no additional fractures are needed. Only the fracture extent needs to be updated based on the microseismic event fitting results, as shown below. Figure 6 As shown.

[0056] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

Claims

1. A method of real-time imaging of artificial fracture networks, characterized in that, The method comprises the following steps: 1) obtaining information of natural fractures, in-situ stress orientation and microseismic event location of the study area, and randomly generating and optimizing new fracture network parameter combinations by Monte Carlo method when new fractures are generated; 2) determining the fracture to which each microseismic event belongs, projecting to determine the range thereof, and completing fracture network imaging in this stage; 3) adding new microseismic event groups, combining with the existing microseismic event groups, and repeating steps 1) and 2).

2. The method of real-time imaging of artificial fracture networks according to claim 1, characterized in that, In step 1), microseismic location and fracturing construction data of the study area are obtained, the microseismic events are divided into stages, and the microseismic event cloud in each stage is de-duplicated and DBSCAN denoised.

3. The method of real-time imaging of artificial fracture networks of claim 1, wherein, In step 1), given the number of fractures and using the Monte Carlo method, a combination of fracture network parameters is randomly generated, defining a maximization objective function D targ As one of the preferred bases: where n f is the total number of fractures, p is the number of microseismic events that are fitted by a fracture at a given distance threshold, d i，j is the distance of the jth microseismic event to the ith fracture, and if it is less than a given threshold, then the jth microseismic event is considered to be contained by the ith fracture.

4. The method of real-time imaging of artificial fracture networks according to claim 3, characterized in that, Another basis for optimizing the fracture network is the fitting rate, and the calculation of the fitting rate comprises the following steps: 1) calculating the distance between the existing fractures and the microseismic events, and excluding the fitted microseismic event points; 2) when the remaining microseismic event points are greater than 5% of the total, a fracture occurrence interval and a microseismic event point are selected to define a fracture; 3) calculating the distance between the fracture and each microseismic event, and excluding the fitted microseismic event points; 4) repeating steps 2) and 3) until the set number of fractures is reached or the remaining microseismic event points are less than 5% of the total; 5) calculating two objective functions; 6) repeating steps 4) and 5) to obtain the optimal fracture network.

5. The method of real-time imaging of artificial fracture networks of claim 1, wherein, In step 2), in combination with the determined fracture parameters, the newly added fitted microseismic events are optimally assigned to the corresponding fracture network by the DBSCAN method, the smaller cluster event points assigned to each fracture are removed after the assignment is completed, the remaining corresponding new microseismic events are projected to the corresponding fracture, and the fracture range is circled after the projected points are combined and screened with the existing microseismic event projected points.

6. The method of real-time imaging of artificial fracture networks of claim 5, wherein, The fracture network imaging generates fractures as parallelograms, whose extent is characterized by an equivalent radius r and a center point x c Characterization: x c = (xl + x2 + x3 + x4) / 4 (3) Wherein S represents the area of the fracture, and x1 to x4 are the coordinates of the four vertices of the fracture.

Images