A cluster analysis method based on geological and engineering parameters
By obtaining the drilling lattice geological engineering parameter vector, calculating the degree of clustering influence, and obtaining the clustering parameter vector, the problem of low clustering efficiency of high-dimensional geological engineering parameters is solved, and fast and accurate geological information analysis is achieved.
Patent Information
- Application Number
- CN202411000725.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-07-25
AI Technical Summary
In the prior art, the cluster analysis and calculation of high-dimensional geological engineering parameters is large and time-consuming, which leads to low efficiency and makes it difficult to quickly and accurately analyze the geological information of mineral resource exploration areas.
By obtaining the geological engineering parameter vectors in the drilling lattice, constructing a parameter sequence set, calculating the degree of clustering influence, obtaining the cluster parameter vector, and finally performing cluster analysis to improve efficiency.
While ensuring the reliability of clustering results, the efficiency and speed of clustering of geological engineering parameters is significantly improved, and the analysis of geological information in mineral resource exploration areas is simplified.
Smart Images

Figure CN118535942B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of parameter cluster analysis, and in particular to a cluster analysis method based on geological and engineering parameters. Background Art
[0002] At present, in the process of mineral resource exploration, it is usually necessary to conduct drilling sampling and sample testing, and obtain multidimensional geological engineering parameters, and the geological engineering parameters are important indicators for describing the characteristics of geological bodies and engineering environments. In addition, since the obtained multidimensional geological engineering parameters are usually clustered and then the geological information of the surveyed area is analyzed based on the clustering results, the geological information can reflect the existence of mineral resources in the surveyed area. Therefore, it is very important to cluster the geological engineering parameters.
[0003] In the prior art, the collected multidimensional geological engineering parameters are usually clustered directly using traditional clustering algorithms, such as using the k-means clustering algorithm to directly cluster the collected multidimensional geological engineering parameters. However, due to the high dimensionality of the collected geological engineering parameters, although the k-means clustering algorithm can achieve the purpose of clustering, the k-means clustering algorithm has a large amount of calculation, a long calculation time, and a low data processing efficiency in the process of clustering the collected multidimensional geological engineering parameters directly. That is, the k-means clustering algorithm is used to directly cluster the collected multidimensional geological engineering parameters, which will result in a low clustering speed or efficiency. Therefore, it is an urgent problem to ensure that the geological information of the surveyed area can be analyzed more reliably based on the clustering results while improving the clustering efficiency. Summary of the invention
[0004] In order to solve the above problems, the present invention provides a cluster analysis method based on geological and engineering parameters. The technical solutions adopted are as follows:
[0005] An embodiment of the present invention provides a cluster analysis method based on geological and engineering parameters, comprising the following steps:
[0006] Obtaining a geological engineering parameter vector corresponding to each borehole in a borehole lattice corresponding to a target area, wherein the geological engineering parameter vector is composed of A geological engineering parameters;
[0007] According to the geological engineering parameter vector corresponding to each borehole, a parameter sequence set is obtained;
[0008] Clustering the geotechnical parameters in each parameter sequence in the parameter sequence set to obtain each cluster cluster corresponding to each parameter sequence, and obtaining the clustering influence degree corresponding to each parameter sequence according to each parameter sequence and the geotechnical parameters in each cluster cluster corresponding to each parameter sequence;
[0009] According to the clustering influence degree, obtaining the clustering parameter vector corresponding to each of the boreholes;
[0010] A cluster of geological engineering parameter vectors is obtained according to the cluster parameter vectors.
[0011] Preferably, the method for obtaining the parameter sequence set includes:
[0012] Among all the geological engineering parameter vectors, the sequence constructed by all the geological engineering parameters at the same position is recorded as a parameter sequence, and the geological engineering parameters in the parameter sequence are arranged in order from small to large; the sequence set constructed by all the obtained parameter sequences is recorded as a parameter sequence set.
[0013] Preferably, the method for obtaining the clustering influence degree corresponding to each parameter sequence includes:
[0014] For any parameter sequence B in the parameter sequence set:
[0015] Obtaining a first difference sequence corresponding to the parameter sequence B, wherein the bth first difference value in the first difference sequence is the absolute value of the difference between the b+1th geological engineering parameter in the parameter sequence B and the bth geological engineering parameter in the parameter sequence B; recording a normalized value of the variance of the first difference sequence as a characteristic distribution representation value corresponding to the parameter sequence B;
[0016] The maximum geoengineering parameter and the minimum geoengineering parameter in each cluster corresponding to the parameter sequence B are recorded as the maximum parameter value and the minimum parameter value corresponding to the corresponding cluster respectively; all clusters corresponding to the parameter sequence B are sorted in ascending order of the maximum parameter value to obtain the cluster sequence B0 corresponding to the parameter sequence B; the parameter difference sequence corresponding to the cluster sequence B0 is obtained, and the cth parameter difference value in the parameter difference sequence is the absolute value of the difference between the minimum parameter value corresponding to the c+1th cluster in the cluster sequence B0 and the maximum parameter value corresponding to the cth cluster in the cluster sequence B0; the sum of all parameter differences in the parameter difference sequence is recorded as the comprehensive interval characterization value corresponding to the parameter sequence B;
[0017] According to the geological engineering parameters in each cluster in the cluster cluster sequence B0, the intra-cluster aggregation characterization value corresponding to the parameter sequence B is obtained; the normalized value of the product of the intra-cluster aggregation characterization value and the comprehensive interval characterization value is recorded as the block aggregation distribution characterization value corresponding to the geological engineering parameter sequence B; the normalized value of the sum of the block aggregation distribution characterization value and the characteristic distribution characterization value is recorded as the clustering influence degree corresponding to the parameter sequence B.
[0018] Preferably, the method for obtaining the intra-cluster aggregation characterization value corresponding to the parameter sequence B includes:
[0019] For any cluster D in the cluster sequence B0: sort all the geotechnical parameters in the cluster D in ascending order and record them as the parameter subsequence corresponding to the cluster D; obtain the second difference sequence corresponding to the parameter subsequence, the d-th second difference in the second difference sequence is the absolute value of the difference between the d+1-th geotechnical parameter in the parameter subsequence and the d-th geotechnical parameter in the parameter subsequence; record the sum of all the second differences in the second difference sequence as the intra-cluster difference characterization value corresponding to the cluster D;
[0020] The inverse of the mean of the intra-cluster difference characterization values corresponding to all clusters in the cluster sequence B0 is used as the intra-cluster aggregation characterization value corresponding to the parameter sequence B.
[0021] Preferably, according to the clustering influence degree, the method for obtaining the clustering parameter vector corresponding to each borehole includes:
[0022] All the sequences to be determined in the parameter sequence set are obtained, and a set constructed by all the sequences to be determined is recorded as a set of sequences to be determined, wherein the sequences to be determined are parameter sequences whose clustering influence is not less than a preset first threshold; according to the cluster cluster sequences corresponding to the sequences to be determined, a set of connected borehole domains and a set of discrete boreholes corresponding to each of the sequences to be determined are obtained; according to the set of connected borehole domains and the set of discrete boreholes corresponding to each of the sequences to be determined, a characterization value of borehole aggregation distribution corresponding to each of the sequences to be determined is obtained;
[0023] All the sequences to be determined in the set of sequences to be determined are arranged and combined to obtain all the combinations of sequences to be determined, and a set constructed by all the combinations of sequences to be determined is recorded as a feature set, wherein the number of the sequences to be determined in the set of sequences to be determined is a preset number value;
[0024] According to the borehole connected domain set corresponding to each to-be-determined sequence in each to-be-determined sequence combination in the feature set, a connected domain superposition characterization value corresponding to each to-be-determined sequence in each to-be-determined sequence combination is obtained; and the cumulative sum of the connected domain superposition characterization values corresponding to each to-be-determined sequence in each to-be-determined sequence combination is recorded as the comprehensive superposition characterization value corresponding to the corresponding to-be-determined sequence combination;
[0025] For any sequence combination J to be determined in the feature set, the normalized value of the cumulative sum of the borehole aggregation distribution characterization values corresponding to all the sequences to be determined in the sequence combination J to be determined is recorded as the comprehensive aggregation distribution characterization value corresponding to the sequence combination J to be determined, and the sum of the normalized value of the comprehensive superposition characterization value corresponding to the sequence combination J to be determined and the comprehensive aggregation distribution characterization value is recorded as the preference degree characterization value corresponding to the sequence combination J to be determined;
[0026] Among all the to-be-determined sequence combinations, the to-be-determined sequence combination corresponding to the maximum preferred degree characterization value is selected and recorded as the target sequence set; in each to-be-determined sequence in the target sequence set, a parameter vector constructed by all the geological engineering parameters of the geological engineering parameter vector corresponding to each borehole is recorded as the clustering parameter vector corresponding to the corresponding borehole.
[0027] Preferably, the method for obtaining the borehole connected domain set and the discrete borehole set corresponding to each sequence to be determined includes:
[0028] Obtaining a neighboring borehole set corresponding to each borehole; all neighboring boreholes in the neighboring borehole set corresponding to each borehole belong to the eight-neighborhood of the corresponding borehole;
[0029] For the fth sequence to be determined in the set of sequences to be determined:
[0030] For any borehole h0 in the borehole lattice, if the hth geotechnical parameter in the gth cluster in the cluster sequence corresponding to the fth sequence to be determined belongs to the geotechnical parameter vector corresponding to the borehole h0, g is used as the fth label value corresponding to the borehole h0; according to the fth label value corresponding to each borehole, all borehole connected domains corresponding to the fth sequence to be determined are obtained; and the set constructed by all borehole connected domains corresponding to the fth sequence to be determined is recorded as the borehole connected domain set corresponding to the fth sequence to be determined;
[0031] For any borehole h0 in the borehole lattice, if the f-th label value corresponding to all neighboring boreholes in the neighboring borehole set corresponding to the borehole h0 is different from the f-th label value corresponding to the borehole h0, then the borehole h0 is recorded as the discrete borehole corresponding to the f-th sequence to be determined; and the set constructed by all discrete boreholes corresponding to the f-th sequence to be determined is recorded as the discrete borehole set corresponding to the f-th sequence to be determined.
[0032] Preferably, the method for obtaining the drilling hole aggregation distribution characterization value corresponding to each sequence to be determined includes:
[0033] For the fth sequence to be determined:
[0034] The sum of the total number of discrete boreholes in the discrete borehole set corresponding to the f-th sequence to be determined and a preset first constant is recorded as the discrete characterization value corresponding to the f-th sequence to be determined; the difference between the total number of boreholes in the borehole lattice and the discrete characterization value is recorded as the characteristic difference value, and the normalized value of the ratio of the characteristic difference value to the discrete characterization value is recorded as the first characterization value corresponding to the f-th sequence to be determined;
[0035] For any borehole connected domain S1 in the borehole connected domain set F0 corresponding to the f-th sequence to be determined: the square value of the total number of boreholes located in the borehole connected domain S1 is recorded as the area representation value corresponding to the borehole connected domain S1; for the j-th borehole located in the borehole connected domain S1, in the neighboring borehole set corresponding to the j-th borehole, the number of neighboring boreholes located in the borehole connected domain S1 is statistically obtained, and recorded as the first quantity representation value corresponding to the j-th borehole; the cumulative sum of the first quantity representation values corresponding to all boreholes in the borehole connected domain S1 is recorded as the comprehensive quantity representation value corresponding to the borehole connected domain S1;
[0036] The normalized value of the cumulative sum of area characterization values corresponding to all the borehole connected domains in the borehole connected domain set F0 is recorded as the second characterization value corresponding to the f-th sequence to be determined; the normalized value of the cumulative sum of comprehensive quantity characterization values corresponding to all the borehole connected domains in the borehole connected domain set F0 is recorded as the third characterization value corresponding to the f-th sequence to be determined; the sum of the first characterization value, the second characterization value and the third characterization value corresponding to the f-th sequence to be determined is recorded as the borehole aggregation distribution characterization value corresponding to the f-th sequence to be determined.
[0037] Preferably, the method for obtaining the connected domain superposition characterization value corresponding to each to-be-determined sequence in each to-be-determined sequence combination includes:
[0038] For any sequence L to be determined in the sequence combination J to be determined:
[0039] In the to-be-determined sequence combination J, all the remaining to-be-determined sequences except the to-be-determined sequence L are recorded as first sequences corresponding to the to-be-determined sequence L, and a set constructed by all first sequences corresponding to the to-be-determined sequence L is recorded as a first sequence set corresponding to the to-be-determined sequence L; according to the borehole connected domain set corresponding to each first sequence in the first sequence set, a first eigenvalue set corresponding to each borehole connected domain in the borehole connected domain set corresponding to the to-be-determined sequence L is obtained;
[0040] Determine whether the set of characteristic quantity characterization values corresponding to each borehole connected domain in the borehole connected domain set corresponding to the sequence L to be determined is an empty set; if not, record the corresponding borehole connected domain as the characteristic connected domain corresponding to the sequence L to be determined; record the set constructed by all characteristic connected domains corresponding to the sequence L to be determined as the characteristic connected domain set corresponding to the sequence L to be determined;
[0041] The second eigenvalue set corresponding to each feature connected domain in the feature connected domain set is obtained, and the cumulative sum of all second eigenvalues in the second eigenvalue set corresponding to each feature connected domain is recorded as the superimposed eigenvalue corresponding to the corresponding feature connected domain; the cumulative sum of the superimposed eigenvalues corresponding to all feature connected domains in the feature connected domain set is recorded as the connected domain superimposed representation value corresponding to the sequence L to be determined; the rth second eigenvalue in the second eigenvalue set corresponding to the qth feature connected domain in the feature connected domain set is the reciprocal of the rth first eigenvalue in the first eigenvalue set corresponding to the qth feature connected domain.
[0042] Preferably, the method for obtaining the first eigenvalue set corresponding to each borehole connected domain in the borehole connected domain set corresponding to the sequence L to be determined includes:
[0043] For the zth borehole connected domain in the set of borehole connected domains corresponding to the sequence L to be determined: obtain the set of eigenvalues to be analyzed corresponding to the zth borehole connected domain, the tth eigenvalue to be analyzed in the set of eigenvalues to be analyzed is the number of borehole connected domains in the set of borehole connected domains corresponding to the tth first sequence in the first sequence set that have an intersection with the zth borehole connected domain; record the set constructed by all eigenvalues to be analyzed whose values are not 0 in the set of eigenvalues to be analyzed as the first eigenvalue set corresponding to the zth borehole connected domain.
[0044] Preferably, the method for obtaining the clustering cluster of the geological engineering parameter vector according to the clustering parameter vector includes:
[0045] A Q-dimensional space is constructed, where Q is the total number of sequences in the target sequence set; all the geological engineering parameters in the clustering parameter vector corresponding to each borehole in the borehole lattice are mapped to the Q-dimensional space, and the data points obtained after mapping are recorded as the data points to be clustered corresponding to the corresponding boreholes; all the data points to be clustered are clustered, and each cluster cluster obtained after clustering is recorded as a feature cluster cluster, and the cluster cluster formed by the geological engineering parameter vectors corresponding to all the boreholes in each feature cluster cluster is recorded as the geological engineering parameter vector cluster cluster corresponding to the corresponding feature cluster cluster.
[0046] Beneficial effects: The present invention first obtains the geological engineering parameter vector corresponding to each borehole in the borehole lattice corresponding to the target area; then obtains a parameter sequence set according to the geological engineering parameter vector corresponding to each borehole; then clusters the geological engineering parameters in each parameter sequence in the parameter sequence set to obtain each cluster cluster corresponding to each parameter sequence, and obtains the clustering influence degree corresponding to each parameter sequence according to each parameter sequence and the geological engineering parameters in each cluster cluster corresponding to each parameter sequence; then obtains the clustering parameter vector corresponding to each borehole according to the clustering influence degree; finally obtains the geological engineering parameter vector cluster cluster according to the clustering parameter vector. The geological engineering parameter vector cluster cluster obtained by the present invention based on the clustering parameter vector corresponding to each borehole can ensure that the purpose of more reliable analysis of the geological information of the surveyed area is achieved based on the clustering result of the geological engineering parameter vector, while also improving the efficiency or rate of obtaining the geological engineering parameter vector cluster cluster. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0048] Figure 1 The present invention is a flow chart of a cluster analysis method based on geological and engineering parameters. DETAILED DESCRIPTION
[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the embodiments of the present invention.
[0050] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0051] This embodiment provides a cluster analysis method based on geological and engineering parameters, which is described in detail as follows:
[0052] like Figure 1 As shown, the cluster analysis method based on geological and engineering parameters includes the following steps:
[0053] Step S001, obtaining a geological engineering parameter vector corresponding to each borehole in a borehole array corresponding to a target area.
[0054] In order to facilitate subsequent analysis, this embodiment selects any mineral source survey area as an example for analysis and records it as the target area; then the relevant staff determines the location where drilling sampling is required in the target area, and conducts drilling sampling to obtain the core sample corresponding to each borehole; since the purpose of drilling sampling in this embodiment is to survey the mineral source, and when conducting mineral source survey, drilling is usually performed in the form of a dot matrix, that is, the boreholes in the target area in this embodiment are arranged in a dot matrix, that is, the drilling dot matrix corresponding to the target area can be obtained through the above process.
[0055] Then, the core samples corresponding to each borehole in the borehole lattice are detected and collected, specifically: for any borehole h0 in the borehole lattice, the geological engineering parameters of the core sample corresponding to the borehole h0 are detected and collected using the detection equipment, and all the collected geological engineering parameters are recorded as the geological engineering parameters corresponding to the borehole h0, and the vector constructed by all the geological engineering parameters corresponding to the borehole h0 is recorded as the geological engineering parameter vector corresponding to the borehole h0, and in this embodiment, the geological engineering parameters of A dimensions of the core sample corresponding to the borehole h0 can be collected, that is, the geological engineering parameter vector corresponding to the borehole h0 is composed of A geological engineering parameters; and in this embodiment, the process of detecting and collecting the geological engineering parameters of the core sample corresponding to the borehole h0 using the detection equipment is a well-known technology, so this embodiment will not be described in detail.
[0056] Therefore, according to the above method of obtaining the geological engineering parameter vector corresponding to the borehole h0, the geological engineering parameter vectors corresponding to all the boreholes in the target area are obtained, and the types and numbers of the geological engineering parameters in all the geological engineering parameter vectors in this embodiment are the same, that is, the geological engineering parameter vectors corresponding to all the boreholes are composed of A different types of geological engineering parameters; in addition, the geological engineering parameters in this embodiment include at least the particle size of the core sample, the density of the core sample, the porosity of the core sample, the saturation of the core sample, the permeability of the core sample, the suction of the core sample, the hydraulic gradient of the core sample, and the shear strength of the core sample. , the compressive strength of the core sample, the acoustic wave velocity of the core sample, the pore water pH value of the core sample and the pore water ion concentration of the core sample. Therefore, the value of A in this embodiment is 12, that is, all geological engineering parameter vectors are composed of A different types of geological engineering parameters, and the parameter types at the same position in each geological engineering parameter vector are consistent. For example, if the pth geological engineering parameter in the geological engineering parameter vector corresponding to the borehole h0 refers to the density of the core sample corresponding to the borehole h0, then the pth geological engineering parameter in the geological engineering parameter vector corresponding to the remaining boreholes in the borehole lattice also refers to the density of the core sample corresponding to the corresponding borehole.
[0057] In addition, as other implementation methods, different A can also be set, for example, A can be set to 6, and the geological engineering parameters constituting the geological engineering parameter vector include the density of the core sample, the porosity of the core sample, the saturation of the core sample, the permeability of the core sample, the hydraulic gradient of the core sample, and the compressive strength of the core sample.
[0058] Therefore, the geological engineering parameter vector corresponding to each borehole in the borehole lattice can be obtained by the above method.
[0059] Step S002: obtaining a parameter sequence set according to the geological engineering parameter vector corresponding to each borehole; clustering the geological engineering parameters in each parameter sequence in the parameter sequence set to obtain each cluster cluster corresponding to each parameter sequence; obtaining the cluster influence degree corresponding to each parameter sequence according to each parameter sequence and the geological engineering parameters in each cluster cluster corresponding to each parameter sequence.
[0060] Since the number of parameters in the geological engineering parameter vector corresponding to each borehole obtained above is large, that is, the dimension of the geological engineering parameters corresponding to each borehole obtained is high. When the dimension is high, if the geological engineering parameter vector is directly clustered, although the clustering purpose can be achieved in the end, due to the high dimension of the geological engineering parameter vector, the amount of calculation in the clustering process of the geological engineering parameter vector will be large, the calculation time will be long, and the data processing efficiency will be low. That is, using the k-means clustering algorithm to cluster the geological engineering parameter vectors corresponding to the obtained boreholes will result in low clustering speed or efficiency. In order to improve the clustering speed or efficiency, this embodiment will subsequently analyze the geological engineering parameter vector to obtain a parameter sequence in a single dimension, and then analyze the influence of the parameter sequence in a single dimension on the clustering result obtained when the geological engineering parameter vector is directly clustered, so as to screen the parameter sequences in multiple dimensions, and obtain the clustering parameter vectors corresponding to each borehole, and then cluster the corresponding boreholes. The parameter vector is used to determine and obtain the clustering clusters of each geological engineering parameter vector, and the dimension of the clustering parameter vector obtained in this embodiment is smaller than the dimension of the geological engineering parameter vector. Therefore, compared with directly clustering the geological parameter vectors, the subsequent clustering of the clustering parameter vectors has a smaller amount of calculation, a shorter calculation time, and a higher data processing efficiency. Therefore, relatively speaking, the clustering speed or efficiency of clustering the clustering parameter vectors is faster, and the clustering clusters of the geological engineering parameter vectors determined based on the clustering parameter vectors will not affect the analysis of the survey results of the target area, nor will it affect the acquisition of geological survey information of the target area. The difference between the clustering clusters of the geological engineering parameter vectors determined based on the clustering parameter vectors and the clustering clusters of the geological engineering parameter vectors determined directly based on the geological engineering parameter vectors is small, that is, the geological survey results of the target area analyzed by the clustering clusters of the geological engineering parameter vectors determined based on the clustering parameter vectors and the clustering clusters of the geological engineering parameter vectors determined directly based on the geological engineering parameter vectors are basically consistent.
[0061] Therefore, in order to avoid the problem of low clustering speed when directly clustering the geological engineering parameter vectors and to ensure the reliability and accuracy of the subsequently obtained geological engineering parameter vector clusters, this embodiment will next obtain a parameter sequence set based on the geological engineering parameter vector corresponding to each borehole. The parameter sequence set is the basis for the subsequent determination of the clustering parameter vector. Therefore, the specific method for obtaining the parameter sequence set is as follows:
[0062] The set constructed by the geo-engineering parameter vectors corresponding to all the boreholes is recorded as a vector set; the a0th geo-engineering parameter in each geo-engineering parameter vector of the vector set is obtained, and the sequence constructed by the a0th geo-engineering parameter in each geo-engineering parameter vector is recorded as the a0th parameter sequence corresponding to the vector set, and the geo-engineering parameters in the a0th parameter sequence are arranged in ascending order, and the parameter type of the a0th geo-engineering parameter in each geo-engineering parameter vector is the same, that is, the types of all parameters in the a0th parameter sequence are consistent.
[0063] Therefore, through the above process, all parameter sequences corresponding to the vector set can be obtained, and the number of all parameter sequences corresponding to the vector set is the same as the total number of parameters in the geological engineering parameter vector, and the total number of parameters in the parameter sequence is the same as the total number of geological engineering parameter vectors in the vector set; then the sequence set constructed by all parameter sequences corresponding to the vector set is recorded as the parameter sequence set.
[0064] Therefore, a parameter sequence set can be obtained through the above process.
[0065] Next, this embodiment uses the K-means clustering algorithm to cluster the geological engineering parameters in each parameter sequence in the parameter sequence set, and obtains each cluster cluster corresponding to each parameter sequence and the geological engineering parameters in each cluster. When clustering each parameter sequence using the K-means clustering algorithm, this embodiment can determine the optimal K value corresponding to each parameter sequence by the elbow method or the silhouette coefficient method, where K is the number of cluster centers.
[0066] Then, according to each parameter sequence in the parameter sequence set and the geological engineering parameters in each cluster cluster corresponding to each parameter sequence, the clustering influence degree corresponding to each parameter sequence is obtained, and the clustering influence degree corresponding to each parameter sequence can reflect the influence degree of the corresponding parameter sequence on the overall clustering result, and the greater the clustering influence degree, the more important the corresponding parameter sequence is for the subsequent acquisition or analysis of geological survey information in the target area, and the overall clustering result refers to the clustering result obtained by directly clustering the geological engineering parameter vector, and the clustering influence degree is an important indicator for the subsequent determination of the clustering parameter vector corresponding to each borehole, so the specific process of obtaining the clustering influence degree in this embodiment is:
[0067] For any parameter sequence B in the parameter sequence set:
[0068] The first difference sequence corresponding to the parameter sequence B is obtained, and the bth first difference in the first difference sequence is the absolute value of the difference between the b+1th parameter in the parameter sequence B and the bth geological engineering parameter in the parameter sequence B; then the normalized value of the variance of the first difference sequence is obtained and recorded as the characteristic distribution representation value corresponding to the parameter sequence B, and the characteristic distribution representation value can reflect the uniformity of the distribution of the parameter sequence B, and is also an important indicator for subsequently determining the degree of clustering influence.
[0069] Then, the maximum and minimum geotechnical parameters in each cluster corresponding to parameter sequence B are obtained, and recorded as the maximum parameter value and the minimum parameter value corresponding to the corresponding cluster respectively; then, all clusters corresponding to parameter sequence B are sorted in order from small to large according to the maximum parameter value, and the obtained sequence is recorded as cluster sequence B0 corresponding to parameter sequence B. Then, the parameter difference sequence corresponding to cluster sequence B0 is obtained, and the cth parameter difference in the parameter difference sequence is the absolute value of the difference between the minimum parameter value corresponding to the c+1th cluster in cluster sequence B0 and the maximum parameter value corresponding to the cth cluster in cluster sequence B0.
[0070] Then, the sum of all parameter differences in the parameter difference sequence is obtained and recorded as the comprehensive interval characterization value corresponding to the parameter sequence B. Then, according to the geological engineering parameters in each cluster in the cluster cluster sequence B0, the intra-cluster aggregation characterization value corresponding to the parameter sequence B is obtained. Both the comprehensive interval characterization value and the intra-cluster aggregation characterization value can reflect whether the parameter sequence B presents a block-like concentrated distribution. Then, the normalized value of the product of the intra-cluster aggregation characterization value corresponding to the parameter sequence B and its corresponding comprehensive interval characterization value is obtained, and recorded as the block-like aggregation distribution characterization value corresponding to the geological engineering parameter sequence B. The block-like aggregation distribution characterization value is an important indicator for determining the degree of clustering influence in the subsequent process.
[0071] In addition, in this embodiment, according to the geological engineering parameters in each cluster in the cluster sequence B0, the specific process of obtaining the cluster clustering characterization value corresponding to the parameter sequence B is as follows:
[0072] For any cluster D in the cluster sequence B0: sort all the geotechnical parameters in the cluster D in ascending order, and record the sorted sequence as the parameter subsequence corresponding to the cluster D; then obtain the second difference sequence corresponding to the parameter subsequence, and the d-th second difference in the second difference sequence is the absolute value of the difference between the d+1-th geotechnical parameter in the parameter subsequence and the d-th geotechnical parameter in the parameter subsequence; then obtain the sum of all the second differences in the second difference sequence, and record it as the intra-cluster difference characterization value corresponding to the cluster D.
[0073] Therefore, through the above process, the intra-cluster difference characterization values corresponding to all clusters in the cluster sequence B0 can be obtained, and then the inverse of the mean of the intra-cluster difference characterization values corresponding to all clusters in the cluster sequence B0 is used as the intra-cluster aggregation characterization value corresponding to the parameter sequence B.
[0074] Therefore, through the above acquisition, the cluster block aggregation distribution representation value corresponding to parameter sequence B and its corresponding feature distribution representation value are obtained, and then the normalized value of the sum of the cluster block aggregation distribution representation value corresponding to parameter sequence B and its corresponding feature distribution representation value is obtained, and recorded as the clustering influence degree corresponding to parameter sequence B.
[0075] In addition, in this embodiment, the specific expression for obtaining the clustering influence degree corresponding to the parameter sequence B is:
[0076]
[0077] Among them, R0 is the clustering influence degree corresponding to the parameter sequence B, is the variance of the first difference sequence corresponding to the parameter sequence B, norm() is the normalization function, is the characteristic distribution representation value corresponding to the parameter sequence B, N0 is the total number of clusters in the cluster sequence B0 corresponding to the parameter sequence B, is the intra-cluster difference representation value corresponding to the nth cluster in the cluster sequence B0, The reciprocal of is the clustered representation value corresponding to the parameter sequence B. is the comprehensive interval characterization value corresponding to the parameter sequence B, N1 is the total number of parameter differences in the parameter difference sequence corresponding to the cluster sequence B0, is the ath parameter difference in the parameter difference sequence corresponding to the cluster sequence B0.
[0078] In this embodiment, when The larger the value, The smaller the value of The larger the value of is, the larger the value of R0 is. The smaller the value of The larger the value, the The smaller the value of is, the smaller the value of R0 is, and after normalization, the value range of R0 is 0 to 1.
[0079] Because when the parameter distribution in the parameter sequence is more uniform, it indicates that the geological engineering parameters in the parameter sequence have less directionality when participating in clustering. The directionality refers to the ability to more clearly determine which cluster the geological engineering parameters belong to during clustering, which also indicates that the corresponding parameter sequence has a smaller influence on the overall clustering result, and the overall clustering result refers to the clustering result obtained by directly clustering all geological engineering parameter vectors; and because when the parameter sequence presents a more blocky clustered distribution, it indicates that the geological engineering parameters in the parameter sequence have more directionality when participating in clustering, which also indicates that the corresponding parameter sequence has a greater influence on the overall clustering result; and It can reflect the uniformity of parameter distribution in parameter sequence B. and It can reflect whether the parameters in parameter sequence B are distributed in a block-like manner. The smaller the value of is, the more uniform the distribution of parameters in parameter sequence B is. The smaller the value of The larger the value of , the more the parameters in parameter sequence B are distributed in a block-like manner. It can be seen that when the value of R0 is larger, the corresponding parameter sequence B has a greater influence on the overall clustering result, and when the value of R0 is smaller, the corresponding parameter sequence B has a smaller influence on the overall clustering result.
[0080] Therefore, this embodiment can obtain the clustering influence degree corresponding to each parameter sequence in the parameter sequence set through the above process, and the clustering influence degree can reflect the importance of the corresponding parameter sequence for the subsequent acquisition or analysis of geological survey information of the target area.
[0081] Step S003: acquiring cluster parameter vectors corresponding to the respective boreholes according to the cluster influence degree.
[0082] Next, based on the clustering influence degree corresponding to the parameter sequence obtained in S002, the present embodiment obtains the clustering parameter vector corresponding to each borehole. The clustering parameter vector is the basis for subsequent clustering to obtain the final geological engineering parameter vector cluster cluster. Therefore, the specific method for obtaining the clustering parameter vector corresponding to each borehole in the present embodiment is as follows:
[0083] First, a preset first threshold is obtained, and in a specific application, the implementer may set the preset first threshold according to actual conditions. For example, in this embodiment, the preset first threshold is set to 0.5.
[0084] According to the clustering influence degree corresponding to each parameter sequence and the preset first threshold, all the sequences to be determined in the parameter sequence set are obtained. The obtaining method is: determine whether the clustering influence degree corresponding to each parameter sequence is not less than the preset first threshold. If so, the corresponding parameter sequence is recorded as the sequence to be determined; and in this embodiment, the clustering influence degree corresponding to the parameter sequence other than the sequence to be determined is relatively small, so when performing clustering subsequently, not referring to the parameters in the parameter sequence with a smaller clustering influence degree will not make the reliability and accuracy of the subsequently determined geological engineering parameter vector cluster cluster low, so this embodiment only needs to analyze the sequence to be determined later.
[0085] Since the number of the sequences to be determined obtained above may be large, if the clustering clusters of the geological engineering parameter vectors are determined based on all the geological engineering parameters in all the sequences to be determined, there will still be a problem of low clustering rate. Therefore, in order to avoid the problem of low clustering rate, this embodiment will further analyze the sequences to be determined obtained above to obtain the clustering parameter vectors corresponding to each borehole. The dimension of the clustering parameter vector is much smaller than the dimension of the geological engineering parameter vector. Therefore, the specific process of obtaining the clustering parameter vector based on the sequence to be determined is as follows:
[0086] First, according to each corresponding cluster sequence to be determined in the set of sequences to be determined, the borehole connected domain set and discrete borehole set corresponding to each sequence to be determined are obtained. The specific process of obtaining the borehole connected domain set and the discrete borehole set is as follows:
[0087] First, the neighboring borehole set corresponding to each borehole in the borehole dot matrix is obtained, and all neighboring boreholes in the neighboring borehole set corresponding to each borehole in the borehole dot matrix belong to the eight-neighborhood of the corresponding borehole, and the boreholes in the neighboring borehole set are recorded as neighboring boreholes; since the boreholes are arranged in a dot matrix, the boreholes in the dot matrix can be regarded as pixel points.
[0088] For the fth sequence to be determined in the set of sequences to be determined:
[0089] First, the f-th marking value corresponding to each borehole in the borehole lattice is obtained according to each cluster in the cluster sequence corresponding to the f-th sequence to be determined. The specific acquisition method is: for any borehole h0 in the borehole lattice, if the h-th geological engineering parameter in the g-th cluster in the cluster sequence corresponding to the f-th sequence to be determined belongs to the geological engineering parameter vector corresponding to the borehole h0, then g is used as the f-th marking value corresponding to the borehole h0, that is, the f-th marking value corresponding to the boreholes corresponding to the geological engineering parameters belonging to the same cluster is the same.
[0090] Then, according to the f-th marking value corresponding to each borehole in the borehole lattice, all borehole connected domains corresponding to the f-th sequence to be determined are obtained, and the set constructed by all borehole connected domains corresponding to the f-th sequence to be determined is recorded as the set of borehole connected domains corresponding to the f-th sequence to be determined; and the f-th marking value corresponding to all boreholes in any borehole connected domain S0 corresponding to the f-th sequence to be determined is the same; in addition, in the borehole lattice, the f-th marking value corresponding to the borehole adjacent to the borehole connected domain S0 and not belonging to the borehole connected domain S0 is different from the f-th marking value corresponding to the borehole in the borehole connected domain S0.
[0091] For any borehole h0 in the borehole lattice, if the f-th label value corresponding to all neighboring boreholes in the neighboring borehole set corresponding to borehole h0 is different from the f-th label value corresponding to borehole h0, then the borehole h0 at this time is determined to be a discrete borehole, that is, the borehole h0 is recorded as the discrete borehole corresponding to the f-th sequence to be determined; therefore, through the above process, all discrete boreholes corresponding to the f-th sequence to be determined can be obtained, and the set constructed by all discrete boreholes corresponding to the f-th sequence to be determined is recorded as the discrete borehole set corresponding to the f-th sequence to be determined.
[0092] Therefore, through the above process, the connected domain set of boreholes and the discrete borehole set corresponding to each sequence to be determined can be obtained, and the connected domain set of boreholes and the discrete borehole set corresponding to the sequence to be determined can characterize the distribution of boreholes under the parameter dimension corresponding to the sequence to be determined, and the parameter types in each sequence to be determined or parameter sequence are consistent, so one sequence to be determined or one parameter sequence can characterize one parameter dimension.
[0093] Then, according to the borehole connected domain set and the discrete borehole set corresponding to each sequence to be determined, the borehole aggregation distribution characterization value corresponding to each sequence to be determined is obtained, and the specific acquisition process of the borehole aggregation distribution characterization value is:
[0094] For the fth sequence to be determined:
[0095] First, the sum of the total number of discrete boreholes in the discrete borehole set corresponding to the f-th sequence to be determined and a preset first constant is recorded as the discrete characterization value corresponding to the f-th sequence to be determined; then, the difference between the total number of boreholes in the borehole lattice and the discrete characterization value corresponding to the f-th sequence to be determined is obtained and recorded as the characteristic difference value, and the normalized value of the ratio of the characteristic difference value corresponding to the f-th sequence to be determined and the corresponding discrete characterization value is recorded as the first characterization value corresponding to the f-th sequence to be determined, and the characteristic difference value is also the total number of boreholes located in the connected domain of boreholes corresponding to the f-th sequence to be determined.
[0096] In a specific application, the implementer needs to set the preset first constant according to the actual situation. For example, in this embodiment, the preset first constant is set to constant 1.
[0097] For any borehole connected domain S1 in the borehole connected domain set F0 corresponding to the f-th sequence to be determined: obtain the total number of boreholes located in the borehole connected domain S1, and record the square value of the total number of boreholes located in the borehole connected domain S1 as the area representation value corresponding to the borehole connected domain S1, and then obtain the normalized value of the cumulative sum of the area representation values corresponding to all borehole connected domains in the borehole connected domain set F0 and record it as the second representation value corresponding to the f-th sequence to be determined; for the j-th borehole located in the borehole connected domain S1, in the neighboring borehole set corresponding to the j-th borehole, the number of neighboring boreholes located in the borehole connected domain S1 is statistically obtained, and recorded as the first quantity representation value corresponding to the j-th borehole; then obtain the cumulative sum of the first quantity representation values corresponding to all boreholes in the borehole connected domain S1, and record it as the comprehensive quantity representation value corresponding to the borehole connected domain S1.
[0098] Therefore, through the above process, the comprehensive quantitative characterization value corresponding to all the borehole connected domains in the borehole connected domain set F0 can be obtained, and then the normalized value of the cumulative sum of the comprehensive quantitative characterization values corresponding to all the borehole connected domains in the borehole connected domain set F0 can be obtained, and recorded as the third characterization value corresponding to the fth sequence to be determined.
[0099] Finally, the sum of the first characterization value, the second characterization value, and the third characterization value corresponding to the f-th sequence to be determined is obtained and recorded as the borehole aggregation distribution characterization value corresponding to the f-th sequence to be determined; the first characterization value, the second characterization value, and the third characterization value can all reflect the degree of aggregation of the boreholes under the parameter dimension corresponding to the f-th sequence to be determined; in addition, in this embodiment, the specific expression for obtaining the borehole aggregation distribution characterization value corresponding to the f-th sequence to be determined is:
[0100]
[0101] Among them, R1 is the drilling aggregation distribution characterization value corresponding to the fth sequence to be determined, is the difference between the total number of holes in the drilling lattice and the discrete representation value corresponding to the fth sequence to be determined, is the total number of discrete boreholes in the discrete borehole set corresponding to the fth sequence to be determined, is the first characterization value corresponding to the f-th sequence to be determined, M0 is the total number of borehole connected domains in the borehole connected domain set F0 corresponding to the f-th sequence to be determined, is the total number of boreholes in the ith borehole connected domain in the borehole connected domain set F0, is the area representation value corresponding to the i-th borehole connected domain in the borehole connected domain set F0, is the second characterization value corresponding to the fth sequence to be determined, is the total number of boreholes in the ith borehole connected domain in the borehole connected domain set F0, is the first quantity characterization value of the mth borehole in the i-th borehole connected domain, and is also the number of neighboring boreholes in the i-th borehole connected domain in the neighboring borehole set corresponding to the m-th borehole. is the comprehensive quantitative representation value corresponding to the i-th borehole connected domain, is the third characterization value corresponding to the fth sequence to be determined.
[0102] when The smaller, When the value is larger, it indicates that the possibility of the holes in the lattice presenting a clustered distribution under the parameter dimension corresponding to the fth sequence to be determined is greater. The bigger, The smaller it is, the less likely it is that the holes in the lattice will be clustered under the parameter dimension corresponding to the fth sequence to be determined. When the value is larger, it indicates that the connected domain area in the connected domain set of the boreholes corresponding to the f-th sequence to be determined is larger, and the larger the connected domain area, the more the boreholes in the lattice are clustered under the parameter dimension corresponding to the f-th sequence to be determined. On the contrary, when The smaller it is, the smaller the area of the connected domain in the borehole connected domain set corresponding to the f-th sequence to be determined is, and the smaller the area of the connected domain is, the less the boreholes in the lattice are clustered under the parameter dimension corresponding to the f-th sequence to be determined are. When the value is larger, the distribution of the boreholes in the corresponding borehole connected domain is denser, and the denser the distribution of the boreholes in the borehole connected domain, the more it indicates that the boreholes in the lattice are more clustered under the parameter dimension corresponding to the fth sequence to be determined. On the contrary, when The smaller the value, the less dense the boreholes in the corresponding borehole connected domain are. The less dense the boreholes in the borehole connected domain are, the less dense the boreholes in the lattice are in the parameter dimension corresponding to the fth sequence to be determined. In addition, when , as well as The larger it is, the larger the value of R1 is. The larger the value of R1 is, the more it indicates that the boreholes in the lattice are clustered under the parameter dimension corresponding to the fth sequence to be determined.
[0103] Because if the boreholes in the lattice show more clustered distribution under the parameter dimension corresponding to a certain sequence to be determined, it indicates that the geological engineering parameters in the sequence to be determined are more important for the analysis of subsequent mineral source exploration in the target area or the analysis of geological exploration information in the target area. On the contrary, if the boreholes in the lattice show less clustered distribution under the parameter dimension corresponding to a certain sequence to be determined, the geological engineering parameters in the sequence to be determined are less important for the analysis of subsequent mineral source exploration in the target area. It can be seen that when the characterization value of the clustered distribution of the boreholes corresponding to a certain sequence to be determined is larger, it indicates that the geological engineering parameters in the corresponding sequence to be determined are more important for the analysis of subsequent mineral source exploration in the target area, and also indicates that the possibility that the corresponding sequence to be determined is the target sequence is greater, that is, it indicates that the possibility that the parameters in the corresponding sequence to be determined exist in the clustering parameter vector determined subsequently is greater.
[0104] Therefore, this embodiment can obtain the borehole cluster distribution characterization value corresponding to each to-be-determined sequence in the to-be-determined sequence set through the above process, and the borehole cluster distribution characterization value can reflect the possibility that the corresponding to-be-determined sequence is the target sequence, and the target sequence is the basis for the subsequent determination of the clustering parameter vector; since the clustering cluster of the geological engineering parameter vector is determined by clustering the multi-dimensional geological engineering parameters, that is, it is determined based on the clustering of the clustering parameter vector, in order to further determine that the to-be-determined sequences are the target sequences, it is also necessary to analyze the borehole connected domains corresponding to the to-be-determined sequences after the combination of the to-be-determined sequences, and determine the target sequence based on the analysis results and the borehole cluster distribution characterization values corresponding to each to-be-determined sequence obtained above, and the specific process of determining the target sequence is:
[0105] First, all the sequences to be determined in the set of sequences to be determined are arranged and combined to obtain all the combinations of sequences to be determined, and the set constructed by all the combinations of sequences to be determined is recorded as a feature set. The number of sequences to be determined in the combination of sequences to be determined is a preset number value, and this embodiment is a non-repetitive arrangement and combination of the sequences to be determined; in addition, in specific applications, the implementer can determine the preset number value according to how much the amount of calculation needs to be reduced, that is, if the implementer wants to reduce the amount of calculation more, then the preset number value can be set to a smaller value, and the minimum value of the preset number value is 1, and the maximum value is the number of all sequences to be determined. In this embodiment, the preset number value is set to 3, that is, the number of sequences to be determined in each combination of sequences to be determined is 3.
[0106] Then, according to the borehole connected domain set corresponding to each sequence to be determined in each sequence combination to be determined in the feature set, the connected domain superposition characterization value corresponding to each sequence to be determined in each sequence combination to be determined is obtained; and in this embodiment, the specific process of obtaining the connected domain superposition characterization value corresponding to each sequence to be determined in each sequence combination to be determined is:
[0107] For any undetermined sequence L in any undetermined sequence combination J in the feature set:
[0108] In the sequence combination J to be determined, all the sequences to be determined except the sequence L to be determined are recorded as the first sequences corresponding to the sequence L to be determined, and the set constructed by all the first sequences corresponding to the sequence L to be determined is recorded as the first sequence set corresponding to the sequence L to be determined.
[0109] Then, according to the borehole connected domain set corresponding to each first sequence in the first sequence set, the first eigenvalue set Z0 corresponding to each borehole connected domain in the borehole connected domain set corresponding to the sequence L to be determined is obtained; and the specific acquisition process of the first eigenvalue set Z0 is: for the zth borehole connected domain in the borehole connected domain set corresponding to the sequence L to be determined, the eigenvalue set Z1 to be analyzed corresponding to the zth borehole connected domain is obtained, and the set constructed by all the eigenvalues to be analyzed whose values in the eigenvalue set Z1 to be analyzed are not 0 is recorded as the first eigenvalue set corresponding to the zth borehole connected domain, and the tth eigenvalue to be analyzed in the eigenvalue set Z1 to be analyzed is the number of borehole connected domains in the borehole connected domain set corresponding to the tth first sequence in the first sequence set that have an intersection with the zth borehole connected domain, and the existence of an intersection means that the two borehole connected domains have the same boreholes.
[0110] Then, determine whether the set of characteristic quantity representation values corresponding to each borehole connected domain in the borehole connected domain set corresponding to the sequence L to be determined is an empty set. If not, the corresponding borehole connected domain is recorded as the characteristic connected domain corresponding to the sequence L to be determined; and the set constructed by all characteristic connected domains corresponding to the sequence L to be determined is recorded as the characteristic connected domain set corresponding to the sequence L to be determined.
[0111] Next, the second eigenvalue set corresponding to each feature connected domain in the feature connected domain set corresponding to the sequence L to be determined is obtained, and then the cumulative sum of all second eigenvalues in the second eigenvalue set corresponding to each feature connected domain is obtained, and recorded as the superimposed eigenvalue corresponding to the corresponding feature connected domain. After that, the cumulative sum of the superimposed eigenvalues corresponding to all feature connected domains in the feature connected domain set corresponding to the sequence L to be determined is obtained, and recorded as the connected domain superimposed representation value corresponding to the sequence L to be determined.
[0112] In addition, for any feature connected domain in the feature connected domain set, the rth second eigenvalue in the second eigenvalue set corresponding to the feature connected domain is the reciprocal of the rth first eigenvalue in the first eigenvalue set corresponding to the feature connected domain.
[0113] In this embodiment, the specific expression for obtaining the connected domain superposition characterization value corresponding to the sequence L to be determined is:
[0114]
[0115] Among them, UO is the superposition representation value of the connected domain corresponding to the sequence L to be determined, W0 is the number of feature connected domains in the feature connected domain set corresponding to the sequence L to be determined, is the number of second eigenvalues in the second eigenvalue set corresponding to the zth feature connected domain in the feature connected domain set corresponding to the sequence L to be determined, is the rth second eigenvalue in the second eigenvalue set corresponding to the zth feature connected domain, It is also the reciprocal of the rth first eigenvalue in the first eigenvalue set corresponding to the zth characteristic connected domain.
[0116] And when The larger the value of is, the larger the value of UO is, and the larger the value of UO is, the more it indicates that there are intersections and boreholes between the feature connected domains in the feature connected domain set corresponding to the sequence to be determined L and the borehole connected domains corresponding to the remaining sequences to be determined in the sequence combination J, but the number of intersections and boreholes is small; on the contrary, when The smaller the value of is, the smaller the value of UO is, and the smaller the value of UO is, the more intersections there are between the feature connected domains in the feature connected domain set corresponding to the sequence to be determined L and the borehole connected domains corresponding to the remaining sequences to be determined in the sequence combination J to be determined, and the number of borehole connected domains is large.
[0117] Therefore, this embodiment can obtain the borehole aggregation distribution characterization values corresponding to all the to-be-determined sequences in the to-be-determined sequence combination J in the manner of obtaining the connected domain superposition characterization values corresponding to the to-be-determined sequence L as described above; then, the cumulative sum of the connected domain superposition characterization values corresponding to all the to-be-determined sequences in the to-be-determined sequence combination J is obtained, and recorded as the comprehensive superposition characterization value corresponding to the to-be-determined sequence combination J; when there are intersection borehole connected domains between the feature connected domain set corresponding to the to-be-determined sequences in the to-be-determined sequence combination J and the borehole connected domains corresponding to the remaining to-be-determined sequences in the to-be-determined sequence combination J, but the fewer the number of intersection borehole connected domains, that is, the larger the comprehensive superposition characterization value corresponding to the to-be-determined sequence combination J, it indicates that the geological engineering parameters in the to-be-determined sequences in the to-be-determined sequence combination J are more important for the analysis of the subsequent mineral source exploration in the target area or for the analysis of the subsequent geological survey information in the target area, and vice versa.
[0118] Therefore, this embodiment obtains the comprehensive superposition characterization value corresponding to the sequence combination J to be determined through the above process, and then this embodiment determines the preferred degree characterization value corresponding to the sequence combination J to be determined based on the borehole aggregation distribution characterization value corresponding to each sequence to be determined in the sequence combination J to be determined and the comprehensive superposition characterization value corresponding to the sequence combination J to be determined, which is specifically:
[0119] The normalized value of the cumulative sum of the drilling aggregation distribution characterization values corresponding to all the sequences to be determined in the sequence combination J to be determined is recorded as the comprehensive aggregation distribution characterization value corresponding to the sequence combination J to be determined, and the sum of the normalized value of the comprehensive superposition characterization value corresponding to the sequence combination J to be determined and its corresponding comprehensive aggregation distribution characterization value is recorded as the preference degree characterization value corresponding to the sequence combination J to be determined.
[0120] Therefore, through the above process, the preferred degree characterization value corresponding to each to-be-determined sequence combination in the feature set can be obtained. The larger the preferred degree characterization value, the more important the geological engineering parameters in the to-be-determined sequence in the corresponding to-be-determined sequence combination are for the analysis of subsequent mineral source exploration in the target area or for the analysis of subsequent geological survey information in the target area. Therefore, in this embodiment, among all the to-be-determined sequence combinations, the to-be-determined sequence combination corresponding to the maximum preferred degree characterization value is selected and recorded as the target sequence set, and all the to-be-determined sequences in the target sequence set are recorded as target sequences; then, in each target sequence in the target sequence set, the parameter vector constructed by all the geological engineering parameters belonging to the geological engineering parameter vector corresponding to each borehole is recorded as the clustering parameter vector corresponding to the corresponding borehole.
[0121] Therefore, the clustering parameter vector corresponding to each borehole in the borehole lattice is obtained through the above acquisition.
[0122] Step S004: obtaining a cluster of geological engineering parameter vectors according to the cluster parameter vectors.
[0123] This embodiment then obtains the geological engineering parameter vector cluster cluster by analyzing the cluster parameter vector, specifically:
[0124] A Q-dimensional space is constructed, where Q is the total number of sequences in the target sequence set, and the parameter type represented by the q0th dimension in the Q-dimensional space is consistent with the type of the geological engineering parameter in the q0th target sequence in the target sequence set.
[0125] Then all the geological engineering parameters in the clustering parameter vector corresponding to each borehole in the borehole lattice are mapped into the Q-dimensional space, and the data points obtained after mapping are recorded as the data points to be clustered corresponding to the corresponding boreholes; then all the data points to be clustered are clustered, and each clustering cluster obtained after clustering is recorded as a characteristic clustering cluster, and the clustering cluster formed by the geological engineering parameter vectors corresponding to all the boreholes in each characteristic clustering cluster is recorded as the geological engineering parameter vector clustering cluster corresponding to the corresponding characteristic clustering cluster, that is, the geological engineering parameter vectors in the geological engineering parameter vector clustering cluster corresponding to each characteristic clustering cluster are all geological engineering parameter vectors corresponding to the boreholes in the corresponding characteristic clustering cluster.
[0126] So far, this embodiment has completed the clustering of the geological engineering parameter vector. Since the dimension of the clustering parameter vector is smaller than the dimension of the geological engineering parameter, and the clustering parameter vector is determined based on the geological engineering parameters in the target sequence, this embodiment can ensure that the geological information of the surveyed area can be analyzed more reliably based on the clustering results of the geological engineering parameter vector while also improving the efficiency of obtaining the clustering clusters of the geological engineering parameter vector.
[0127] In summary, the present embodiment first obtains the geological engineering parameter vector corresponding to each borehole in the borehole lattice corresponding to the target area; then, according to the geological engineering parameter vector corresponding to each borehole, obtains a parameter sequence set; then, clusters the geological engineering parameters in each parameter sequence in the parameter sequence set to obtain each cluster cluster corresponding to each parameter sequence, and according to each parameter sequence and the geological engineering parameters in each cluster cluster corresponding to each parameter sequence, obtains the clustering influence degree corresponding to each parameter sequence; then, according to the clustering influence degree, obtains the clustering parameter vector corresponding to each borehole; finally, obtains the clustering cluster of the geological engineering parameter vector according to the clustering parameter vector. The clustering cluster of the geological engineering parameter vector obtained by the present embodiment based on the clustering parameter vector corresponding to each borehole can ensure that the purpose of more reliable analysis of the geological information of the surveyed area is achieved based on the clustering result of the geological engineering parameter vector, while also improving the efficiency or rate of obtaining the clustering cluster of the geological engineering parameter vector.
[0128] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A cluster analysis method based on geological and engineering parameters, characterized in that: The method comprises the following steps: Obtaining a geological engineering parameter vector corresponding to each borehole in a borehole lattice corresponding to a target area, wherein the geological engineering parameter vector is composed of A geological engineering parameters; According to the geological engineering parameter vector corresponding to each borehole, a parameter sequence set is obtained; Clustering the geotechnical parameters in each parameter sequence in the parameter sequence set to obtain each cluster cluster corresponding to each parameter sequence, and obtaining the clustering influence degree corresponding to each parameter sequence according to each parameter sequence and the geotechnical parameters in each cluster cluster corresponding to each parameter sequence; According to the clustering influence degree, obtaining the clustering parameter vector corresponding to each of the boreholes; Obtaining a cluster of geological engineering parameter vectors according to the cluster parameter vectors; The method for obtaining a parameter sequence set includes: Among all the geo-engineering parameter vectors, a sequence constructed by all the geo-engineering parameters at the same position is recorded as a parameter sequence, and the geo-engineering parameters in the parameter sequence are arranged in ascending order; a sequence set constructed by all the obtained parameter sequences is recorded as a parameter sequence set; The method for obtaining the clustering influence degree corresponding to each parameter sequence includes: For any parameter sequence B in the parameter sequence set: Obtaining a first difference sequence corresponding to the parameter sequence B, wherein the bth first difference value in the first difference sequence is the absolute value of the difference between the b+1th geological engineering parameter in the parameter sequence B and the bth geological engineering parameter in the parameter sequence B; recording a normalized value of the variance of the first difference sequence as a characteristic distribution representation value corresponding to the parameter sequence B; The maximum geoengineering parameter and the minimum geoengineering parameter in each cluster corresponding to the parameter sequence B are recorded as the maximum parameter value and the minimum parameter value corresponding to the corresponding cluster respectively; all clusters corresponding to the parameter sequence B are sorted in ascending order of the maximum parameter value to obtain the cluster sequence B0 corresponding to the parameter sequence B; the parameter difference sequence corresponding to the cluster sequence B0 is obtained, and the cth parameter difference value in the parameter difference sequence is the absolute value of the difference between the minimum parameter value corresponding to the c+1th cluster in the cluster sequence B0 and the maximum parameter value corresponding to the cth cluster in the cluster sequence B0; the sum of all parameter differences in the parameter difference sequence is recorded as the comprehensive interval characterization value corresponding to the parameter sequence B; According to the geological engineering parameters in each cluster in the cluster sequence B0, the intra-cluster aggregation characterization value corresponding to the parameter sequence B is obtained; the normalized value of the product of the intra-cluster aggregation characterization value and the comprehensive interval characterization value is recorded as the block aggregation distribution characterization value corresponding to the geological engineering parameter sequence B; the normalized value of the sum of the block aggregation distribution characterization value and the characteristic distribution characterization value is recorded as the clustering influence degree corresponding to the parameter sequence B; The method for obtaining the intra-cluster aggregation characterization value corresponding to the parameter sequence B includes: For any cluster D in the cluster sequence B0: sort all the geotechnical parameters in the cluster D in ascending order and record them as the parameter subsequence corresponding to the cluster D; obtain the second difference sequence corresponding to the parameter subsequence, the d-th second difference in the second difference sequence is the absolute value of the difference between the d+1-th geotechnical parameter in the parameter subsequence and the d-th geotechnical parameter in the parameter subsequence; record the sum of all the second differences in the second difference sequence as the intra-cluster difference characterization value corresponding to the cluster D; The inverse of the mean of the intra-cluster difference characterization values corresponding to all clusters in the cluster sequence B0 is used as the intra-cluster aggregation characterization value corresponding to the parameter sequence B; The method of obtaining the clustering parameter vector corresponding to each borehole according to the clustering influence degree includes: All the sequences to be determined in the parameter sequence set are obtained, and a set constructed by all the sequences to be determined is recorded as a set of sequences to be determined, wherein the sequences to be determined are parameter sequences whose clustering influence is not less than a preset first threshold; according to the cluster cluster sequences corresponding to the sequences to be determined, a set of connected borehole domains and a set of discrete boreholes corresponding to each sequence to be determined are obtained; according to the connected borehole domain sets and the set of discrete boreholes corresponding to each sequence to be determined, a characterization value of borehole aggregation distribution corresponding to each sequence to be determined is obtained; All the sequences to be determined in the set of sequences to be determined are arranged and combined to obtain all the combinations of sequences to be determined, and a set constructed by all the combinations of sequences to be determined is recorded as a feature set, wherein the number of the sequences to be determined in the set of sequences to be determined is a preset number value; According to the borehole connected domain set corresponding to each to-be-determined sequence in each to-be-determined sequence combination in the feature set, a connected domain superposition characterization value corresponding to each to-be-determined sequence in each to-be-determined sequence combination is obtained; and the cumulative sum of the connected domain superposition characterization values corresponding to each to-be-determined sequence in each to-be-determined sequence combination is recorded as the comprehensive superposition characterization value corresponding to the corresponding to-be-determined sequence combination; For any sequence combination J to be determined in the feature set, the normalized value of the cumulative sum of the borehole aggregation distribution characterization values corresponding to all the sequences to be determined in the sequence combination J to be determined is recorded as the comprehensive aggregation distribution characterization value corresponding to the sequence combination J to be determined, and the sum of the normalized value of the comprehensive superposition characterization value corresponding to the sequence combination J to be determined and the comprehensive aggregation distribution characterization value is recorded as the preference degree characterization value corresponding to the sequence combination J to be determined; Among all the sequences to be determined, the sequence combination to be determined corresponding to the maximum preferred degree characterization value is selected and recorded as the target sequence set; in each sequence to be determined in the target sequence set, a parameter vector constructed by all the geotechnical parameters belonging to the geotechnical parameter vector corresponding to each borehole is recorded as the clustering parameter vector corresponding to the corresponding borehole; The method for obtaining the borehole connected domain set and the discrete borehole set corresponding to each sequence to be determined includes: Obtaining a neighboring borehole set corresponding to each borehole; all neighboring boreholes in the neighboring borehole set corresponding to each borehole belong to the eight-neighborhood of the corresponding borehole; For the fth sequence to be determined in the set of sequences to be determined: For any borehole h0 in the borehole lattice, if the hth geotechnical parameter in the gth cluster in the cluster sequence corresponding to the fth sequence to be determined belongs to the geotechnical parameter vector corresponding to the borehole h0, g is used as the fth label value corresponding to the borehole h0; according to the fth label value corresponding to each borehole, all borehole connected domains corresponding to the fth sequence to be determined are obtained; and the set constructed by all borehole connected domains corresponding to the fth sequence to be determined is recorded as the borehole connected domain set corresponding to the fth sequence to be determined; For any borehole h0 in the borehole lattice, if the f-th label value corresponding to all neighboring boreholes in the neighboring borehole set corresponding to the borehole h0 is different from the f-th label value corresponding to the borehole h0, the borehole h0 is recorded as the discrete borehole corresponding to the f-th sequence to be determined; the set constructed by all discrete boreholes corresponding to the f-th sequence to be determined is recorded as the discrete borehole set corresponding to the f-th sequence to be determined; The method for obtaining the drilling hole aggregation distribution characterization value corresponding to each sequence to be determined includes: For the fth sequence to be determined: The sum of the total number of discrete boreholes in the discrete borehole set corresponding to the f-th sequence to be determined and a preset first constant is recorded as the discrete characterization value corresponding to the f-th sequence to be determined; the difference between the total number of boreholes in the borehole lattice and the discrete characterization value is recorded as the characteristic difference value, and the normalized value of the ratio of the characteristic difference value to the discrete characterization value is recorded as the first characterization value corresponding to the f-th sequence to be determined; For any borehole connected domain S1 in the borehole connected domain set F0 corresponding to the f-th sequence to be determined: the square value of the total number of boreholes located in the borehole connected domain S1 is recorded as the area representation value corresponding to the borehole connected domain S1; for the j-th borehole located in the borehole connected domain S1, in the neighboring borehole set corresponding to the j-th borehole, the number of neighboring boreholes located in the borehole connected domain S1 is statistically obtained, and recorded as the first quantity representation value corresponding to the j-th borehole; the cumulative sum of the first quantity representation values corresponding to all boreholes in the borehole connected domain S1 is recorded as the comprehensive quantity representation value corresponding to the borehole connected domain S1; Record the normalized value of the cumulative sum of the area characterization values corresponding to all the borehole connected domains in the borehole connected domain set F0 as the second characterization value corresponding to the f-th sequence to be determined; record the normalized value of the cumulative sum of the comprehensive quantity characterization values corresponding to all the borehole connected domains in the borehole connected domain set F0 as the third characterization value corresponding to the f-th sequence to be determined; record the sum of the first characterization value, the second characterization value and the third characterization value corresponding to the f-th sequence to be determined as the borehole aggregation distribution characterization value corresponding to the f-th sequence to be determined; The method for obtaining the connected domain superposition characterization value corresponding to each to-be-determined sequence in each to-be-determined sequence combination includes: For any sequence L to be determined in the sequence combination J to be determined: In the to-be-determined sequence combination J, all the remaining to-be-determined sequences except the to-be-determined sequence L are recorded as first sequences corresponding to the to-be-determined sequence L, and a set constructed by all first sequences corresponding to the to-be-determined sequence L is recorded as a first sequence set corresponding to the to-be-determined sequence L; according to the borehole connected domain set corresponding to each first sequence in the first sequence set, a first eigenvalue set corresponding to each borehole connected domain in the borehole connected domain set corresponding to the to-be-determined sequence L is obtained; Determine whether the set of characteristic quantity characterization values corresponding to each borehole connected domain in the borehole connected domain set corresponding to the sequence L to be determined is an empty set; if not, record the corresponding borehole connected domain as the characteristic connected domain corresponding to the sequence L to be determined; record the set constructed by all characteristic connected domains corresponding to the sequence L to be determined as the characteristic connected domain set corresponding to the sequence L to be determined; The method comprises: obtaining a second eigenvalue set corresponding to each feature connected domain in the feature connected domain set, and recording the cumulative sum of all second eigenvalues in the second eigenvalue set corresponding to each feature connected domain as the superposition eigenvalue corresponding to the corresponding feature connected domain; recording the cumulative sum of the superposition eigenvalues corresponding to all feature connected domains in the feature connected domain set as the connected domain superposition characterization value corresponding to the sequence L to be determined; the rth second eigenvalue in the second eigenvalue set corresponding to the qth feature connected domain in the feature connected domain set is the reciprocal of the rth first eigenvalue in the first eigenvalue set corresponding to the qth feature connected domain; and obtaining the first eigenvalue set corresponding to each borehole connected domain in the borehole connected domain set corresponding to the sequence L to be determined, comprising: For the zth borehole connected domain in the set of borehole connected domains corresponding to the sequence L to be determined: obtaining a set of eigenvalues to be analyzed corresponding to the zth borehole connected domain, wherein the tth eigenvalue to be analyzed in the set of eigenvalues to be analyzed is the number of borehole connected domains in the set of borehole connected domains corresponding to the tth first sequence in the first sequence set that have an intersection with the zth borehole connected domain; recording a set constructed from all eigenvalues to be analyzed whose values are not 0 in the set of eigenvalues to be analyzed as the first eigenvalue set corresponding to the zth borehole connected domain; The method for obtaining a cluster of geological engineering parameter vectors according to the cluster parameter vectors includes: A Q-dimensional space is constructed, where Q is the total number of sequences in the target sequence set; all the geological engineering parameters in the clustering parameter vector corresponding to each borehole in the borehole lattice are mapped to the Q-dimensional space, and the data points obtained after mapping are recorded as the data points to be clustered corresponding to the corresponding boreholes; all the data points to be clustered are clustered, and each cluster cluster obtained after clustering is recorded as a feature cluster cluster, and the cluster cluster formed by the geological engineering parameter vectors corresponding to all the boreholes in each feature cluster cluster is recorded as the geological engineering parameter vector cluster cluster corresponding to the corresponding feature cluster cluster.
