Hidden fault identification method based on fracture trend feature analysis

Abstract

The invention relates to a hidden fault identification method based on fracture trend characteristic analysis, and belongs to the technical field of fault identification of oil and gas field seismic exploration. Based on the coherence attribute data volume of the three-dimensional post-stack seismic data of the reference horizon and the research area, analyzing the development characteristics of the fault by angle scanning, and then calculating the fault development characteristics of the corresponding target horizon by adopting the fault angle parameter of the reference horizon; and hidden fault identification of a target layer with weak continuous seismic reflection characteristics or poor seismic data imaging is realized.

Description

Technical Field

[0001] The invention relates to the technical field of fault identification in oil and gas field seismic exploration, and in particular to a hidden fault identification method based on fault strike feature analysis. Background Art

[0002] Faults are a type of geological structure formed when underground rock layers are squeezed or stretched by crustal movement, causing the rocks to break and shift along the fracture surface or fracture zone. The geological structure of faults has important geological significance. Oil and gas are easily enriched near faults, which are the main channels for oil and gas migration and accumulation. In the process of oil and gas exploration and development, fault interpretation is the key to seismic data interpretation. Clearly locating the location of faults and understanding the distribution of faults are of great significance to oil and gas exploration. The results of fault interpretation are directly related to the accuracy of reservoir prediction and affect the efficiency and benefits of oil and gas production.

[0003] At present, the commonly used fault prediction methods at home and abroad are mainly: fault identification method based on seismic attribute fusion, automatic fault identification method and fault identification method based on deep learning, among which the fault identification method based on seismic attribute fusion is the most widely used.

[0004] 1. Principle of fault identification method based on seismic attribute fusion

[0005] The seismic attribute analysis method is to analyze the displacement or interruption of the reflective layer shown by the fault in the seismic data, and to achieve it by enhancing the seismic attributes. The methods based on seismic attributes mainly include coherence volume technology and curvature attributes. Among the fault identification methods, coherence volume technology is widely used. It uses coherence volume attributes to represent the similarity of seismic data, that is, through the calculation of attributes such as inter-channel similarity, the three-dimensional amplitude data volume is converted into a coherence attribute volume, thereby highlighting the development characteristics of faults and geological structures. The curvature attribute has good noise resistance. This method mainly identifies faults through the changes in the undulations (curvature) of the strata. The use of fusion attributes can fully tap the information contained in the data, remove repeated and mixed information, reduce the multi-solution of anomaly prediction, and further improve the accuracy of geological anomaly prediction.

[0006] 2. Principle of automatic fault identification method

[0007] Automatic fault identification methods can automatically identify faults in seismic data, effectively overcoming the error problem caused by subjective human experience. For example, ant tracking technology can identify fault planes and remove the influence of noise and non-faults. This method generally achieves the purpose of suppressing noise and non-fault responses through high-resolution processing, noise reduction, enhancing the continuity of event axes, and eliminating false structures. On this basis, ant tracking technology is used to identify faults.

[0008] 3. Principle of fault identification method based on deep learning

[0009] At present, deep learning is widely used in various fields, and has achieved good results in various fields, with great potential. The amount of seismic data is large, which just meets the premise that deep learning is driven by big data. The application of deep learning in various fields is similar to fault identification, and both require the extraction of image features. Traditional fault interpretation methods need to first determine the location of the fault, which is very similar to the process of extracting the area of ​​interest in image segmentation. Therefore, using deep learning for fault identification is a feasible and efficient method. Although fault identification methods and fault identification methods based on deep learning are gradually being realized, there are still some challenges and limitations.

[0010] 4. Deficiencies of Existing Methods

[0011] 1. The representative RGB attribute fusion technology among the multi-attribute fusion technologies has certain shortcomings. This technology can only be used for three or at most four seismic attributes and cannot meet the requirements for analyzing a larger number of seismic attributes. Therefore, there is a problem of insufficient analysis of seismic attribute information.

[0012] 2. How does machine learning technology work in actual working areas? Different models require different parameters. These model-specific hyperparameters will largely affect the final recognition effect of the model. If you rely solely on manual adjustment of parameters, it will take a lot of time and it is difficult to try as many parameter combinations as possible. Therefore, how to optimize different model hyperparameters through a more reasonable and convenient method is also an important issue that needs to be solved urgently.

[0013] 3. Although the automatic fault identification method is gradually being automated, it still faces some problems. One of the main problems is that the accuracy of the identification results depends on multiple parameters that are set. The selection of these parameters is crucial to the final identification effect and requires careful manual adjustment and optimization. Different parameter selections will lead to different results, so experienced experts are required to adjust the parameters, which to a certain extent limits the full automation of the method. The fault identification method based on deep learning uses a convolutional neural network model as the basis to predict faults. This method realizes the automatic and intelligent identification of faults, and improves the accuracy and efficiency of fault identification. However, there are still some challenges and problems in the prediction of faults in actual work areas, such as: the lack of guidance of geological knowledge when using seismic data for learning, the difficulty of model training, and the low recognition accuracy when the imaging quality of seismic data is poor.

[0014] 4. Existing technologies are limited by data quality. Coherence attributes and curvature attributes require good continuity of the event axis. Faced with the problem of poor quality of some deep seismic data and unclear event axis imaging, the accuracy of fault identification with these two attributes is not high. The accuracy and efficiency of fault automatic identification methods and fault identification methods based on deep learning are also significantly reduced in this case.

[0015] Therefore, in view of the fact that the above conventional methods and techniques are difficult to achieve good results when the seismic data of the target layer is of poor quality, the continuity of the phase axis is poor, or the quality of seismic imaging is poor, the present invention has invented a hidden fault identification method based on fault strike characteristic analysis, which uses known reference horizons and seismic data with good imaging quality to identify hidden faults in the target layer, thereby improving the reliability of fault identification. Summary of the invention

[0016] The present invention relates to a hidden fault identification method based on fault strike feature analysis, and belongs to the technical field of fault identification in oil and gas field seismic exploration. The method first inputs a three-dimensional post-stack seismic data volume of the study area, the seismic horizon of the target layer segment and the reference horizon, and the reference horizon selects the overlying stratum with a relatively continuous phase axis adjacent to the target layer or a good seismic imaging; performs coherence volume analysis on the three-dimensional post-stack seismic data volume to obtain a three-dimensional coherent data volume; then extracts the coherence volume profile of a single main survey line along the direction of the seismic main survey line, and sets a certain layer point of the reference horizon on the profile as the analysis point, and sets a rectangular analysis window centered on the analysis point and scans the fault development at each angle in the analysis window at different angles (calculates all samples at a certain angle in the rectangular window) to obtain the fault development situation. The root mean square of the coherence value of the point is calculated. If the root mean square is less than the set threshold value, it is judged that a fault is developed at this angle), and for the angle with fault development, the fault extension position (target point) of the target layer is determined based on the angle value and the position of the analysis point of the reference layer, and the value of the target point is modified to the root mean square value corresponding to the angle; all analysis points of the reference layer on the profile are calculated cyclically, and all main survey lines of the study area are calculated cyclically to obtain the fault analysis array of the target layer in the study area; finally, the fault analysis array of the target layer is normalized to characterize the fault development characteristics of the target layer.

[0017] The specific steps of the present invention include:

[0018] (1) Input the 3D post-stack seismic data volume of the study area, the reference layer data volume (select the overlying strata with relatively continuous events or good seismic imaging adjacent to the target layer) and the target layer data volume (with poor continuity of seismic events);

[0019] (2) Calculate the coherent attributes of the three-dimensional post-stack data in the study area (A. Gersztenkorn and KJ Marfurt, 1999) to obtain the three-dimensional coherent volume attribute data volume of the study area;

[0020] (3) Extract the coherent volume profile of a single main survey line along the direction of the seismic main survey line, and calculate the distance between the reference layer and the target layer;

[0021] (4) For a single seismic coherence profile, a single CMP point on the reference horizon:

[0022] (4-1) Open an N×N rectangular time window for a single CMP point (the value of N is twice the distance between the reference layer and the target layer, obtained from step 3), and the CMP point is located at the center of the time window;

[0023] (4-2) Extract N×N data in the time window, scan and calculate the root mean square of the coherence volume attribute value in this direction in the rectangular time window every 10° (the total angle range is 0-180°). When the root mean square of the coherence volume attribute value is less than the threshold value, it indicates that a fault has developed in this direction;

[0024] (4-3) For the N×N time window in step (4-1), the discrete data in the time window is extracted along the fault development direction in step (4-2), and the discrete data is fitted by the least square fitting method to obtain a two-dimensional straight line representing the discrete points of the reference horizon fault data, wherein the horizontal coordinate of the straight line is the distance from the CMP point (the difference from the CMP point track number), and the vertical coordinate is time;

[0025] (4-4) Extending the seismic fault data according to the two-dimensional straight line trend obtained in step (4-3) until it passes through the target layer, and obtaining the intersection of the fitting straight line and the target layer, where the intersection is the development position of the fault in the target layer;

[0026] (4-5) Using the RMS value of the coherent attribute in the N×N rectangular time window of the reference layer, replace the RMS value of the coherent attribute in the N×N rectangular time window of the target layer intersection position, and set the angular direction coherence value of the target layer without fault development to 1;

[0027] (5) Circularly calculate all the CMP points of the seismic reference layer in the current section to obtain the target layer fault prediction data volume in the section, that is, the predicted hidden fault data volume;

[0028] (6) All main and side line directions are calculated cyclically to obtain the three-dimensional hidden fault prediction data volume of the study area, and the fault analysis array of the target layer is normalized to characterize the fault development characteristics of the target layer.

[0029] A hidden fault identification method based on fault strike feature analysis has the following characteristics, mainly manifested in:

[0030] (1) The present invention takes advantage of the reference layer, the continuity of the event axis and the clear seismic fault section to establish the connection between the reference layer fault and the hidden fault of the target layer. By calculating the intersection of the fitting straight line of the reference layer fault and the target layer, the development position of the hidden fault of the target layer is obtained. Therefore, it is not limited by the quality of the seismic data of the target layer and has strong innovation.

[0031] (2) The core basis of the present invention is whether the connection between the reference layer fault and the hidden fault of the target layer can be established? Can the intersection position of the fitting straight line of the reference layer fault and the target layer be used to indicate the development position of the hidden fault of the target layer? After consulting a large amount of data, the inventor found that the earthquake source is often deep, the earthquake fault plane mostly extends from bottom to top, and the section can be fitted with two-dimensional straight lines within a smaller longitudinal range; through the result of two-dimensional straight line fitting within the longitudinal range of the fault, the connection between the reference layer fault and the hidden fault of the target layer is established, and then the hidden fault of the target layer is identified; and there is methodological theoretical support: ① Earthquakes are mainly caused by the mutual compression and collision between plates on the earth. At this time, the earth's interior releases a large amount of energy to the surface, causing the plate edges and the interior of the plates to shift and rupture, thereby forming sections from bottom to top; ② The straight line obtained by fitting the earthquake section using the least squares method within a small range can be equivalent to the extension trend of the fault plane within this range; therefore, the present invention has a good methodological theoretical basis. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a technical flow chart of the present invention;

[0033] Figure 2 This is the predicted plan view of hidden faults in the target layer obtained by conventional methods;

[0034] Figure 3 It is a prediction plane map of hidden faults of the target layer obtained by the method of the present invention; DETAILED DESCRIPTION

[0035] Example 1

[0036] A hidden fault identification method based on fault strike feature analysis, comprising the following steps:

[0037] Step 1: Input the 3D seismic data volume of the study area, the reference layer and the target layer data volume. The reference layer data volume is denoted as H 上 (line, xline, t), the target layer data volume is denoted as H 下 (line,xline,t), where line is the line number, xline is the track number, and t is the time;

[0038] Step 2: Calculate the coherent attributes of the three-dimensional data volume in the study area to obtain the three-dimensional coherent attribute data volume of the reference layer in the study area. The calculation method is as follows:

[0039] Given a covariance matrix of a three-dimensional data volume within an analysis window,

[0040]

[0041] The coherence value of the eigenvalue-based C3 coherence algorithm is defined as,

[0042]

[0043] The denominator Tr(C) is the trace of the matrix, representing the energy of the covariance matrix, and the numerator λ max is the maximum eigenvalue, usually the first eigenvalue λ1, representing the dominant energy, λ j (j=1, 2, ..., J) is the eigenvalue of the covariance matrix C, arranged in descending order. When the waveforms of all traces in the analysis window are consistent, that is, when no faults are developed, the eigenvalue coherence coefficient is equal to 1;

[0044] Step 3: extract the attribute profile of the 3D coherent data along the main survey line, that is, pr(xline, t, value), where value is the seismic coherent attribute value, and calculate the distance between the reference layer and the target layer N = 2×(t 下 -t 上 ), t 下 is the time of the destination layer, t 上 is the time of the reference layer;

[0045] Step 4: For a single seismic attribute section, refer to a single CMP point in the horizon, with the coordinates marked as (xline0, t0).

[0046] Step 4-1, open an N×N time window for a single CMP point, and the CMP point is located at the center of the time window. The value of N is twice the distance between the reference layer and the target layer, which is obtained from step 3;

[0047] Step 4-2, extract N×N data in the time window, calculate the root mean square of the coherence volume attribute value in this direction in the time window every 10°, with a total angle range of 0-180°. When the root mean square of the coherence volume attribute value is less than or equal to 0.3, 0.3 is the set threshold value, that is, assuming that the coherence volume attribute value is lower than 0.3, it is a fault development point, and this direction is a fault development direction in the time window;

[0048] Step 4-3, for the N×N time window of step 4-1, extract all discrete data in this direction in the time window along the fault development direction of step 4-2, that is, extract the abscissa channel number xline and the ordinate time t of the discrete data, record it as the reference layer fault data f(xline-xline0, t-t0), and use the least squares fitting method to fit the earthquake fault f(xline-xline0, t-t0) data. The fitting abscissa range is xline to xline0, and a two-dimensional straight line T(xline1)=a+b×xline1 representing the discrete points of the fault data f(xline-xline0, t-t0) is obtained. The abscissa of this straight line is the distance between the earthquake fault f(xline-xline0, t-t0) and the CMP point (xline0, t0), xline1=xline n -xline0,n=1,2,…,n, the ordinate is time t, and the least squares formula is as follows:

[0049] The fitting straight line is

[0050] y=a+bx (3)

[0051] and

[0052]

[0053] Where a and b are the fitting straight line parameters, which are obtained by combining Formula 4 and Formula 5. x is the track number difference of the horizontal coordinate, y is the time difference of the vertical coordinate, and n is the number of discrete data fitted.

[0054] Step 4-4, extend the seismic fault data f(xline-xline0, t-t0) according to the trend of the two-dimensional straight line T(xline1)=a+b×xline1 obtained in step 4-2, and the extension point data is the value of the two-dimensional straight line data at the extension point until it passes through the target layer, and use the extended two-dimensional straight line data body as the target layer hidden fault data fpre(xline, t) corresponding to the reference layer fault data f(xline-xline0, t-t0);

[0055] Step 4-5, using the RMS value of the coherent attribute in the N×N rectangular time window of the reference layer, replace the RMS value of the coherent attribute in the N×N rectangular time window of the intersection position of the target layer, and set the angular direction coherence value of the target layer without fault development to 1;

[0056] Step 5, looping and calculating all CMP points in the current section to obtain the hidden fault prediction data volume fpre(xline,t) of the target layer in the section, that is, the hidden fault data volume predicted by the section;

[0057] Step 6: Circularly calculate all the main survey line directions to obtain the three-dimensional hidden fault prediction data volume Fre(line, xline, t) of the study area, and perform normalization processing to obtain a coherent data volume that characterizes the fault development characteristics of the target layer.

[0058] Example 2

[0059] Figure 2 and Figure 3 They are examples of hidden fault prediction in a target layer section of a certain study area. Figure 2 This is the plane diagram of the prediction results of conventional superimposed hidden faults in the study area. Figure 3 From the hidden fault prediction plan view obtained by the present invention, it can be seen that the crack prediction effect of the present invention is better.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying hidden faults based on fault strike characteristics analysis includes the following specific steps: Step 1, input the 3D seismic data volume of the study area, the reference layer and the target layer data volume, the reference layer data volume is recorded as Hup (line, xline, t), and the target layer data volume is recorded as Hdown (line, xline, t), where line is the line number, xline is the trace number, and t is the time; Step 2: Calculate the coherent attributes of the three-dimensional data volume in the study area to obtain the three-dimensional coherent attribute data volume of the reference layer in the study area. The calculation method is as follows: Given a covariance matrix of a three-dimensional data volume within an analysis window, The coherence value of the eigenvalue-based C3 coherence algorithm is defined as, The denominator Tr(C) is the trace of the matrix, representing the energy of the covariance matrix, and the numerator λ max is the maximum eigenvalue, usually the first eigenvalue λ1, representing the dominant energy, λ j (j=1, 2, ..., J) is the eigenvalue of the covariance matrix C, arranged in descending order. When the waveforms of all traces in the analysis window are consistent, that is, when no faults are developed, the eigenvalue coherence coefficient is equal to 1; Step 3, extract the attribute profile of the three-dimensional coherent data along the main survey line direction, that is, pr(xline, t, value), where value is the seismic coherent attribute value, and calculate the distance between the reference layer and the target layer N = 2×(t down - t up), where t down is the time of the target layer and t up is the time of the reference layer; Step 4: For a single seismic attribute section, refer to a single CMP point in the horizon, with the coordinates marked as (xline0, t0). Step 4-1, open an N×N time window for a single CMP point, and the CMP point is located at the center of the time window. The value of N is twice the distance between the reference layer and the target layer, which is obtained from step 3; Step 4-2, extract N×N data in the time window, calculate the root mean square of the coherence volume attribute value in this direction in the time window every 10°, with a total angle range of 0-180°. When the root mean square of the coherence volume attribute value is less than or equal to 0.3, 0.3 is the set threshold value, that is, assuming that the coherence volume attribute value is lower than 0.3, it is a fault development point, and this direction is a fault development direction in the time window; Step 4-3, for the N×N time window of step 4-1, extract all discrete data in this direction in the time window along the fault development direction of step 4-2, that is, extract the abscissa channel number xline and the ordinate time t of the discrete data, record it as the reference layer fault data f(xline-xline0, t-t0), and use the least squares fitting method to fit the earthquake fault f(xline-xline0, t-t0) data. The fitting abscissa range is xline to xline0, and a two-dimensional straight line T(xline1)=a+b×xline1 representing the discrete points of the fault data f(xline-xline0, t-t0) is obtained. The abscissa of this straight line is the distance between the earthquake fault f(xline-xline0, t-t0) and the CMP point (xline0, t0), xline1=xlinen-xline0, n=1,2,…,n, and the ordinate is time t. The least squares method formula is as follows: The fitting straight line is y=a+bx (3) and Where a and b are the fitting straight line parameters, which are obtained by combining Formula 4 and Formula 5. x is the track number difference of the horizontal coordinate, y is the time difference of the vertical coordinate, and n is the number of discrete data fitted. Step 4-4, extend the seismic fault data f(xline-xline0, t-t0) according to the trend of the two-dimensional straight line T(xline1)=a+b×xline1 obtained in step 4-2, and the extension point data is the value of the two-dimensional straight line data at the extension point until it passes through the target layer, and use the extended two-dimensional straight line data body as the target layer hidden fault data fpre(xline, t) corresponding to the reference layer fault data f(xline-xline0, t-t0); Step 4-5, using the RMS value of the coherent attribute in the N×N rectangular time window of the reference layer, replace the RMS value of the coherent attribute in the N×N rectangular time window of the intersection position of the target layer, and set the angular direction coherence value of the target layer without fault development to 1; Step 5, looping and calculating all CMP points in the current section to obtain the hidden fault prediction data volume fpre(xline,t) of the target layer in the section, that is, the hidden fault data volume predicted by the section; Step 6: Circularly calculate all the main survey line directions to obtain the three-dimensional hidden fault prediction data volume Fre(line, xline, t) of the study area, and perform normalization processing to obtain a coherent data volume that characterizes the fault development characteristics of the target layer.

Images