Deep complex fracture identification method and device based on particle swarm, medium and equipment

Abstract

The invention relates to a deep complex fracture identification method and device based on a particle swarm, a medium and equipment. The method comprises the following steps: extracting a variance attribute body; setting initialization parameters of fracture identification; carrying out normalization on the variance attribute body; calculating the fitness of each particle; performing iterative calculation according to the initialization parameters, and determining an individual target fitness value and an individual target position; determining a population target fitness value and a population target position of the particle swarm based on the individual target fitness and the individual target position of all particles in the particle swarm; calculating a flight speed and a direction vector of each particle; executing fault marking; and judging whether the window circulation is completed or not, if not, updating and identifying the position coordinates of the window, and continuing the fracture identification of the next window, and if so, outputting a fracture identification result. According to the invention, fracture identification is carried out based on the particle swarm optimization, the curve of the best fitting fault can be searched in various fracture sensitive seismic attribute bodies, and global optimization is realized.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration technology, and more specifically, to a method, apparatus, medium, and equipment for identifying deep complex fractures based on particle swarm optimization. Background Technology

[0002] In seismic data interpretation, precise fault interpretation is crucial for understanding regional evolution, hydrocarbon migration, and trap identification. As exploration and development progresses into deeper strata, frequent deep tectonic activity and complex faults make manual fault interpretation time-consuming and highly subjective. Fault identification technology can effectively assist interpreters in fault interpretation, shortening the interpretation cycle and improving efficiency.

[0003] Current industrial fracture identification methods are based on ant colony algorithms, which use pheromones at grid points to calculate the probability of each ant moving to the next grid point. However, the search direction of ant colony algorithms only depends on the experience of individuals in the population, making it easy to get trapped in local optima and unable to obtain the global optimum. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, device, medium and equipment for identifying deep complex fractures based on particle swarm optimization, which addresses the problems existing in the prior art.

[0005] The technical solution adopted by this invention to solve its technical problem is: to construct a deep complex fracture identification method based on particle swarm optimization, comprising the following steps: Step S10: Extract data from the target seismic data to obtain the variance attribute volume of the target seismic data; Step S20: Set the initialization parameters for fracture detection; Step S30: Normalize the variance attribute volume to obtain a particle swarm; Step S40: Calculate the fitness of each particle in the particle swarm to obtain the fitness of each particle; Step S50: Perform iterative calculations based on the initialization parameters to determine the individual target fitness value and individual target position of each particle; Step S60: Based on the individual target fitness and individual target position of all particles in the particle swarm, determine the group target fitness value and group target position of the particle swarm; Step S70: Calculate the flight velocity and direction vector of each particle to update the position and direction values ​​of the particles; Step S80: Perform fault marking; Step S90: Determine whether the window loop has been completed. If not, update the position coordinates of the recognition window and return to step S30 to continue the break recognition of the next window. If completed, output the break recognition result.

[0006] In the particle swarm optimization method for identifying deep complex fractures described in this invention, the initialization parameters in step S20 include: identification window side length, maximum number of iterations, particle swarm size, learning factor, inertia coefficient, particle velocity, particle length, and population target fitness value.

[0007] In the particle swarm optimization method for identifying deep complex fractures described in this invention, step S30, which involves normalizing the variance attribute volume to obtain the particle swarm, includes: Obtain the recognition window from the variance attribute body; Extract the pixel values ​​of each grid point within the recognition window; Compare the pixel values ​​of the grid points with a set threshold; If the pixel value of the grid point is greater than the set threshold, the grid point is marked as processed; If the pixel value of the grid point is less than the set threshold, then a particle swarm is randomly initialized at the grid point.

[0008] In the particle swarm optimization method for identifying deep complex fractures described in this invention, step S50, which involves iteratively calculating based on the initialization parameters to determine the individual target fitness value and individual target position of each particle, includes: In the first round of iteration, the individual target fitness value and the corresponding individual target position of the particle are determined based on the fitness value calculated in step S40. After iterative calculations based on the initialization parameters, the subsequent iterative calculations are determined according to the following method: If the fitness value calculated in the current round is less than the individual target fitness value determined in the first round, then the currently calculated fitness value will be used as the individual target fitness value. If the fitness value calculated in the current round is greater than the individual target fitness value determined in the first round, then the individual target fitness value and the individual target position remain unchanged.

[0009] In the particle swarm optimization method for identifying deep complex fractures described in this invention, step S60, determining the swarm target fitness value and swarm target position based on the individual target fitness and individual target position of all particles in the particle swarm, includes: In each round of iterative calculation, the individual target fitness values ​​of all particles are compared; Select the minimum value based on the comparison results; The minimum value is taken as the population target fitness value of the particle swarm; The individual target position of the minimum value is taken as the group target position of the particle swarm.

[0010] In the particle swarm optimization-based deep complex fracture identification method of the present invention, step S80, performing fault marking includes: Determine whether a curve encoding a group of target particles exists; If present, the inclination angle of the target particle encoding curve of the swarm is calculated during identification on the profile. Fault marking is performed based on the dip angle.

[0011] In the particle swarm optimization-based method for identifying deep complex fractures described in this invention, the step of fault marking based on the dip angle includes: Compare the tilt angle with the set standard; If the tilt angle meets the set criteria, then all pixels on the target particle encoding curve of the population are marked as faults; If the tilt angle does not meet the set standard, all pixels within the identification area will be marked as processed.

[0012] The present invention also provides a particle swarm optimization-based device for identifying deep complex fractures, comprising: The data extraction unit is used to extract data from the target seismic data and obtain the variance attribute volume of the target seismic data; The parameter setting unit is used to set the initialization parameters for fracture recognition; A normalization processing unit is used to normalize the variance attribute volume to obtain a particle swarm. The fitness calculation unit is used to calculate the fitness of each particle in the particle swarm to obtain the fitness of each particle. An individual iterative calculation unit is used to perform iterative calculations based on the initialization parameters to determine the individual target fitness value and individual target position of each particle. A swarm update unit is used to determine the swarm target fitness value and the swarm target position based on the individual target fitness and individual target position of all particles in the swarm. The velocity and direction calculation unit is used to calculate the flight velocity and direction vector of each particle in order to update the particle's position and direction values; The fault identification unit is used to perform fault marking; The loop determination unit is used to determine whether the window loop has been completed. If it has not been completed, the recognition window position coordinates are updated and the break recognition of the next window continues. If it has been completed, the break recognition result is output.

[0013] The present invention also provides a storage medium storing a computer program adapted for loading by a processor to execute the steps of the particle swarm optimization method for identifying deep complex fractures as described above.

[0014] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the particle swarm-based deep complex fracture identification method as described above by calling the computer program stored in the memory.

[0015] The method, apparatus, medium, and device for identifying deep complex fractures based on particle swarm optimization (PSO) of this invention have the following beneficial effects: They include: extracting the variance attribute volume; setting initialization parameters for fracture identification; normalizing the variance attribute volume; calculating the fitness of each particle; performing iterative calculations based on the initialization parameters to determine the individual target fitness value and individual target position; determining the swarm target fitness value and swarm target position based on the individual target fitness and individual target positions of all particles in the particle swarm; calculating the flight velocity and direction vector of each particle; performing fault marking; determining whether the window loop has been completed; if not, updating the identification window position coordinates and continuing fracture identification for the next window; if completed, outputting the fracture identification result. This invention, based on the particle swarm optimization algorithm for fracture identification, can find the best-fitting fault curve among various fracture-sensitive seismic attribute volumes, achieving global optimization. Attached Figure Description

[0016] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart illustrating the deep complex fracture identification method based on particle swarm optimization provided by the present invention. Figure 2 This is a horizontal slice of the target seismic data provided by the present invention; Figure 3 This is a schematic diagram of the extracted variance attribute volume provided by the present invention; Figure 4 This is a schematic diagram of the particle encoding curve provided by the present invention; Figure 5 This is a diagram before constraints are added; Figure 6 This is a schematic diagram of the present invention after adding constraints; Figure 7 This is a diagram showing the fracture identification results provided by the present invention; Figure 8 This is a logic block diagram of the particle swarm optimization-based deep complex fracture identification device provided by the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] This invention is based on the Particle Swarm Optimization (PSO) algorithm, which encodes particles as curves to find the best-fitting fault curve among various fault-sensitive seismic attribute volumes. By adjusting the particle length, the dimension of the solution space can be directly determined, thus effectively distinguishing faults of different scales. Furthermore, the search direction of the PSO algorithm is constrained by both individual experience and population experience, making it easier to achieve a global optimum compared to the Ant Colony Optimization (ACO) algorithm.

[0019] refer to Figure 1 In a preferred embodiment, the particle swarm optimization method for identifying deep complex fractures provided by the present invention may include steps S10, S20, S30, S40, S50, S60, S70, S80 and S90.

[0020] Step S10: Extract data from the target seismic data to obtain the variance attribute volume of the target seismic data.

[0021] In this step, for example Figure 2 The variance attribute of the seismic data (i.e., the target seismic data) that needs to be identified for fracture identification is extracted to obtain its variance attribute volume, as shown in the figure below. Figure 3 As shown. Existing methods can be used to extract the variance attribute; this invention does not impose specific limitations.

[0022] Step S20: Set the initialization parameters for fracture recognition.

[0023] In some embodiments, before fracture detection, initialization parameters for fracture detection need to be set. These initialization parameters include, but are not limited to: recognition window side length, maximum number of iterations, particle swarm size, learning factor, inertia coefficient, particle velocity, particle length, and population target fitness value.

[0024] Step S30: Normalize the variance attribute volume to obtain the particle swarm.

[0025] In some embodiments, step S30, normalizing the variance attribute volume to obtain a particle swarm includes: obtaining a recognition window in the variance attribute volume; extracting the pixel value of a grid point within each recognition window; comparing the pixel value of the grid point with a set threshold; if the pixel value of the grid point is greater than the set threshold, marking the grid point as processed; if the pixel value of the grid point is less than the set threshold, randomly initializing the particle swarm at the grid point.

[0026] Specifically, after completing the initial parameter settings, fracture identification can be performed. In this step, the variance attribute volume extracted in step S10 is normalized, thus normalizing the pixel values ​​of all grid points to between 0 and 1. For example, taking a certain identification window as an example, the pixel values ​​of all grid points within the window are read and compared with a set threshold. Grid points with pixel values ​​greater than the set threshold are then marked as processed, while particle swarms are randomly initialized on grid points with pixel values ​​less than the set threshold.

[0027] For example, such as Figure 4 As shown, the red dashed box represents the recognition window. The curve formed by the particles extends from the center point to both ends, controlled by the particle's center point position and direction of movement. The coordinates of the particle's center point (point P) represent the individual position of the particle. Figure 4 In the example, the particle's individual position is represented as [6, 4], and its direction of movement is represented as: [0,0,7,0,0,0,7,0,4,3,4,4,5,4,4,3]. The numbers 0 to 7 are directional values, representing the direction of the particle's movement from the previous point to the next point.

[0028] Step S40: Calculate the fitness of each particle in the particle swarm to obtain the fitness of each particle.

[0029] Specifically, in this step, the fitness of each particle can be calculated using a formula: ; Where L is the length of the particle, b is the gain coefficient of the sigmoid function, c is the point of maximum gain, and Avg(p) is used to measure the averageness of the particle curve p, and its specific calculation formula is as follows: ; Among them, Mag i Uni(p) is the pixel value of the i-th grid point on the particle curve p. Uni(p) measures the uniformity of the particle curve p, and its specific calculation formula is: ; Cur(p) is used to measure the curvature of the particle curve p, and its specific calculation formula is as follows: ; Where, m i This represents the value in the i-th direction of the particle curve.

[0030] Step S50: Perform iterative calculations based on the initialization parameters to determine the individual target fitness value and individual target position of each particle.

[0031] In some embodiments, step S50, performing iterative calculations based on initialization parameters to determine the individual target fitness value and individual target position of each particle includes: in the first round of iteration, determining the individual target fitness value and corresponding individual target position of the particle based on the fitness value calculated in step S40; after performing iterative calculations based on initialization parameters, determining the individual target fitness value and individual target position in subsequent iterations in the following manner: if the fitness value calculated in the current round is less than the individual target fitness value determined in the first round, then the currently calculated fitness value is used as the individual target fitness value; if the fitness value calculated in the current round is greater than the individual target fitness value determined in the first round, then the individual target fitness value and individual target position remain unchanged.

[0032] This step is primarily used to update the individual target fitness value and individual target position of particles with better fitness values. The individual target fitness value can also be called the individual optimal fitness value, and correspondingly, the individual target position can be called the individual optimal position.

[0033] In some embodiments, updating the individual target fitness value and individual target position of a particle with a better fitness value specifically involves the following steps: In the first iteration, the fitness value calculated for the first time according to the particle fitness calculation formula in step S40 is the particle's individual best fitness, and its individual position is the individual best position. In subsequent iterations, if the fitness calculated in the current round is less than the individual best fitness, the individual best fitness is updated, that is, the fitness calculated in the current round is used as the individual best fitness, and the individual position in the current round is used as the individual best position; if the fitness calculated in the current round is greater than the individual best fitness, the individual best fitness and individual best position remain unchanged.

[0034] Step S60: Based on the individual target fitness and individual target position of all particles in the particle swarm, determine the group target fitness value and group target position of the particle swarm.

[0035] In some embodiments, step S60, determining the swarm target fitness value and swarm target position of the particle swarm based on the individual target fitness and individual target position of all particles in the particle swarm, includes: comparing the individual target fitness values ​​of all particles in each round of iterative calculation; selecting the minimum value according to the comparison result; using the minimum value as the swarm target fitness value of the particle swarm; and using the individual target position of the minimum value as the swarm target position of the particle swarm.

[0036] Specifically, this step mainly corrects the population's best fitness value and population's best position based on the current best fitness value within the population. That is, in each iteration, the individual target fitness values ​​of all particles are compared, and the smallest one is the population's best fitness value (i.e., the population's target fitness value), and its individual best position is the population's best position (i.e., the population's target position).

[0037] Step S70: Calculate the flight velocity and direction vector of each particle to update the position and direction values ​​of the particles.

[0038] In some embodiments, the flight velocity and direction vector of each particle are calculated using the following formula: ; ; in, This represents the velocity vector of particle i in the kth generation; This represents the position vector of particle i in the kth generation; This represents the optimal individual position vector of particle i in the kth generation; This represents the global optimal position vector, and c1 and c2 are learning factors. and The parameter is random, taking a value randomly between [0, 1], which can give the particle a certain degree of randomness; w As a linearly decreasing inertia weight, this parameter allows particles to have different search ranges at different stages. Particles have a larger exploration range in the early stages of iteration, while particles have higher exploration accuracy in the later stages of iteration.

[0039] Linearly decreasing inertia weight w The specific calculation formula is as follows: in, It is the inertial weight used in each iteration. The initial inertia weight values ​​are determined. To achieve the maximum number of iterations, The maximum allowed number of iterations, This represents the current iteration number.

[0040] The position and orientation values ​​of the particles are updated by this formula, and a new generation of loops is performed. The algorithm is then run in step S40 until the maximum number of iterations is reached. The process is repeated for the next particle until all grid points in the recognition window that meet the predetermined threshold conditions are marked as processed.

[0041] Step S80: Perform fault marking.

[0042] In some embodiments, step S80, performing fault marking, includes: determining whether a swarm target particle encoding curve exists; if it exists, calculating the inclination angle of the swarm target particle encoding curve during identification on the profile; and performing fault marking based on the inclination angle. Specifically, performing fault marking based on the inclination angle includes: comparing the inclination angle with a set standard; if the inclination angle meets the set standard, marking all pixels on the swarm target particle encoding curve as faults; if the inclination angle does not meet the set standard, marking all pixels within the identification area as processed. The swarm target particle encoding curve can also be referred to as the swarm optimal particle encoding curve.

[0043] Specifically, after marking all grid points within the identification window in step S70, it is determined whether a curve representing the optimal particle encoding for the population exists. If it does, it is then determined whether the curve conforms to the set criteria (i.e., the tilt angle criterion) during identification on the profile. If identification is performed on a plane, it is not necessary to determine whether the tilt angle criterion is met. The specific formula for calculating the tilt angle is as follows: ; in, The tilt angle is calculated using the particle curve (i.e., the curve representing the optimal particle encoding of the swarm). The minimum and maximum inclination angles are represented by [30°, 90°], which vary depending on the region. Generally, they can be taken as [30°, 90°], but can be adjusted according to the actual situation. k is the slope of the linear function fitted to the particle curve, specifically expressed as: Where x and y are the coordinates of the grid points where the particle curve is located, and c is the constant term of the fitted linear function.

[0044] If the curve of the optimal particle coding for the swarm conforms to the tilt angle criterion, then all pixels on that curve are marked as faults. If there is no optimal particle coding curve for the swarm, or if it exists but does not conform to the tilt angle criterion, then all pixels within the recognition area are marked as processed.

[0045] Step S90: Determine whether the window loop has been completed. If not, update the position coordinates of the recognition window and return to step S30 to continue the break recognition of the next window. If completed, output the break recognition result.

[0046] This invention presents a deep complex fracture identification method based on particle swarm optimization. It employs particle swarm optimization and encodes the particles in the particle swarm algorithm as curves, enabling effective identification of complex fractures at different scales. Figure 5 and Figure 6 A comparison is given before and after adding constraints. Figure 7The fracture identification results of this invention are shown. It can be seen that this invention can effectively distinguish fractures of different scales. Furthermore, the search direction of the particle swarm optimization algorithm is constrained by both individual experience and population experience, making it easier to achieve a global optimum compared to the ant colony optimization algorithm.

[0047] refer to Figure 8 The present invention also provides a deep complex fracture identification device based on particle swarm optimization.

[0048] like Figure 8 As shown, the particle swarm optimization-based deep complex fracture identification device includes: The data extraction unit 801 is used to extract data from the target seismic data and obtain the variance attribute volume of the target seismic data.

[0049] The parameter setting unit 802 is used to set the initialization parameters for fracture recognition.

[0050] The normalization processing unit 803 is used to normalize the variance attribute volume to obtain the particle swarm.

[0051] The fitness calculation unit 804 is used to calculate the fitness of each particle in the particle swarm and obtain the fitness of each particle.

[0052] The individual iterative calculation unit 805 is used to perform iterative calculations based on the initialization parameters to determine the individual target fitness value and individual target position of each particle.

[0053] The swarm update unit 806 is used to determine the swarm target fitness value and swarm target position based on the individual target fitness and individual target position of all particles in the swarm.

[0054] The velocity and direction calculation unit 807 is used to calculate the flight velocity and direction vector of each particle in order to update the position and direction values ​​of the particles.

[0055] The fault identification unit 808 is used to perform fault marking.

[0056] The loop determination unit 809 is used to determine whether the window loop has been completed. If it has not been completed, the position coordinates of the recognition window are updated and the break recognition of the next window continues. If it has been completed, the break recognition result is output.

[0057] Specifically, the specific operational process of the cooperation between the units in the particle swarm-based deep complex fracture identification device can be referred to the aforementioned particle swarm-based deep complex fracture identification method, and will not be repeated here.

[0058] Furthermore, an electronic device of the present invention includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the particle swarm optimization-based deep complex fracture identification method as described above. Specifically, according to embodiments of the present invention, the processes described above with reference to the flowchart can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, when the computer program is downloaded, installed, and executed by an electronic device, it performs the functions defined in the methods of the embodiments of the present invention. The electronic device of the present invention can be a terminal such as a laptop, desktop computer, tablet computer, or smartphone, or it can be a server.

[0059] Furthermore, one type of storage medium of the present invention stores a computer program thereon, which, when executed by a processor, implements the particle swarm optimization-based deep complex fracture identification method described above. Specifically, it should be noted that the storage medium described above in the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein computer-readable program code is carried. The transmitted data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0060] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0061] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0062] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0063] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0064] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They do not limit the scope of protection of the present invention. All equivalent changes and modifications made within the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A method for identifying deep, complex fractures based on particle swarm optimization, characterized in that, Includes the following steps: step S10: Extract data from the target seismic data to obtain the variance attribute volume of the target seismic data; Step S20: Set the initialization parameters for fracture detection; Step S30: Normalize the variance attribute volume to obtain a particle swarm; Step S40: Calculate the fitness of each particle in the particle swarm to obtain the fitness of each particle; Step S50: Perform iterative calculations based on the initialization parameters to determine the individual target fitness value and individual target position of each particle; Step S60: Based on the individual target fitness and individual target position of all particles in the particle swarm, determine the group target fitness value and group target position of the particle swarm; Step S70: Calculate the flight velocity and direction vector of each particle to update the position and direction values ​​of the particles; Step S80: Perform fault marking; Step S90: Determine whether the window loop has been completed. If not, update the position coordinates of the recognition window and return to step S30 to continue the break recognition of the next window. If completed, output the break recognition result.

2. The method for identifying deep complex fractures based on particle swarm optimization according to claim 1, characterized in that, In step S20, the initialization parameters include: recognition window side length, maximum number of iterations, particle swarm size, learning factor, inertia coefficient, particle velocity, particle length, and population target fitness value.

3. The method for identifying deep complex fractures based on particle swarm optimization according to claim 1, characterized in that, In step S30, normalizing the variance attribute volume to obtain the particle swarm includes: Obtain the recognition window from the variance attribute body; Extract the pixel values ​​of each grid point within the recognition window; Compare the pixel values ​​of the grid points with a set threshold; If the pixel value of the grid point is greater than the set threshold, the grid point is marked as processed; If the pixel value of the grid point is less than the set threshold, then a particle swarm is randomly initialized at the grid point.

4. The method for identifying deep complex fractures based on particle swarm optimization according to claim 1, characterized in that, In step S50, the iterative calculation based on the initialization parameters to determine the individual target fitness value and individual target position of each particle includes: In the first round of iteration, the individual target fitness value and the corresponding individual target position of the particle are determined based on the fitness value calculated in step S40. After iterative calculations based on the initialization parameters, the subsequent iterative calculations are determined according to the following method: If the fitness value calculated in the current round is less than the individual target fitness value determined in the first round, then the currently calculated fitness value will be used as the individual target fitness value. If the fitness value calculated in the current round is greater than the individual target fitness value determined in the first round, then the individual target fitness value and the individual target position remain unchanged.

5. The method for identifying deep complex fractures based on particle swarm optimization according to claim 1, characterized in that, In step S60, determining the swarm target fitness value and swarm target position based on the individual target fitness and individual target position of all particles in the particle swarm includes: In each round of iterative calculation, the individual target fitness values ​​of all particles are compared; Select the minimum value based on the comparison results; The minimum value is taken as the population target fitness value of the particle swarm; The individual target position of the minimum value is taken as the group target position of the particle swarm.

6. The method for identifying deep complex fractures based on particle swarm optimization according to claim 1, characterized in that, In step S80, performing fault marking includes: Determine whether a curve encoding a group of target particles exists; If present, the inclination angle of the target particle encoding curve of the swarm is calculated during identification on the profile. Fault marking is performed based on the dip angle.

7. The method for identifying deep complex fractures based on particle swarm optimization according to claim 6, characterized in that, The fault marking based on the dip angle includes: Compare the tilt angle with the set standard; If the tilt angle meets the set criteria, then all pixels on the target particle encoding curve of the population are marked as faults; If the tilt angle does not meet the set standard, all pixels within the identification area will be marked as processed.

8. A particle swarm optimization-based device for identifying deep complex fractures, characterized in that, include: The data extraction unit is used to extract data from the target seismic data and obtain the variance attribute volume of the target seismic data; The parameter setting unit is used to set the initialization parameters for fracture recognition; A normalization processing unit is used to normalize the variance attribute volume to obtain a particle swarm. The fitness calculation unit is used to calculate the fitness of each particle in the particle swarm to obtain the fitness of each particle. An individual iterative calculation unit is used to perform iterative calculations based on the initialization parameters to determine the individual target fitness value and individual target position of each particle. A swarm update unit is used to determine the swarm target fitness value and the swarm target position based on the individual target fitness and individual target position of all particles in the swarm. The velocity and direction calculation unit is used to calculate the flight velocity and direction vector of each particle in order to update the particle's position and direction values; The fault identification unit is used to perform fault marking; The loop determination unit is used to determine whether the window loop has been completed. If it has not been completed, the recognition window position coordinates are updated and the break recognition of the next window continues. If it has been completed, the break recognition result is output.

9. A storage medium, characterized in that, The storage medium stores a computer program adapted for loading by a processor to perform the steps of the particle swarm-based deep complex fracture identification method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the particle swarm-based deep complex fracture identification method as described in any one of claims 1 to 7 by calling the computer program stored in the memory.

Images