Prediction method for small fault of complex fault block oil reservoir

Abstract

The invention relates to a prediction method for a small fault of a complex fault block oil reservoir, which calculates a structural gradient tensor seismic body, reflects the characteristics of curvature change of a seismic wave event structure, can finely identify the small fluctuation of the structure caused by the dislocation of the seismic event, has a good identification capability for small and medium faults, and can be used for predicting the small fault of the complex fault block oil reservoir. The method comprises the following steps: firstly, constructing a gradient tensor seismic body, then carrying out curvature change gradient attribute calculation on the constructed gradient tensor seismic body, reflecting the change gradients of the constructed curvature in X and Y directions in the data body and highlighting the discontinuity of an event, finally calculating a curvature change gradient variance body of multi-channel seismic data, and completing minor fault tracking according to the variance value by extracting a horizontal slice of a target layer; a visual basis is provided for interpretation of the small fault in the complex fault block oil reservoir, and the problem of complex fault block oil reservoir structure interpretation is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and natural gas exploitation, and in particular relates to a method for predicting small faults in complex fault-block reservoirs. Background Art

[0002] Complex fault-block reservoirs generally have faults of various levels. Currently, fault identification is mainly achieved through fault interpretation of seismic data. Due to the limitation of seismic data resolution, the seismic responses of some small faults are weak and fuzzy, resulting in unclear understanding of such small faults, which in turn affects the understanding of the distribution of complex fault-block reservoirs and the ultimate exploration and development effects.

[0003] Large faults are mainly identified through manual interpretation of seismic profiles, and small faults are mainly identified through technical means such as curvature, dip, and coherence analysis. However, the identification effect is relatively poor in areas with relatively poor seismic data. Therefore, it is necessary to summarize new small fault interpretation methods to achieve more accurate small-scale fault predictions and provide technical support for subsequent complex block structure interpretations.

[0004] At present, the interpretation of small faults in complex fault blocks requires overcoming the following technical difficulties: (1) How to select appropriate seismic attributes to identify small faults; (2) How to use data enhancement technology to make coherent slices clearer and improve the accuracy of fault interpretation; Therefore, in response to the above problems, this application uses the structural gradient tensor attribute to reflect the change in the structural curvature of the seismic wave phase axis, and then performs a secondary gradient attribute calculation on the structural gradient tensor seismic body, and finally calculates the curvature change gradient variance body of multi-channel seismic data to complete the tracking of small faults based on the high and low coherence values. Summary of the invention

[0005] This patent calculates the structural gradient tensor seismic body, which reflects the characteristics of the change in the structural curvature of the seismic wave event axis. It can accurately identify the slight structural fluctuations caused by the faulting of the seismic event axis, and has a good recognition ability for small and medium-sized faults. The curvature change gradient attribute of the structural gradient tensor seismic body is then calculated to reflect the gradient change of the structural curvature in the X and Y directions in the data body, and continue to highlight the discontinuity of the event axis. Finally, the curvature change gradient variance body of the multi-channel seismic data is calculated, and the small faults are tracked according to the variance value by extracting the horizontal slices of the target layer. This technology provides an intuitive basis for the interpretation of small faults in complex fault block reservoirs, and overcomes the difficulty of complex fault block structure interpretation.

[0006] The present invention solves the technical problem by adopting the following technical solutions:

[0007] The prediction method of small faults in complex fault-block reservoirs includes the following steps:

[0008] S1. Establish a stratigraphic comparison section of the drilled well to determine the depth range of the top and bottom interfaces of the target layer;

[0009] S2. Complete the time-depth conversion of each well, track the top and bottom interfaces of the target layer, and determine its time range in the seismic data body;

[0010] S3, intercepting the seismic data volume to ensure that the time range of the target layer is included;

[0011] S4, performing structural gradient tensor seismic volume calculation on the intercepted seismic data volume;

[0012] S5, calculating the curvature change gradient attribute of the structural gradient tensor seismic volume;

[0013] S6, calculating the curvature change gradient variance volume of the multi-channel seismic data based on the curvature change gradient attribute volume;

[0014] S7. Extract the horizontal slice of the curvature change gradient variance body of the target layer, and determine the position of the fault according to the variance value of the target layer.

[0015] Furthermore, the structural gradient tensor seismic volume calculation is performed on the intercepted seismic data body, which reflects the characteristics of the change in the structural curvature of the seismic wave event axis. It can accurately identify the slight structural fluctuations caused by the misalignment of the seismic event axis and has a good recognition ability for small and medium-sized faults.

[0016] Furthermore, the method for calculating the curvature change gradient attribute of the construction gradient tensor seismic volume in step S5 is:

[0017] ① Obtain the gradient of seismic data;

[0018] ② Construct a structure tensor matrix and smooth or integrate the elements of the structure tensor matrix respectively;

[0019] ③Eigenvalue decomposition and sorting.

[0020] Furthermore, the method for obtaining the gradient of seismic data is:

[0021] By calculating the first-order partial derivatives g1, g2, and g3 of the three-dimensional seismic data in the vertical, inline, and crossline directions, the gradient vector of each data sample point corresponding to the original three-dimensional seismic data can be obtained:

[0022] g=(g1, g2, g3)T (1)

[0023] g cannot directly construct a tensor matrix to solve the eigenvalue, because the rank of the directly constructed tensor matrix is ​​not greater than 1, even if g1, g2, and g3 are smoothed and then the tensor matrix is ​​reconstructed;

[0024]

[0025] When T grad When all elements of are not zero, T grad Only one of the three eigenvalues ​​λ1, λ2, and λ3 of is non-zero, then

[0026] λ1=g1 2 +g2 2 +g3 2 (3)

[0027] Then λ1 is the energy of the gradient.

[0028] Furthermore, constructing the structure tensor matrix includes the following steps:

[0029] Multiply the gradient vector by its transpose to form a 3rd order square matrix (also called dyadic product);

[0030] Each element in the matrix is ​​smoothed (or locally averaged) to obtain the average structure tensor matrix within a certain scale range:

[0031]

[0032] Where 〈·〉 represents the smoothing of the objects in the diagonal brackets, usually using Gaussian filtering. From the symmetry of the tensor matrix, it can be seen that only 6 elements need to be smoothed to complete the smoothing of 9 elements. If g1, g2, and g3 are also smoothed, a total of 9 three-dimensional data volumes need to be smoothed. The smoothing calculation of gradient elements can suppress the local gradient anomaly caused by seismic noise, thereby enhancing the robustness of subsequent calculations. The smoothing calculation of tensor elements makes the tensor matrix contain not only the characteristics of the maximum gradient energy direction, but also the change characteristics of all orthogonal directions.

[0033] Furthermore, when calculating the curvature change gradient attribute of the structural gradient tensor seismic volume in step S5, the curvature change gradient formula is:

[0034]

[0035] In the formula, K ab is the curvature change gradient of points a and b; Dx and Dy are the curvature differences between the two points in the horizontal and vertical directions.

[0036] Furthermore, the method for calculating the curvature change gradient variance volume of multi-channel seismic data from the curvature change gradient attribute volume in step S6 is:

[0037] The curvature change gradient variance body is mainly obtained by calculating the variance value of the curvature change gradient between the average seismic trace and the seismic trace in the trace set. It represents the difference in reflection characteristics (curvature change) between each seismic trace, and mines the seismic abnormal reflection information caused by abnormal geological bodies or faults, so as to realize the identification of small faults. The variance of a certain sampling point and all the sample points of the surrounding adjacent seismic traces within a given time window is obtained, and then the variance value of the point is obtained after weighted normalization;

[0038]

[0039]

[0040] ω=sinθ(0≤θ≤90°)

[0041] In the formula, δ 2 is the variance value of a sampling point in a certain channel; δ 2 ω is the weighted variance value of the sampling points; A ij is the amplitude of seismic trace i at time j; is the average amplitude of the seismic trace included in the time window calculation at time j; L is the time window length for calculating the variance, in ms; n is the number of adjacent traces required to calculate the variance of a certain sampling point; ω j-t is the triangular weighted function of a sampling point in a time window.

[0042] Furthermore, when intercepting the seismic data body, the seismic data body is intercepted according to the time range of the target layer in the seismic data body to ensure that the intercepted data body contains the time range of the target layer, which can avoid non-target layer data interfering with the calculation results of the target layer and save calculation time.

[0043] The advantages and positive effects of the present invention are:

[0044] The method for predicting small faults in complex fault-block reservoirs of the present invention provides an effective technical means and an on-site implementation method for predicting small faults in complex fault-block reservoirs. According to the method of the present invention, the technical problem of inaccurate interpretation of small faults can be solved, and important technical information can be provided for the study of complex fault-block reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments, but it should be understood that these drawings are designed only for explanation purposes and are not intended to limit the scope of the present invention. In addition, unless otherwise specified, these drawings are intended only to conceptually illustrate the structural configurations described herein and are not necessarily drawn to scale.

[0046] Figure 1 The time range of determining the target layer in the seismic data volume provided by the embodiment of the present invention;

[0047] Figure 2 The time range of the target layer included in the truncated seismic data volume provided by the embodiment of the present invention;

[0048] Figure 3 The coherent body technology based on intrinsic structure provided by the embodiment of the present invention;

[0049] Figure 4 The curvature change gradient variance body of the multi-channel seismic data provided by the embodiment of the present invention;

[0050] Figure 5 A horizontal slice of the curvature variation gradient variance volume provided by an embodiment of the present invention; DETAILED DESCRIPTION

[0051] First of all, it should be noted that the specific structure, features and advantages of the present invention will be specifically described below by way of example, but all descriptions are only for illustration and should not be understood as limiting the present invention in any way. In addition, any single technical feature described or implied in the embodiments mentioned herein, or any single technical feature displayed or implied in the drawings, can still be combined or deleted between these technical features (or their equivalents) to obtain more other embodiments of the present invention that may not be directly mentioned in this document. In addition, in order to simplify the drawings, the same or similar technical features may be marked only in one place in the same drawing.

[0052] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other.

[0053] Example 1

[0054] The method for predicting small faults in complex fault-block reservoirs provided in this embodiment includes the following steps:

[0055] S1. Establish a stratigraphic comparison section of the drilled well to determine the depth range of the top and bottom interfaces of the target layer;

[0056] S2. Complete the time-depth conversion of each well, track the top and bottom interfaces of the target layer, and determine its time range in the seismic data body;

[0057] S3, intercepting the seismic data volume to ensure that the time range of the target layer is included;

[0058] S4, performing structural gradient tensor seismic volume calculation on the intercepted seismic data volume;

[0059] S5, calculating the curvature change gradient attribute of the structural gradient tensor seismic volume;

[0060] S6, calculating the curvature change gradient variance volume of the multi-channel seismic data based on the curvature change gradient attribute volume;

[0061] S7. Extract the horizontal slice of the curvature change gradient variance body of the target layer, and determine the position of the fault according to the variance value of the target layer.

[0062] Among them, the structural gradient tensor seismic volume calculation is performed on the intercepted seismic data body, which reflects the characteristics of the change of the structural curvature of the seismic wave event axis. It can accurately identify the slight structural fluctuations caused by the misalignment of the seismic event axis, and also has a good recognition ability for small and medium-sized faults.

[0063] Specifically, the method for calculating the curvature change gradient attribute of the structural gradient tensor seismic volume in step S5 is:

[0064] ① Obtain the gradient of seismic data;

[0065] ② Construct a structure tensor matrix and smooth or integrate the elements of the structure tensor matrix respectively;

[0066] ③Eigenvalue decomposition and sorting.

[0067] The method for obtaining the gradient of seismic data is:

[0068] By calculating the first-order partial derivatives g1, g2, and g3 of the three-dimensional seismic data in the vertical, inline, and crossline directions, the gradient vector of each data sample point corresponding to the original three-dimensional seismic data can be obtained:

[0069] g=(g1, g2, g3)T (1)

[0070] g cannot directly construct a tensor matrix to solve the eigenvalue, because the rank of the directly constructed tensor matrix is ​​not greater than 1, even if g1, g2, and g3 are smoothed and then the tensor matrix is ​​reconstructed;

[0071] T grad is the constructed tensor matrix

[0072]

[0073] When T grad When all elements of are not zero, T grad Only one of the three eigenvalues ​​λ1, λ2, and λ3 is not zero (e.g., λ1≠0, λ2=λ3=0), then

[0074] λ1=g1 2 +g2 2 +g3 2 (3)

[0075] Then λ1 is the energy of the gradient.

[0076] Specifically, constructing the structure tensor matrix includes the following steps:

[0077] Multiply the gradient vector by its transpose to form a 3rd order square matrix (also called dyadic product);

[0078] Each element in the matrix is ​​smoothed (or locally averaged) to obtain the average structure tensor matrix within a certain scale range:

[0079]

[0080] Where 〈·〉 represents the smoothing of the objects in the diagonal brackets, usually using Gaussian filtering. From the symmetry of the tensor matrix, it can be seen that only 6 elements need to be smoothed to complete the smoothing of 9 elements. If g1, g2, and g3 are also smoothed, a total of 9 three-dimensional data volumes need to be smoothed. The smoothing calculation of gradient elements can suppress the local gradient anomaly caused by seismic noise, thereby enhancing the robustness of subsequent calculations. The smoothing calculation of tensor elements makes the tensor matrix contain not only the characteristics of the maximum gradient energy direction, but also the change characteristics of all orthogonal directions.

[0081] When calculating the curvature change gradient attribute of the structural gradient tensor seismic volume in step S5, the curvature change gradient formula is:

[0082]

[0083] In the formula, K ab is the curvature change gradient of points a and b; Dx and Dy are the curvature differences between the two points in the horizontal and vertical directions.

[0084] The method for calculating the curvature change gradient variance volume of multi-channel seismic data from the curvature change gradient attribute volume in step S6 is:

[0085] The curvature change gradient variance volume is mainly obtained by calculating the variance value of the curvature change gradient between the average seismic trace and the seismic trace in the gather, which represents the difference in reflection characteristics (curvature change) between the seismic traces, and mines the seismic abnormal reflection information caused by abnormal geological bodies or faults, thereby realizing the identification of small faults; the variance of a certain sampling point and all the sample points of the surrounding adjacent seismic traces within a given time window is obtained, and then the variance value of the point is obtained after weighted normalization;

[0086]

[0087]

[0088] ω=sinθ(0≤θ≤90°)

[0089] In the formula, δ2 is the variance value of a sampling point in a certain channel; δ 2 ω is the weighted variance value of the sampling points; A ij is the amplitude of seismic trace i at time j; is the average amplitude of the seismic trace included in the time window calculation at time j; L is the time window length for calculating the variance, in ms; n is the number of adjacent traces required to calculate the variance of a certain sampling point; ω j-t It is the triangular weighted function of a sampling point in a time window, and its value ranges from 0 to 1.

[0090] When intercepting the seismic data body, the seismic data body is intercepted according to the time range of the target layer in the seismic data body to ensure that the intercepted data body contains the time range of the target layer, which can avoid non-target layer data from interfering with the calculation results of the target layer and save calculation time.

[0091] Example 2

[0092] The present embodiment provides a method for predicting small faults in complex fault-block reservoirs, which can be regarded as an effective method for using the technology for fine interpretation of small faults in complex fault-block reservoirs. The specific implementation method is as follows:

[0093] (1) Establish a stratigraphic comparison profile of the drilled well to determine the depth range of the top and bottom interfaces of the target layer;

[0094] Establish a stratigraphic comparison profile of the drilled wells to ensure that the wells that encounter the top and bottom ends of the target layer are in the profile, and calculate the depth range of the top and bottom interfaces of the target layer;

[0095] (2) Complete the time-to-depth conversion of each well, track the top and bottom interfaces of the target layer, and determine its time range within the seismic data volume;

[0096] The top and bottom interfaces of the target layers of each well in the work area are converted into time and depth, the top and bottom interfaces of the target layers are tracked, and the time range of the target layers in the seismic data volume is determined, such as Figure 1 As shown;

[0097] (3) Truncate the seismic data volume to ensure that the time range of the target layer is included:

[0098] According to the time range of the target layer in the seismic data volume, the seismic data volume is intercepted to ensure that the intercepted data volume contains the time range of the target layer, such as Figure 2 As shown, the data body is intercepted in order to prevent the non-target layer data from interfering with the target layer calculation results in subsequent calculations, while saving calculation time;

[0099] (4) Perform structural gradient tensor seismic volume calculation on the intercepted seismic data volume:

[0100] The structural gradient tensor seismic volume is calculated for the intercepted seismic data volume, such as Figure 3 As shown in the figure, the gradient structural tensor attribute is to construct the gradient structural tensor by calculating the seismic data volume, which reflects the characteristics of the change of the structural curvature of the seismic wave phase axis. It can finely identify the slight structural fluctuations caused by the fault of the seismic phase axis, and also has a good recognition ability for small and medium faults.

[0101] (5) Calculate the curvature change gradient attribute of the structural gradient tensor seismic volume:

[0102] The curvature change gradient attribute of the structural gradient tensor seismic volume is calculated. The curvature change gradient attribute reflects the gradient change of the structural curvature in the X and Y directions in the data volume, and continues to highlight the discontinuity of the event axis.

[0103] (6) Calculate the curvature change gradient variance volume of multi-channel seismic data based on the curvature change gradient attribute volume:

[0104] The curvature change gradient variance volume of multi-channel seismic data is calculated for the curvature change gradient attribute volume. The curvature change gradient variance volume is mainly obtained by calculating the variance value of the curvature change gradient between the average seismic trace and the seismic traces in the trace set. It represents the reflection characteristic difference (curvature change) between each seismic trace, and mines the seismic abnormal reflection information caused by abnormal geological bodies or faults, so as to realize the identification of small faults; the variance of a certain sampling point and all the sample points of the surrounding adjacent seismic traces within a given time window is obtained, and then the variance value of the point is obtained after weighted normalization, such as Figure 4 As shown, it is the variance body.

[0105] (7) Extract the horizontal slice of the curvature change gradient variance body of the target layer and determine the location of the fault according to the time value of the target layer:

[0106] Extract the horizontal slice of the curvature change gradient variance volume of the target layer, such as Figure 5 As shown, the location of the fault is determined according to the variance value of the target layer, and a variety of processing methods are used to help accurately combine the fracture planes.

[0107] The above embodiments describe the present invention in detail, but the contents are only preferred embodiments of the present invention and cannot be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A method for predicting small faults in complex fault-block reservoirs, characterized in that: The steps include: S1. Establish a stratigraphic comparison section of the drilled well to determine the depth range of the top and bottom interfaces of the target layer; S2. Complete the time-depth conversion of each well, track the top and bottom interfaces of the target layer, and determine its time range in the seismic data body; S3, intercepting the seismic data volume to ensure that the time range of the target layer is included; S4, performing structural gradient tensor seismic volume calculation on the intercepted seismic data volume; S5, calculating the curvature change gradient attribute of the structural gradient tensor seismic volume; S6, calculating the curvature change gradient variance volume of the multi-channel seismic data based on the curvature change gradient attribute volume; S7. Extract the horizontal slice of the curvature change gradient variance body of the target layer, and determine the position of the fault according to the variance value of the target layer.

2. The method for predicting small faults in complex fault-block reservoirs according to claim 1, characterized in that: The structural gradient tensor seismic volume calculation is performed on the intercepted seismic data body, which reflects the characteristics of the change in the structural curvature of the seismic wave phase axis. It can accurately identify the slight structural fluctuations caused by the misalignment of the seismic phase axis, and also has a good recognition ability for small and medium-sized faults.

3. The method for predicting small faults in complex fault-block reservoirs according to claim 1, characterized in that: The method for calculating the curvature change gradient attribute of the structural gradient tensor seismic volume in step S5 is: ① Obtain the gradient of seismic data; ② Construct a structure tensor matrix and smooth or integrate the elements of the structure tensor matrix respectively; ③Eigenvalue decomposition and sorting.

4. The method for predicting small faults in complex fault-block reservoirs according to claim 3, characterized in that: The method for obtaining the gradient of seismic data is: By calculating the first-order partial derivatives g1, g2, and g3 of the three-dimensional seismic data in the vertical, inline, and crossline directions, the gradient vector of each data sample point corresponding to the original three-dimensional seismic data can be obtained: g=(g1, g2, g3)T (1) g cannot directly construct a tensor matrix to solve the eigenvalues, because the rank of the directly constructed tensor matrix is ​​not greater than 1, even if g1, g2, and g3 are smoothed and then the tensor matrix is ​​reconstructed.

5. The method for predicting small faults in complex fault-block reservoirs according to claim 3, characterized in that: Constructing the structure tensor matrix includes the following steps: Multiply the gradient vector and its transpose to form a 3rd order square matrix; Smooth each element in the matrix separately to obtain the average structure tensor matrix within a certain scale range: Where 〈·〉 represents the smoothing of the objects in the diagonal brackets, usually using Gaussian filtering. From the symmetry of the tensor matrix, it can be seen that only 6 elements need to be smoothed to complete the smoothing of 9 elements. If g1, g2, and g3 are also smoothed, a total of 9 three-dimensional data volumes need to be smoothed. The smoothing calculation of gradient elements can suppress the local gradient anomaly caused by seismic noise, thereby enhancing the robustness of subsequent calculations. The smoothing calculation of tensor elements makes the tensor matrix contain not only the characteristics of the maximum gradient energy direction, but also the change characteristics of all orthogonal directions.

6. The method for predicting small faults in complex fault-block reservoirs according to claim 1, characterized in that: When calculating the curvature change gradient attribute of the structural gradient tensor seismic volume in step S5, the curvature change gradient formula is: In the formula, K ab is the curvature change gradient of points a and b; Dx and Dy are the curvature differences between the two points in the horizontal and vertical directions.

7. The method for predicting small faults in complex fault-block reservoirs according to claim 1, characterized in that: The method for calculating the curvature change gradient variance volume of multi-channel seismic data from the curvature change gradient attribute volume in step S6 is: The curvature change gradient variance volume is mainly obtained by calculating the variance value of the curvature change gradient between the average seismic trace and the seismic trace in the gather. It represents the difference in reflection characteristics between the seismic traces, and mines the seismic abnormal reflection information caused by abnormal geological bodies or faults, so as to realize the identification of small faults. The variance of a certain sampling point and all the sample points of the surrounding adjacent seismic traces within a given time window is obtained, and then the variance value of the point is obtained after weighted normalization. ω=sinθ(0≤θ≤90°) In the formula, δ 2 is the variance value of a sampling point in a certain channel; δ 2 ω is the weighted variance value of the sampling points; A ij is the amplitude of seismic trace i at time j; is the average amplitude of the seismic trace included in the time window calculation at time j; L is the time window length for calculating the variance; n is the number of adjacent traces required to calculate the variance of a certain sampling point; ω j-t is the triangular weighted function of a sampling point in a time window.

8. The method for predicting small faults in complex fault-block reservoirs according to claim 1, characterized in that: When intercepting the seismic data body, the seismic data body is intercepted according to the time range of the target layer in the seismic data body to ensure that the intercepted data body contains the time range of the target layer, which can avoid non-target layer data from interfering with the calculation results of the target layer and save calculation time.

Images