A method for fine fault identification based on ant tracking technology

Abstract

The present invention discloses a method for fine fault identification based on ant tracking technology, which is characterized in that the slice data of typical tracks are taken from the original seismic coherence volume data as the seismic coherence volume image, and the high-pass filter is applied to intercept the high-frequency area to reduce the influence of noise; the data blocks are divided according to the number of ants to deploy the ants, the initial search window is set according to the limited range of the fault strike angle, the ant moving point is selected by roulette, the fitness value of the moving path is retained and the search window is adjusted, the pheromone concentration is superimposed according to the ant moving trajectory, and the fine fault identification result is output after the tracking is completed. The present invention uses an adaptive window strategy to plan the action trajectory of the ant tracking technology, so that it has higher search efficiency in simple fault zones and better fault identification accuracy in complex fault zones; and introduces a random ant mutation mechanism to improve the description ability of hidden faults.

Description

Technical Field

[0001] The present invention belongs to the field of geophysical exploration and artificial intelligence, and in particular relates to a method for finely identifying faults based on ant tracking technology. Background Art

[0002] As the oilfield development stage continues to deepen, contradictions such as rising water content and decreasing production are becoming increasingly prominent. Improving the ability to interpret seismic data based on seismic data can provide guarantees for adjusting the injection and production system and finding external expansion potential. Fault interpretation is the core and foundation of seismic data interpretation. Enhancing the ability to finely characterize faults can effectively improve the level of seismic data interpretation. However, when the fault distribution in the seismic data is a relatively complex fault zone, or there are multiple faults intersecting, traditional fault interpretation and analysis methods will lead to difficulties in fault interpretation. Conventional ant tracking technology cannot well identify and analyze fault characteristics, and the accuracy of fault model description is limited. Therefore, in the actual geological exploration process, how to improve the ability to finely describe faults is a very important issue.

[0003] Aiming at the problem that conventional ant tracking technology has low recognition accuracy for complex faults and limited accuracy in describing fault areas, in order to enable it to have better fault description capabilities, an adaptive window is used to plan ant movement trajectories, and combined with a random ant mutation mechanism, a fault fine identification method based on ant tracking technology is proposed. Summary of the invention

[0004] In order to overcome the problem of low accuracy in automatic fault identification of conventional ant tracking technology, the present invention proposes an adaptive window strategy to plan the movement trajectory of ants in ant tracking technology, so that it has higher search efficiency in simple fault zones and better fault identification accuracy in complex fault zones; and introduces a random ant mutation mechanism to increase the global search accuracy of ant tracking technology and improve the description ability of ant tracking technology for hidden faults.

[0005] To achieve the above object, the technical solution of the present invention mainly includes the following steps:

[0006] 1. Seismic coherence volume enhanced image acquisition:

[0007] Since seismic attributes are disturbed by noise and other factors, directly selecting the original seismic coherence volume image will affect the tracking effect. Digital image processing technology is used to process the seismic coherence volume image to enhance the characteristics of the fault area in the image and weaken the characteristics of the non-fault area to obtain the seismic coherence volume enhanced image.

[0008] The steps for acquiring seismic coherence volume enhanced images are as follows:

[0009] (1) Obtaining original seismic coherence volume data from data sources such as exploration databases and seismic data volumes, and extracting slice data of typical traces from them as seismic coherence volume images I0;

[0010] (2) Extracting frequency domain features based on the spatial features of the seismic coherence volume image, applying a Basworth high-pass filter to intercept the high-frequency area of ​​the seismic coherence volume image, retaining a small amount of low-frequency area, and obtaining the seismic coherence volume sharpened image I1;

[0011] (3) Sobel operator and Canny operator are used to process the seismic coherence volume sharpening image I1 to obtain the fault edge image I A ,I B , using the weighted method to process I A ,I B Generate a seismic coherence volume enhanced image I2.

[0012] 2. Adopt ant tracking technology based on adaptive form strategy:

[0013] (1) Set the number of ants in the ant tracking technology to m, the number of iterations to t, the objective function to f, the initial value of the pheromone concentration to c, the pheromone volatility factor to ρ, the mutation probability to q, the fault strike angle to θ, the search window to W, and the adaptive threshold to y. In I2, divide the data blocks according to the number of ants m, find the local extreme point in each data block as the seed point, and randomly select the seed point to deploy ants;

[0014] (2) Set the transfer value of ant tracking, which is obtained by weighting the image pixel value in I2 and the global pheromone. The ant defines the initial search window W according to the limited range of the fault strike angle θ, and determines whether there is a moving point in W according to the transfer value. If so, calculate the transfer probability of all moving points, randomly select moving points according to the roulette method, and calculate the moving path and the corresponding average transfer value y0, compare y0 with the adaptive threshold y, if y0 is larger, reduce the search window W within the window limit, if y0 is smaller, increase the search window W within the window limit; if it does not exist, it is considered that the fault line has been searched and the ant stops moving;

[0015] (3) When the number of iterations t exceeds a certain number, the ants can mutate according to q. During the search process, the mutated ants can find movable points beyond the limited range of the fault strike angle θ and plan the moving path according to the default minimum search window;

[0016] (4) After all ant tracking is completed, the global pheromone concentration c is updated according to the ant movement trajectory and the pheromone volatility factor ρ;

[0017] (5) Repeat steps (2)-(4) until the maximum number of iterations is met or the set threshold is reached, the ant tracking stops, and the result is output according to the global pheromone concentration c, which is the fine automatic identification of the fault.

[0018] The beneficial effects of the present invention are as follows: in view of the problem that conventional ant tracking technology has low accuracy in identifying faults, an adaptive window strategy is proposed for applying to ant tracking technology; when planning the movement trajectory of ant tracking according to the adaptive window strategy, the search window in complex fault zone areas is smaller, thereby improving the accuracy of fault description and making the movement trajectory more consistent with the actual trend of fault distribution; in simple fault zone areas, a relatively large search window is used to speed up fault identification efficiency; and an ant random mutation mechanism is introduced, whereby ants have a wider search space than ordinary ants under mutation conditions, and have a higher ability to track faults that are suitable for spatial characteristics, thereby increasing the global search accuracy of the ant tracking technology and improving the ability to search for hidden faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is the model structure diagram of the present invention DETAILED DESCRIPTION

[0020] Combine the following Figure 1 The present invention is described in further detail:

[0021] 1. Seismic coherence volume enhanced image acquisition:

[0022] (1) Obtaining original seismic coherence volume data from data sources such as exploration databases and seismic data volumes, and extracting slice data of typical traces from them as seismic coherence volume images I0;

[0023] (2) Extracting frequency domain features based on the spatial features of the seismic coherence volume image, applying the Basworth high-pass filter to intercept the high-frequency area of ​​the seismic coherence volume image, retaining a small amount of low-frequency area, reducing noise interference in the image, and enhancing the characteristics of the fault area to obtain the seismic coherence volume sharpened image I1;

[0024] (3) The Sobel operator, Canny operator and other edge extraction operators are used to process the seismic coherence volume sharpening image I1 to obtain the fault edge image I A ,I B Since the two operators have different extraction principles, the weighted method is used to process I A ,I B Generate a seismic coherence volume enhanced image I2.

[0025] 2. Adopt ant tracking technology based on adaptive form strategy:

[0026] (1) Set the number of ants in the ant tracking technology to m, the number of iterations to t, the objective function to f, the initial value of pheromone concentration to c, the pheromone volatilization factor to ρ, the mutation probability to q, the fault strike angle to θ, the search window to W, and the adaptive threshold to y. In I2, the data blocks are divided according to the number of ants m, and local extreme points are found as seed points in each data block. Ants are randomly selected to deploy seed points, so that the ants are distributed more evenly at the initial generation, thereby improving the overall search capability of the image;

[0027] (2) Set the transfer value of ant tracking. This function is obtained by weighting the image pixel value in I2 and the global pheromone. When calculating the transfer value, the ant needs to take into account the spatial characteristics of the fault and the previous generation search results. The ant defines the initial search window W according to the limited range of the fault strike angle θ, and determines whether there is a moving point in W based on the transfer value. If so, calculate the transfer probability of all moving points, randomly select moving points according to the roulette method, and calculate the moving path and the corresponding average transfer value y0. y0 represents the fitness value of the moving path. The higher the fitness value, the lower the fault search interference. The area may be more inclined to a simple fault zone. The lower the fitness value, the more likely the area is more inclined to a complex fault zone, and the search accuracy needs to be improved. Compare y0 with the adaptive threshold y. If y0 is larger, reduce the search window W within the window limit. If y0 is smaller, increase the search window W within the window limit. If it does not exist, it is considered that the fault line has been searched and the ant stops moving.

[0028] (3) When the number of iterations t exceeds a certain number, the ants can mutate according to q. During the search process, the mutated ants can find movable points beyond the limited range of the fault strike angle θ and plan the moving path according to the default minimum search window, thereby improving the search ability of a small number of ants and facilitating the identification of hidden faults.

[0029] (4) After all ants have finished tracking, pheromones are superimposed according to the ants’ movement trajectories, and the global pheromone concentration c is updated using the pheromone volatility factor ρ;

[0030] (5) Repeat steps (2)-(4) until the maximum number of iterations is met or the set threshold is reached, the ant tracking stops, and the result is output according to the global pheromone concentration c, which is the result of the fine automatic identification of the fault.

[0031] The above is only a preferred implementation case of the present invention. Any technician familiar with the profession may use the above-described technical solution to modify or change it into an equivalent example with equivalent changes. Any simple modification, change or modification of the above embodiment based on the technical solution of the invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the invention.

Claims

1. A method for fine fault identification based on ant tracking technology, characterized in that: The following steps are involved: The original seismic coherence volume data is obtained from the exploration database and seismic data volume, and the slice data of the typical track is extracted from it as the seismic coherence volume image. The high-frequency area in the seismic coherence volume image is intercepted by the Basworth high-pass filter and the edge extraction algorithm, and a small amount of low-frequency area is retained to obtain the seismic coherence volume enhanced image; the fault data blocks are divided in the seismic coherence volume enhanced image according to the number of ants, and the local extreme value points are found in each data block as the seed points for fault tracking. The seed points of random fault tracking are selected to deploy ants. The ants define the initial search window according to the fault strike angle range, and search for the table in the search window. The moving points indicating the direction of the fault are selected, the transfer probability of all moving points is calculated, and the moving points are randomly selected according to the roulette method while retaining the fitness value of the moving path. The search window is adjusted by the fitness value until the fault line is searched and the ants stop moving. When the number of iterations exceeds a certain number, the ants can mutate and obtain a stronger tracking ability for the fault. After all ant tracking is completed, pheromones are superimposed according to the ant's movement trajectory, and the global pheromone concentration is updated. Iterate until the maximum number of iterations is met or the set threshold is reached, and the ant tracking stops. The result of the ant tracking is the result of fine fault identification.

Images