Rock mass structural surface advantage grouping method, medium and equipment

Through the improved K-mean algorithm and particle swarm optimization algorithm, combined with structural plane inclination angle and tendency, the global optimal problem caused by local optimization of the K-mean algorithm is solved, and the global optimality and high accuracy of the dominant grouping of rock mass structure surfaces is achieved.

CN119089231APending Publication Date: 2024-12-06河南省地质矿产勘查开发局第四地质勘查院
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411179117.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

When the existing K-mean algorithm is a local optimization algorithm when grouping the structural surface of the rock mass, the final grouping results cannot reach global optimality, and it is impossible to accurately understand the development characteristics of the structural surface in the rock mass.

Method used

By using the improved K-mean algorithm, the objective function is established by obtaining the inclination angle and tendency of the structural plane, and the objective function is constrained by the distance from the structural plane to the cluster center, and the optimal position is updated using the particle swarm optimization algorithm to determine the global optimal particle, combined with Structural plane tendency and inclination data are grouped.

Benefits of technology

The global optimization of the structural surface advantage grouping results is achieved, the accuracy and consistency of the grouping results are improved, and the error rate is reduced, so that the structural surface grouping results are more in line with the actual development characteristics of the rock mass.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119089231A_ABST
    Figure CN119089231A_ABST
Patent Text Reader

Abstract

The invention discloses a rock mass structural plane advantage grouping method, medium and equipment, and relates to the technical field of geotechnical engineering, and the method comprises the steps: obtaining the inclination angle and the inclination of the occurrence of a structural plane; according to the inclination angle and the inclination direction of the structural plane occurrence, establishing a target function of structural plane occurrence grouping with the distance from the structural plane to the clustering center as a constraint; according to the objective function, performing optimal position updating through an improved K-means algorithm, and determining a global optimal particle; and according to the globally optimal particles and in combination with the inclination angles and the tendencies of the occurrence of the structural surfaces, dividing the structural surfaces with similar inclination angle data into one group, and determining a structural surface advantage grouping result. According to the method, the K-means method is optimized by adopting the particle swarm optimization algorithm, so that the structural surface advantage grouping result is globally optimal, and the accuracy of structural surface advantage grouping is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of geotechnical engineering, and in particular to a method, medium and equipment for advantageous grouping of rock mass structural surfaces. Background Art

[0002] Affected by tectonic movement, a large number of structural planes are developed in the rock mass. The development, distribution characteristics and mechanical properties of the structural planes in the rock mass have an important impact on the stability of the project. When studying the structural characteristics of the rock mass, it is necessary to divide the dominant groups of the rock mass structural planes. Grouping the rock mass structural planes can clearly understand the development characteristics of different occurrences of cracks in the rock mass, which is of great significance for understanding the rock mass structure.

[0003] In engineering, the K-means algorithm is usually used to perform structural surface advantage grouping. However, the K-means method is a local optimization algorithm. When this method is used for grouping, the final grouping result is often the local optimum, making it impossible for the structural surface advantage grouping result to reach the global optimum. Summary of the invention

[0004] Based on this, it is necessary to address the above technical issues. The present invention provides a method for advantageously grouping rock mass structural surfaces. The method comprises the following steps:

[0005] Obtain the dip and inclination of the structural surface;

[0006] According to the dip and inclination of the structural surface, an objective function for grouping the structural surface with the distance from the structural surface to the cluster center as a constraint is established;

[0007] According to the objective function, the optimal position is updated by the improved K-means algorithm to determine the global optimal particle; wherein the improved K-means algorithm is as follows: the parameters and the particle swarm are initialized, the position and velocity of the particle are updated according to the current position and velocity of the particle, and the position of the global optimal and individual optimal particles, the fitness of each particle is evaluated, and the particle position and velocity are updated and the particle fitness is evaluated repeatedly until a certain stopping condition is met to determine the global optimal particle;

[0008] According to the global optimal particle and combined with the dip and inclination of the structural surface, the data with similar dip and inclination are grouped together to determine the structural surface advantage grouping results.

[0009] Optionally, the parameters and particle swarm are initialized, specifically including:

[0010] Parameter initialization includes: setting initial values ​​for learning factors μ1, μ2, iteration weight w and total number of particles N in the particle swarm; wherein the total number of particles N in the particle swarm is the number of structural surfaces;

[0011] The initialization of the particle swarm includes: dividing the discontinuity surface into K groups and obtaining the cluster center of each group using the eigenvalue analysis method;

[0012] The initial position and velocity of the particle are determined according to the cluster center of the obtained structural surface.

[0013] Optionally, the updating of the position and velocity of the particle specifically includes:

[0014] The speed and position of the new particle are obtained using the following formula:

[0015] v id =wv id +μ1r1(p id -x id )+μ2r2(p gd -x id )

[0016] x id =x id +v id

[0017] Among them, d is the search direction, w is the iteration weight, v is the speed of particle movement, μ1 and μ2 are learning factors, and r1 and r2 are random variables in the interval [0,1].

[0018] Optionally, the evaluating the fitness of each particle specifically includes:

[0019] The fitness of the particle is obtained using the following formula:

[0020]

[0021] Among them, f is the fitness of the particle, d(e i ,c j ) represents the i-th discontinuity surface e i To the jth cluster center c j The distance, S j is the jth group of structural surfaces;

[0022] Determine the particle's historical optimal position P i The fitness f(P i ) is greater than the current position X i The fitness f(X i ), if so, update the optimal position P of the corresponding particle i , let P i =X i , otherwise keep the particle's historical optimal position P i ;

[0023] Determine the optimal position P of the particle swarm gThe fitness f(P g ) is greater than the particle's optimal position P i The fitness f(P i ), if so, update the optimal position P of the particle swarm g , let P g =P i , otherwise keep the optimal position P of the particle swarm g .

[0024] The present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned rock mass structural surface advantage grouping method is implemented.

[0025] The present invention provides a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned rock mass structural surface advantage grouping method when executing the program.

[0026] Compared with the prior art, the present invention has the following beneficial effects: the present invention provides a method for grouping rock mass structural plane advantages, which obtains the inclination and dip of the rock mass structural plane, adopts a particle swarm optimization algorithm, and optimizes the learning factors μ1, μ2, and iterative weights w min 、w max The total number of particles N in the particle swarm is initialized, and the position and velocity of the particle are updated according to the current position and velocity of the particle, as well as the position of the global optimal and individual optimal particles. The fitness of each particle is evaluated based on the inclination and tendency of the structural surface occurrence, and the particle position and velocity are updated, as well as the particle fitness evaluation is repeated until a certain stopping condition is met, and the global optimal particle is determined, thereby obtaining the structural surface advantage grouping result, so that the structural surface grouping result reaches the global optimal. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0028] Figure 1 A flowchart of a method for advantageously grouping rock mass structural surfaces provided by an embodiment of the present invention;

[0029] Figure 2 A pole diagram of structural surface data provided by an embodiment of the present invention;

[0030] Figure 3 A schematic diagram of grouping results of the K-means algorithm provided in an embodiment of the present invention;

[0031] Figure 4 A schematic diagram of grouping results of the algorithm of the present invention provided in an embodiment of the present invention;

[0032] Figure 5 A schematic diagram of a computer device for a method for advantageously grouping rock mass structural surfaces provided in an embodiment of the present invention.

[0033] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] The development and distribution characteristics of structural surfaces in rock mass have an important impact on the stability of the project. When studying the structural characteristics of rock mass, it is necessary to divide the rock mass structural surfaces into dominant groups, that is, to divide the structural surfaces with similar occurrences into the same group, so as to understand the development characteristics of the structural surfaces in the rock mass.

[0036] In engineering practice, the K-means algorithm is used to perform structural surface advantage grouping. However, this method is a local optimization algorithm, and the final grouping result is often the local optimal one, and it is impossible to obtain the global optimal grouping result.

[0037] In view of the above problems, the present invention provides a method for advantageous grouping of rock structural surfaces. The present invention adopts a particle swarm optimization algorithm to optimize the K-means method, and then uses the optimized algorithm to divide structural surfaces with similar occurrences into the same group, thereby achieving global optimal advantageous grouping of rock structural surfaces.

[0038] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0039] Example 1

[0040] Figure 1 is a flow chart of a method for advantageously grouping rock mass structural surfaces provided by an embodiment of the present invention, such as Figure 1 As shown, this embodiment shows a method for advantageously grouping rock mass structural surfaces, including:

[0041] S101: Obtain the inclination and dip of the structural surface.

[0042] S102: According to the dip angle and inclination of the structural surface, an objective function for grouping the structural surface with the distance from the structural surface to the cluster center as a constraint is established.

[0043] S103: According to the objective function, the optimal position is updated by the improved K-means algorithm to determine the global optimal particle; wherein the improved K-means algorithm is as follows: the parameters and the particle swarm are initialized, the position and velocity of the particle are updated according to the current position and velocity of the particle, and the position of the global optimal and individual optimal particles, the fitness of each particle is evaluated, and the particle position and velocity are updated and the particle fitness is evaluated repeatedly until a certain stop condition is met to determine the global optimal particle.

[0044] Optionally, initial values ​​are set for the learning factors μ1, μ2, the iteration weight w, and the total number of particles N in the particle swarm; wherein the total number of particles N in the particle swarm is the number of structural surfaces; the discontinuous surfaces are divided into K groups, and the cluster center of each group is obtained by the eigenvalue analysis method; and the initial position and speed of the particles are determined according to the obtained cluster center of the structural surface.

[0045] Optionally, use the following formula to get the new particle's speed and position:

[0046] v id =wv id +μ1r1(p id -x id )+μ2r2(p gd -x id )

[0047] x id =x id +v id

[0048] Among them, d is the search direction, w is the iteration weight, v is the speed of particle movement, μ1 and μ2 are learning factors, and r1 and r2 are random variables in the interval [0,1].

[0049] The fitness of the particle is obtained using the following formula:

[0050]

[0051] Among them, f is the fitness of the particle, d(e i ,c j ) represents the i-th discontinuity surface e i To the jth cluster center c j The distance, S j is the jth group of structural surfaces;

[0052] Determine the particle's historical optimal position P i The fitness f(P i ) is greater than the current position X i The fitness f(X i ), if so, update the optimal position P of the corresponding particle i , let P i =X i , otherwise keep the particle's historical optimal position P i ;

[0053] Determine the optimal position P of the particle swarm g The fitness f(P g ) is greater than the particle's optimal position P i The fitness f(P i ), if so, update the optimal position P of the particle swarm g , let P g =P i , otherwise keep the optimal position P of the particle swarm g .

[0054] S104: According to the global optimal particle, particles with similar inclination and dip angle data are grouped together to determine the structural surface advantage grouping result.

[0055] Example 2

[0056] This embodiment uses a computer simulation method to randomly generate 5 groups of structural surface data with different distribution forms and different discreteness. For the convenience of visualization, the randomly generated structural surface is represented by an upper hemisphere equal area map projection as shown in FIG. Figure 2 shown.

[0057] The structural advantage grouping is explained in detail below.

[0058] S201: Generate initial cluster centers using Monte Carlo simulation technology.

[0059] The Monte Carlo simulation technique is used to randomly generate initial cluster centers, which can avoid interference from human factors.

[0060] S202: Setting initial parameter values.

[0061] Optionally, take learning factors μ1=2, μ2=2, iteration weight w=2, and the total number of particles in the particle swarm N=360.

[0062] S203: Dominantly group the rock mass structural surfaces.

[0063] The rock mass structural surfaces are advantageously grouped according to the above steps S101 to S104, which will not be described in detail here.

[0064] S204: The iteration ends and the structural surface advantage grouping result is obtained.

[0065] After the iteration is completed, the classification results of the inclination and dip angle are obtained. Table 1 is a comparison of the grouping results of the algorithm of the present invention and the K-means algorithm.

[0066] Table 1 Comparison of grouping results between the algorithm of the present invention and the K-means algorithm

[0067]

[0068] Table 1 shows that the error rate of the K-means algorithm grouping is 29.7%, while the error rate of the algorithm of the present invention grouping is only 6.7%. It can be seen that the classification result of the algorithm of the present invention is not affected by the initial cluster center, and the accuracy of the grouping result is higher.

[0069] like Figure 3 and Figure 4 As shown, Figure 3 and Figure 4 The grouping results of the K-means algorithm and the algorithm of the present invention are shown. Figure 3 and Figure 4 It can be seen that when the randomly generated initial cluster centers are used, the K-means algorithm mistakenly classifies some cracks that originally belonged to the fourth group into the third group; the grouping results of the algorithm of the present invention are highly consistent with the pre-given grouping results. Compared with the traditional K-means algorithm, the algorithm of the present invention has higher accuracy in performing advantageous grouping of rock mass structural surfaces.

[0070] The present embodiment provides a method for grouping rock mass structural plane advantages. By obtaining the inclination and dip of the rock mass structural plane, the Monte Carlo simulation technology is used to generate the initial cluster center, and the particle swarm optimization algorithm is used to optimize the learning factors μ1, μ2, and the iterative weight w. min 、w max The total number of particles N in the particle swarm is initialized, and the position and velocity of the particle are updated according to the current position and velocity of the particle, as well as the position of the global optimal and individual optimal particles. The fitness of each particle is evaluated based on the inclination and tilt of the structural surface. The particle position and velocity are updated, as well as the particle fitness evaluation is repeated, and iterative optimization is performed until the stopping condition is met to determine the global optimal particle, thereby obtaining the structural surface advantage grouping result. The grouping result is more in line with the development characteristics of the structural surface in the rock mass, so that the structural surface grouping result reaches the global optimal.

[0071] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program can be used to execute the rock mass structural surface advantage grouping method provided above.

[0072] The present invention also provides Figure 5The structural diagram of the computer device shown in FIG. Figure 5 As shown, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the rock mass structural surface advantage grouping method provided in the above embodiment.

[0073] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0074] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present invention.

Claims

1. A method for grouping rock mass structural planes. It is characterized in that include: Obtain the dip and inclination of the structural surface; According to the dip and inclination of the structural surface, an objective function for grouping the structural surface with the distance from the structural surface to the cluster center as a constraint is established; According to the objective function, the optimal position is updated by the improved K-means algorithm to determine the global optimal particle; wherein the improved K-means algorithm is as follows: the parameters and the particle swarm are initialized, the position and velocity of the particle are updated according to the current position and velocity of the particle, and the position of the global optimal and individual optimal particles, the fitness of each particle is evaluated, and the particle position and velocity are updated and the particle fitness is evaluated repeatedly until a certain stopping condition is met to determine the global optimal particle; According to the global optimal particle and combined with the dip and inclination of the structural surface, the data with similar dip and inclination are grouped together to determine the structural surface advantage grouping results.

2. The method for grouping rock mass structural surfaces according to claim 1, It is characterized in that The parameters and particle swarm are initialized, specifically including: Parameter initialization includes: learning factor μ 1 , μ 2 , the iteration weight w and the total number of particles N in the particle swarm are set to initial values; wherein the total number of particles N in the particle swarm is the number of structural surfaces; The initialization of the particle swarm includes: dividing the discontinuity surface into K groups and obtaining the cluster center of each group using the eigenvalue analysis method; The initial position and velocity of the particle are determined according to the cluster center of the obtained structural surface.

3. The method for grouping rock mass structural surfaces according to claim 1, It is characterized in that The updating of the particle position and velocity specifically includes: The speed and position of the new particle are obtained using the following formula: v id =wv id +μ 1 r 1 (p id -x id )+μ 2 r 2 (p gd -x id ) x id =x id +v id Among them, d is the search direction, w is the iteration weight, v is the speed of particle movement, μ 1 , μ 2 is the learning factor, r 1 、r 2 are random variables with an interval between [0,1].

4. The method for grouping rock mass structural surfaces in a dominant manner as claimed in claim 1, It is characterized in that The fitness of each particle is evaluated, specifically including: The fitness of the particle is obtained using the following formula: Among them, f is the fitness of the particle, d(e i ,c j ) represents the i-th discontinuity surface e i To the jth cluster center c j The distance, S j is the jth group of structural surfaces; Determine the particle's historical optimal position P i The fitness f(P i ) is greater than the current position X i The fitness f(X i ), if so, update the optimal position P of the corresponding particle i , let P i =X i , otherwise keep the particle's historical optimal position P i ; Determine the optimal position P of the particle swarm g The fitness f(P g ) is greater than the particle's optimal position P i The fitness f(P i ), if so, update the optimal position P of the particle swarm g , let P g =P i , otherwise keep the optimal position P of the particle swarm g .

5. A computer-readable storage medium, It is characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the method for advantageous grouping of rock mass structural surfaces as described in any one of claims 1 to 4 is implemented.

6. A computer device, It is characterized in that The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for advantageous grouping of rock mass structural surfaces as described in any one of claims 1 to 4 is implemented.