CN116415161B - Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave - Google Patents
Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave Download PDFInfo
- Publication number
- CN116415161B CN116415161B CN202310407991.4A CN202310407991A CN116415161B CN 116415161 B CN116415161 B CN 116415161B CN 202310407991 A CN202310407991 A CN 202310407991A CN 116415161 B CN116415161 B CN 116415161B
- Authority
- CN
- China
- Prior art keywords
- soil
- gaussian mixture
- physical wave
- drilling
- cluster
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 133
- 238000005553 drilling Methods 0.000 title claims abstract description 85
- 238000003169 complementation method Methods 0.000 title abstract description 7
- 239000002689 soil Substances 0.000 claims abstract description 127
- 239000000203 mixture Substances 0.000 claims abstract description 90
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000005192 partition Methods 0.000 claims description 2
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000003384 imaging method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011202 physical detection method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V9/00—Prospecting or detecting by methods not provided for in groups G01V1/00 - G01V8/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Biology (AREA)
- Algebra (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Probability & Statistics with Applications (AREA)
- Operations Research (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Geophysics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a fitting complementation method of geological drilling detection and different physical wave detection data of a string-shaped karst cave, which can deeply analyze stratum distribution trend and provide data for the design of foundation pit support and foundation of the string-shaped karst cave; the method comprises the following steps: firstly classifying soil according to the result of geological drilling detection of the string-shaped karst cave, carrying out hierarchical division on the soil classified in the same way, and carrying out physical wave characteristic assignment on the soil of each hierarchy; adopting a physical wave detection instrument to detect physical waves in the area where the drilling hole is positioned; carrying out Gaussian mixture clustering on the detected data samples, and matching the samples to different clusters; bringing the data of the borehole detection assignment into a final model of Gaussian mixture clustering; finally, determining whether the drilling detection result is consistent with the physical wave detection result; the method can embody the trend of the change of the soil quality of the same cluster of the string-shaped karst cave along with the density; the soil layer distribution is well known, and the mechanical analysis is fully performed on single soil.
Description
Technical Field
The invention belongs to the technical field of engineering exploration, and particularly relates to a fitting complementation method of geological drilling detection and different physical wave detection data of a string-shaped karst cave.
Background
String karst cave geology is a relatively special geological morphology that is difficult to effectively detect and understand in geological exploration.
The current common physical detection methods are mainly divided into drilling detection and physical wave detection; wherein, physical wave detection mainly includes: seismic exploration, electromagnetic exploration, and the like.
When the geology of the string-shaped karst cave is detected, a mode of combining drilling detection and physical wave detection is often adopted for detection, firstly, soil is adopted through drilling detection, and then the accuracy of drilling detection is checked through physical wave detection; the accuracy of borehole detection is often related to fitting of data through physical wave detection, so that three-dimensional imaging of underground cavities is realized.
The existing three-dimensional imaging generated by the physical wave detection data is often simply divided into types of soil, and cannot reflect the gradual change trend of the soil in each type or describe the gradual change trend of the soil in each type inaccurately, so that the detailed understanding of soil layer distribution and the full mechanical analysis of single soil are not facilitated.
Disclosure of Invention
The invention aims to overcome the problems existing in the prior art and greatly improve the technical effect on the basis of the prior art; for this reason, the invention provides a fitting complementation method of geological drilling detection of a string-shaped karst cave and detection data of different physical waves, which comprises the following steps:
1. analyzing the drilling detection result according to the string-shaped karst cave geological drilling detection result, and classifying the soil detected by the drilling;
2. carrying out hierarchical division on the soil with the same classification, and carrying out physical wave characteristic assignment on the soil of each hierarchy according to the propagation characteristics of different physical waves in the soil; the physical wave characteristic assignment refers to the propagation characteristic value of the physical wave in the corresponding soil hierarchy after the soil hierarchy is divided;
3. according to the region where the drilling is detected, adopting a physical wave detection instrument to detect physical waves of the region where the drilling is detected; and extracting detection data sample { x }, based on the result of physical wave detection 1 ，x 2 ，…x n }，x j Representing physical wave characteristic values corresponding to various places of the string-shaped karst cave, wherein j=1, 2 and … n;
4. carrying out Gaussian mixture clustering on the detected data samples, matching the samples to different clusters, and recording posterior probability of sample elements in each cluster and a final model of the Gaussian mixture clustering;
the posterior probability determines the final cluster division of the sample, and the posterior probability formula of the Gaussian mixture cluster is as follows:
wherein alpha is i 、u i Sum sigma i Representing sample x j The parameters of the corresponding ith Gaussian mixture component, k represents the total of k Gaussian mixture component compositions, alpha h 、u h Sum sigma h Parameters representing any one of the gaussian mixture components in the gaussian mixture distribution; p (x) j |u i ，∑ i ) Representation about x j 、u i Sum sigma i Probability density function, p (x) j |u h ，∑ h ) Representation about x j 、u h Sum sigma h Probability density functions of (2); the Gaussian mixture distribution is composed of k Gaussian mixture components, each of the mixture components corresponds to one Gaussian mixture distribution, and the decision formula of the Gaussian mixture distribution is as follows:
wherein x is an element in the sample, alpha h 、u h Sum sigma h For each parameter of the gaussian mixture component in the gaussian mixture distribution, p (x|u h ，∑ h ) Representation about x, u h Sum sigma h Probability density functions of (2);
the final model of the Gaussian mixture cluster is obtained by a parameter alpha of posterior probability h 、u h Sum sigma h The final set of parameters alpha is determined by continuous iterative updating of the parameters until the stopping condition is met h 、u h Sum sigma h The determined posterior probability formula is the final model of Gaussian mixture clustering;
5. carrying out physical wave characteristic assignment on data of different layers in the same soil property classification of drilling detection into a final model of Gaussian mixture clustering to obtain posterior probability of each layer in all soil property classifications of drilling detection, and solving cluster marks of each layer in all soil property classifications by the posterior probability, wherein a calculation formula of the cluster marks is as follows:
wherein, gamma ji Representing a sample determined by the ith Gaussian mixture component as x j Posterior probability, eta j Representing sample x j A final cluster partition label of (a);
6. the final cluster division marks of all layers in all soil classifications are detected through the obtained drill holes, the cluster division marks of all layers in the same type of soil are respectively matched into the same cluster of Gaussian mixture clusters, and whether the drill hole detection result is consistent with the physical wave detection result is determined; if the probabilities corresponding to the final cluster division marks are consistent, combining the probabilities corresponding to the final cluster division marks obtained by the drilling detection data and the physical wave detection data, and carrying out fitting analysis on the probabilities corresponding to the final cluster division marks of the same cluster of soil to obtain a probability layering curve of the same cluster of soil; and determining the density distribution curve of the geological soil layer of the string-shaped karst cave according to the probability layering curve.
Further, classifying the earth detected by the borehole includes: and classifying the soil detected by the drilling according to the soil condition corresponding to the drilling detection result.
Further, the grading of the same classified soil property includes: hierarchical division is carried out according to the density of the soil of the same class, and the soil of the same class is divided into different layers: (A) 1 ，A 2 ，…A n ) Wherein A is 1 To A n The density of (2) gradually increases.
Further, the assigning physical wave characteristics to the soil of each level includes: according to the intensity change characteristics of the physical wave at the propagation speeds of the soil with different densities, the propagation speed of the physical wave at different soil layers is determined, and the determined propagation speed is taken as a propagation characteristic value of the corresponding soil layer.
Further, the method is characterized in that the result according to the physical wave detection comprises: extracting detection data sample { x }, based on the result of physical wave detection 1 ，x 2 ，…x n -wherein the sample element x 1 To x n And representing the propagation speed corresponding to each point of the string-shaped karst cave extracted from the physical wave detection result, namely the physical wave characteristic value corresponding to each point of the string-shaped karst cave.
Further, the flow of Gaussian mixture clustering includes: a) Input sample set d= { x 1 ，x 2 ，…x n Setting the number k of Gaussian mixture components, and initializing model parameters { (alpha) of Gaussian mixture distribution i ，u i ，∑ i ) I 1 is less than or equal to i is less than or equal to k; b) Calculation of x by posterior probability formula j Posterior probability in each mixed component; c) Through posterior testProbability will sample x j Model parameters { (α) i ，u i ，∑ i ) I1.ltoreq.i.ltoreq.k is updated to { (α) i ＇，u i ＇，∑ i ' i 1 is less than or equal to i is less than or equal to k }; d) Bringing the newly updated model parameters into step b) and generating new model parameters, and iterating the process until a stop condition is met; e) Determining x from cluster marking formula j Cluster marking η j And x is taken as j Grouping into corresponding clusters C ηj The method comprises the steps of carrying out a first treatment on the surface of the f) Output cluster division c= { C 1 ，C 2 ，…C k }；
Further, the matching the cluster division marks of each level in the same soil texture into the same cluster of the Gaussian mixture cluster comprises the following steps: if the cluster division marks of all layers in the same soil property detected by drilling are completely consistent, verifying that the result of Gaussian mixture clustering is matched with drilling data, and proving that the drilling detection result is consistent with the physical wave detection result; if the cluster division marks of all layers in the same soil texture detected by drilling are not completely consistent, the result of Gaussian mixture clustering is verified to be not matched with drilling data, and the fact that the drilling detection result is inconsistent with the physical wave detection result is proved.
The beneficial effects of the invention are as follows:
the invention provides a fitting complementation method of geological drilling detection and different physical wave detection data of a string-shaped karst cave; the invention divides the same soil sample of drilling data into different layers, and assigns values to the soil samples of different layers; meanwhile, probability distribution analysis is carried out on the detection samples of the physical waves through a Gaussian mixture model, and the detection samples of the physical waves are divided into different clusters; bringing assignment values of different layers of the same soil sample of drilling data into a final Gaussian mixture model posterior probability function, so as to judge whether the drilling data and the physical wave detection data are consistent; finally, combining the probability corresponding to the final cluster division mark obtained by the drilling detection data and the physical wave detection data, performing data fitting to obtain a probability layering curve of the soil quality of the same cluster, and determining a density distribution curve of a geological soil layer of the string-shaped karst cave according to the probability layering curve, wherein the curve can reflect the trend of the soil quality of the same cluster of the string-shaped karst cave along with the change of the density; on the basis of the density distribution curve, the three-dimensional model showing the gradient of the same soil sample can be established, so that the detailed understanding of soil layer distribution and the full mechanical analysis of single soil property are facilitated.
Drawings
Fig. 1: the invention discloses a fitting complementation method flow chart of geological drilling detection and different physical wave detection data of a string-shaped karst cave.
Detailed Description
Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings; it should be understood that the particular embodiments presented herein are illustrative and explanatory only and are not restrictive of the invention.
Referring to fig. 1, a flow chart of a complementary fitting method of geological borehole detection and different physical wave detection data of a string-shaped karst cave according to the present invention is shown, the flow chart comprises: step S100, analyzing the drilling detection result according to the string-shaped karst cave geological drilling detection result, and classifying the drilling detection soil; step S101, carrying out hierarchical division on the same classified soil, and carrying out physical wave characteristic assignment on the soil of each hierarchy according to the propagation characteristics of different physical waves in the soil; the physical wave characteristic assignment refers to the propagation characteristic value of the physical wave in the corresponding soil hierarchy after the soil hierarchy is divided; step S102, according to the region where the drilling detection is located, adopting a physical wave detection instrument to detect the physical wave of the region where the drilling is located; and extracting detection data sample { x }, based on the result of physical wave detection 1 ，x 2 ，…x n }，x j Representing physical wave characteristic values corresponding to various places of the string-shaped karst cave, wherein j=1, 2 and … n; step S103, carrying out Gaussian mixture clustering on the detected data samples, matching the samples to different clusters, and recording posterior probability of sample elements in each cluster and a final model of the Gaussian mixture clustering; step S104, bringing the data of physical wave characteristic assignment for different layers in the same soil property classification of the drilling detection into a final model of Gaussian mixture clustering to obtain posterior probabilities of all layers in all the soil property classification of the drilling detection, and obtaining all the soil property classification by the posterior probabilitiesCluster marks of each level in the soil classification; step S105, detecting final cluster division marks of all layers in all soil classifications through the obtained drill holes, respectively matching the cluster division marks of all layers in the same soil classification into the same cluster of Gaussian mixture clusters, and determining whether the drill hole detection result is consistent with the physical wave detection result; if the probabilities corresponding to the final cluster division marks are consistent, combining the probabilities corresponding to the final cluster division marks obtained by the drilling detection data and the physical wave detection data, and carrying out fitting analysis on the probabilities corresponding to the final cluster division marks of the same cluster of soil to obtain a probability layering curve of the same cluster of soil; and determining the density distribution curve of the geological soil layer of the string-shaped karst cave according to the probability layering curve.
Specifically, firstly classifying the soil according to the result of geological drilling detection of the string-shaped karst cave, carrying out hierarchical division on the soil classified in the same way, and carrying out physical wave characteristic assignment on the soil of each hierarchy; according to the region where the drilling is detected, adopting a physical wave detection instrument to detect physical waves of the region where the drilling is detected; extracting detection data sample { x } from the result of physical wave detection 1 ，x 2 ，…x n -a }; then, carrying out Gaussian mixture clustering on the detected data samples, matching the samples to different clusters, and recording posterior probability of sample elements in each cluster and a final model of the Gaussian mixture clustering; carrying out physical wave characteristic assignment on the drilling detection data into a final model of Gaussian mixture clustering to obtain posterior probability of each level in all soil classifications of the drilling detection, and solving cluster marks of each level in all soil classifications by the posterior probability; the method comprises the steps of determining whether a drilling detection result is consistent with a physical wave detection result or not by matching cluster division marks of all layers in the same soil in drilling detection into the same cluster of Gaussian mixture clusters; if the probabilities corresponding to the final cluster division marks are consistent, combining the probabilities corresponding to the final cluster division marks obtained by the drilling detection data and the physical wave detection data, and carrying out fitting analysis on the probabilities corresponding to the final cluster division marks of the same cluster of soil to obtain a probability layering curve of the same cluster of soil; and determining the density distribution curve of the geological soil layer of the string-shaped karst cave according to the probability layering curve.
Step S100, analyzing the drilling detection result according to the string-shaped karst cave geological drilling detection result, and classifying the drilling detection soil; specifically, the method for classifying the soil properties detected by drilling holes comprises the following steps: and classifying the soil detected by the drilling according to the soil condition corresponding to the drilling detection result.
Step S101, carrying out hierarchical division on the same classified soil, and carrying out physical wave characteristic assignment on the soil of each hierarchy according to the propagation characteristics of different physical waves in the soil; the physical wave characteristic assignment refers to the propagation characteristic value of the physical wave in the corresponding soil hierarchy after the soil hierarchy is divided; specifically, the method for hierarchical division of the same classified soil texture comprises the following steps: hierarchical division is carried out according to the density of the soil of the same class, and the soil of the same class is divided into different layers: (A) 1 ，A 2 ，…A n ) Wherein A is 1 To A n The density of (2) gradually increases.
In the foregoing embodiment, specifically, the assigning the physical wave characteristics to the soil of each level includes: according to the intensity change characteristics of the physical wave at the propagation speeds of the soil with different densities, the propagation speed of the physical wave at different soil layers is determined, and the determined propagation speed is taken as a propagation characteristic value of the corresponding soil layer.
Step S102, according to the region where the drilling detection is located, adopting a physical wave detection instrument to detect the physical wave of the region where the drilling is located; and extracting detection data sample { x }, based on the result of physical wave detection 1 ，x 2 ，…x n }，x j Representing physical wave characteristic values corresponding to various places of the string-shaped karst cave, wherein j=1, 2 and … n; specifically, from the result of physical wave detection, a detection data sample { x } is extracted 1 ，x 2 ，…x n -wherein the sample element x 1 To x n And representing the propagation speed corresponding to each point of the string-shaped karst cave extracted from the physical wave detection result, namely the physical wave characteristic value corresponding to each point of the string-shaped karst cave.
Step S103, carrying out Gaussian mixture clustering on the detected data samples, matching the samples to different clusters, and recording the posterior probability and Gaussian mixture of the sample elements in each clusterCombining the final models of the clusters; in particular, the different clusters refer to the classification of samples into different classes; the Gaussian mixture clustering process comprises the following steps: a) Input sample set d= { x 1 ，x 2 ，…x n Setting the number k of Gaussian mixture components, and initializing model parameters { (alpha) of Gaussian mixture distribution i ，u i ，∑ i ) I 1 is less than or equal to i is less than or equal to k; b) Calculation of x by posterior probability formula j Posterior probability in each mixed component; c) Sample x by posterior probability j Model parameters { (α) i ，u i ，∑ i ) I1.ltoreq.i.ltoreq.k is updated to { (α) i ＇，u i ＇，∑ i ' i 1 is less than or equal to i is less than or equal to k }; d) Bringing the newly updated model parameters into step b) and generating new model parameters, and iterating the process until a stop condition is met; e) Determining x from cluster marking formula j Cluster marking η j And x is taken as j Grouping into corresponding clusters C ηj The method comprises the steps of carrying out a first treatment on the surface of the f) Output cluster division c= { C 1 ，C 2 ，…C k }。
In the above embodiment, specifically, the posterior probability determines the final cluster division of the sample, and the posterior probability formula of the gaussian mixture cluster is:
wherein alpha is i 、u i Sum sigma i Representing sample x j The parameters of the corresponding ith Gaussian mixture component, k represents the total of k Gaussian mixture component compositions of the Gaussian mixture distribution, alpha h 、u h Sum sigma h Parameters representing any one of the gaussian mixture components in the gaussian mixture distribution; p (x) j |u i ，∑ i ) Representation about x j 、u i Sum sigma i Probability density function, p (x) j |u h ，∑ h ) Representation about x j 、u h Sum sigma h Probability density functions of (2); the Gaussian mixture distribution is composed of k Gaussian mixture components, and each mixture component corresponds to one Gaussian mixtureThe decision formula of the resultant distribution is:
wherein x is an element in the sample, alpha h 、u h Sum sigma h For each parameter of the gaussian mixture component in the gaussian mixture distribution, p (x|u h ，∑ h ) Representation about x, u h Sum sigma h Probability density function of (a).
In the above embodiment, specifically, the final model of the gaussian mixture cluster is a parameter α of a posterior probability h 、u h Sum sigma h The final set of parameters alpha is determined by continuous iterative updating of the parameters until the stopping condition is met h 、u h Sum sigma h The determined posterior probability formula is the final model of Gaussian mixture clustering;
step S104, carrying data of physical wave characteristic assignment on different layers in the same soil property classification of drilling detection into a final model of Gaussian mixture clustering to obtain posterior probability of each layer in all soil property classifications of drilling detection, and solving cluster marks of each layer in all soil property classifications according to the posterior probability, wherein a calculation formula of the cluster marks is as follows:
wherein, gamma ji Representing a sample determined by the ith Gaussian mixture component as x j Posterior probability, eta j Representing sample x j Is marked by the final cluster division.
Step S105, detecting final cluster division marks of all layers in all soil classifications through the obtained drill holes, respectively matching the cluster division marks of all layers in the same soil classification into the same cluster of Gaussian mixture clusters, and determining whether the drill hole detection result is consistent with the physical wave detection result; if the probabilities corresponding to the final cluster division marks are consistent, combining the probabilities corresponding to the final cluster division marks obtained by the drilling detection data and the physical wave detection data, and carrying out fitting analysis on the probabilities corresponding to the final cluster division marks of the same cluster of soil to obtain a probability layering curve of the same cluster of soil; and determining the density distribution curve of the geological soil layer of the string-shaped karst cave according to the probability layering curve.
In the above embodiment, specifically, the matching the cluster division marks of each level in the same soil property into the same cluster of the gaussian mixture cluster includes: cluster division of Gaussian mixture model is based on physical wave in sample x j Is divided by the propagation speed; the physical wave characteristics of each layer of the same earth quality of the drilling data are assigned, and the physical waves of the same kind are detected according to the physical waves to be converted; therefore, if the cluster division marks of all layers in the same type of soil detected by drilling are completely consistent, the propagation speed of the physical wave in all layers of soil samples in the same type of soil detected by drilling is proved to be completely matched with the speed of data in a certain cluster analyzed by physical wave detection, the result of Gaussian mixture clustering is verified to be matched with drilling data, and the drilling detection result and the physical wave detection result are proved to be consistent; if the cluster division marks of all layers in the same soil texture detected by drilling are not completely consistent, the result of Gaussian mixture clustering is verified to be not matched with drilling data, and the fact that the drilling detection result is inconsistent with the physical wave detection result is proved.
It is to be understood that the above-described embodiments are one or more embodiments of the invention, and that many other embodiments and variations thereof are possible in accordance with the invention; variations and modifications of the invention, which are intended to be within the scope of the invention, will occur to those skilled in the art without any development of the invention.
Claims (7)
1. A method for fitting and complementing geological borehole detection and different physical wave detection data of a string-shaped karst cave, which is characterized by comprising the following steps:
1) Analyzing the drilling detection result according to the string-shaped karst cave geological drilling detection result, and classifying the soil detected by the drilling;
2) Carrying out hierarchical division on the soil with the same classification, and carrying out physical wave characteristic assignment on the soil of each hierarchy according to the propagation characteristics of different physical waves in the soil; the physical wave characteristic assignment refers to the propagation characteristic value of the physical wave in the corresponding soil hierarchy after the soil hierarchy is divided;
3) According to the region where the drilling is detected, adopting a physical wave detection instrument to detect physical waves of the region where the drilling is detected; and extracting detection data sample { x }, based on the result of physical wave detection 1 ，x 2 ，…x n }，x j Representing physical wave characteristic values corresponding to various places of the string-shaped karst cave, wherein j=1, 2 and … n;
4) Carrying out Gaussian mixture clustering on the detected data samples, matching the samples to different clusters, and recording posterior probability of sample elements in each cluster and a final model of the Gaussian mixture clustering;
the posterior probability determines the final cluster division of the sample, and the posterior probability formula of the Gaussian mixture cluster is as follows:
wherein alpha is i 、u i Sum sigma i Representing sample x j The parameters of the corresponding ith Gaussian mixture component, k represents the total of k Gaussian mixture component compositions, alpha h 、u h Sum sigma h Parameters representing any one of the gaussian mixture components in the gaussian mixture distribution; p (x) j |u i ，∑ i ) Representation about x j 、u i Sum sigma i Probability density function, p (x) j |u h ，∑ h ) Representation about x j 、u h Sum sigma h Probability density functions of (2); the Gaussian mixture distribution is composed of k Gaussian mixture components, each of the mixture components corresponds to one Gaussian mixture distribution, and the decision formula of the Gaussian mixture distribution is as follows:
wherein x is an element in the sample, alpha h 、u h Sum sigma h For each parameter of the gaussian mixture component in the gaussian mixture distribution, p (x|u h ，∑ h ) Representation about x, u h Sum sigma h Probability density functions of (2);
the final model of the Gaussian mixture cluster is obtained by a parameter alpha of posterior probability h 、u h Sum sigma h The final set of parameters alpha is determined by continuous iterative updating of the parameters until the stopping condition is met h 、u h Sum sigma h The determined posterior probability formula is the final model of Gaussian mixture clustering;
5) Carrying out physical wave characteristic assignment on data of different layers in the same soil property classification of drilling detection into a final model of Gaussian mixture clustering to obtain posterior probability of each layer in all soil property classifications of drilling detection, and solving cluster marks of each layer in all soil property classifications by the posterior probability, wherein a calculation formula of the cluster marks is as follows:
wherein, gamma ji Representing a sample determined by the ith Gaussian mixture component as x j Posterior probability, eta j Representing sample x j A final cluster partition label of (a);
6) The final cluster division marks of all layers in all soil classifications are detected through the obtained drill holes, the cluster division marks of all layers in the same type of soil are respectively matched into the same cluster of Gaussian mixture clusters, and whether the drill hole detection result is consistent with the physical wave detection result is determined; if the probabilities corresponding to the final cluster division marks are consistent, combining the probabilities corresponding to the final cluster division marks obtained by the drilling detection data and the physical wave detection data, and carrying out fitting analysis on the probabilities corresponding to the final cluster division marks of the same cluster of soil to obtain a probability layering curve of the same cluster of soil; and determining the density distribution curve of the geological soil layer of the string-shaped karst cave according to the probability layering curve.
2. The method of fitting and complementing a string-like karst cave geological borehole survey with different physical wave detection data according to claim 1, wherein classifying the borehole detected soil comprises: and classifying the soil detected by the drilling according to the soil condition corresponding to the drilling detection result.
3. The method of fitting and complementing a string-like karst cave geological borehole survey with different physical wave detection data according to claim 1, wherein said hierarchically partitioning the same classified soil comprises: hierarchical division is carried out according to the density of the soil of the same class, and the soil of the same class is divided into different layers: (A) 1 ，A 2 ，…A n ) Wherein A is 1 To A n The density of (2) gradually increases.
4. The method of fitting and complementing a string-like karst cave geological borehole survey with different physical wave detection data according to claim 1, wherein said assigning physical wave characteristics to each level of soil comprises: according to the intensity change characteristics of the physical wave at the propagation speeds of the soil with different densities, the propagation speed of the physical wave at different soil layers is determined, and the determined propagation speed is taken as a propagation characteristic value of the corresponding soil layer.
5. The method of fitting and complementing a string-like karst cave geological borehole survey with different physical wave detection data according to claim 1, wherein the results from the physical wave detection comprise: extracting detection data sample { x }, based on the result of physical wave detection 1 ，x 2 ，…x n -wherein the sample element x 1 To x n And representing the propagation speed corresponding to each point of the string-shaped karst cave extracted from the physical wave detection result, namely the physical wave characteristic value corresponding to each point of the string-shaped karst cave.
6. The method for fitting and complementing a string-shaped karst cave geological borehole detection and different physical wave detection data according to claim 1, wherein the gaussian mixture clustering process comprises the following steps: a) Input sample set d= { x 1 ，x 2 ，…x n Setting the number k of Gaussian mixture components, and initializing model parameters { (alpha) of Gaussian mixture distribution i ，u i ，∑ i ) I 1 is less than or equal to i is less than or equal to k; b) Calculation of x by posterior probability formula j Posterior probability in each mixed component; c) Sample x by posterior probability j Model parameters { (α) i ，u i ，∑ i ) I1.ltoreq.i.ltoreq.k is updated to { (α) i ＇，u i ＇，∑ i ' i 1 is less than or equal to i is less than or equal to k }; d) Bringing the newly updated model parameters into step b) and generating new model parameters, and iterating the process until a stop condition is met; e) Determining x from cluster marking formula j Cluster marking η j And x is taken as j Grouping into corresponding clusters C ηj The method comprises the steps of carrying out a first treatment on the surface of the f) Output cluster division c= { C 1 ，C 2 ，…C k }。
7. The method of fitting and complementing a string-like karst cave geological borehole survey with different physical wave detection data according to claim 1, wherein matching cluster division markers of each level in the same class of soil into the same cluster of gaussian mixture clusters comprises: if the cluster division marks of all layers in the same soil property detected by drilling are completely consistent, verifying that the result of Gaussian mixture clustering is matched with drilling data, and proving that the drilling detection result is consistent with the physical wave detection result; if the cluster division marks of all layers in the same soil texture detected by drilling are not completely consistent, the result of Gaussian mixture clustering is verified to be not matched with drilling data, and the fact that the drilling detection result is inconsistent with the physical wave detection result is proved.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310407991.4A CN116415161B (en) | 2023-04-17 | 2023-04-17 | Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310407991.4A CN116415161B (en) | 2023-04-17 | 2023-04-17 | Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116415161A CN116415161A (en) | 2023-07-11 |
| CN116415161B true CN116415161B (en) | 2023-09-08 |
Family
ID=87052773
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310407991.4A Active CN116415161B (en) | 2023-04-17 | 2023-04-17 | Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116415161B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117077535A (en) * | 2023-08-31 | 2023-11-17 | 广东电白建设集团有限公司 | High formwork construction monitoring method based on Gaussian mixture clustering algorithm |
| CN117077066B (en) * | 2023-10-17 | 2024-03-12 | 合肥联宝信息技术有限公司 | Waveform abnormality detection method, waveform abnormality detection device, electronic device and storage medium |
| CN117421615B (en) * | 2023-12-13 | 2024-03-08 | 中交路桥建设有限公司 | Geological drilling detection and physical wave detection data combination analysis method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104280771A (en) * | 2014-10-27 | 2015-01-14 | 中国石油集团川庆钻探工程有限公司地球物理勘探公司 | Three-dimensional seismic data waveform semi-supervised clustering method based on EM algorithm |
| CN104700617A (en) * | 2015-04-02 | 2015-06-10 | 武汉大学 | High-precision lane information extracting method based on low-precision GPS track data |
| CN105353644A (en) * | 2015-09-29 | 2016-02-24 | 中国人民解放军63892部队 | Radar target track derivative system on the basis of information mining of real-equipment data and method thereof |
| CN107121677A (en) * | 2017-06-02 | 2017-09-01 | 太原理工大学 | Avoidance radar method and device based on ultra wide band cognition CPPM signals |
| CN115243119A (en) * | 2022-07-18 | 2022-10-25 | 中交四航局第一工程有限公司 | Intelligent monitoring device, system and method for bridge construction |
| CN115407407A (en) * | 2021-05-26 | 2022-11-29 | 中国石油化工股份有限公司 | Three-dimensional geological model construction method for carbonate rock ancient karst cave and filling thereof |
| CN115494560A (en) * | 2022-10-10 | 2022-12-20 | 广东省工程勘察院 | Novel modeling method based on combination of CT (computed tomography) and drilling data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220049606A1 (en) * | 2018-12-18 | 2022-02-17 | Technological Resources Pty. Limited | Automated Updating of Geological Model Boundaries for Improved Ore Extraction |
-
2023
- 2023-04-17 CN CN202310407991.4A patent/CN116415161B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104280771A (en) * | 2014-10-27 | 2015-01-14 | 中国石油集团川庆钻探工程有限公司地球物理勘探公司 | Three-dimensional seismic data waveform semi-supervised clustering method based on EM algorithm |
| CN104700617A (en) * | 2015-04-02 | 2015-06-10 | 武汉大学 | High-precision lane information extracting method based on low-precision GPS track data |
| CN105353644A (en) * | 2015-09-29 | 2016-02-24 | 中国人民解放军63892部队 | Radar target track derivative system on the basis of information mining of real-equipment data and method thereof |
| CN107121677A (en) * | 2017-06-02 | 2017-09-01 | 太原理工大学 | Avoidance radar method and device based on ultra wide band cognition CPPM signals |
| CN115407407A (en) * | 2021-05-26 | 2022-11-29 | 中国石油化工股份有限公司 | Three-dimensional geological model construction method for carbonate rock ancient karst cave and filling thereof |
| CN115243119A (en) * | 2022-07-18 | 2022-10-25 | 中交四航局第一工程有限公司 | Intelligent monitoring device, system and method for bridge construction |
| CN115494560A (en) * | 2022-10-10 | 2022-12-20 | 广东省工程勘察院 | Novel modeling method based on combination of CT (computed tomography) and drilling data |
Non-Patent Citations (1)
| Title |
|---|
| 基于贝叶斯ICA的多工况非高斯过程故障检测;邓晓刚等;《控制工程》;第25卷(第3期);402-407 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116415161A (en) | 2023-07-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116415161B (en) | Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave | |
| US8374974B2 (en) | Neural network training data selection using memory reduced cluster analysis for field model development | |
| US10656068B2 (en) | Method for providing a numerical model of a sample of rock | |
| CN108952699B (en) | Intelligent stratum lithology identification method in complex geological drilling process | |
| WO2009134685A2 (en) | System and method for interpretation of well data | |
| CN106062310A (en) | Facies definition using unsupervised classification procedures | |
| Eskandari et al. | Reservoir Modelling of Complex Geological Systems--A Multiple-Point Perspective | |
| CN104570109B (en) | A kind of method of reservoir hydrocarbons prediction | |
| CN116168224A (en) | Machine learning lithology automatic identification method based on imaging gravel content | |
| CN114638255A (en) | Tight sandstone fracture logging comprehensive identification method combined with deep learning | |
| CN111626377A (en) | Lithofacies identification method, device, equipment and storage medium | |
| CN117421615B (en) | Geological drilling detection and physical wave detection data combination analysis method | |
| CN110414085A (en) | It has developed fault block oil reservoir original oil-water level and has determined method | |
| CN115809411A (en) | Improved decision tree lithology identification method based on logging data | |
| CN106405638B (en) | A kind of gas saturation determines method and apparatus | |
| CN114202036A (en) | Static sounding automatic soil layer dividing technology based on pattern recognition technology | |
| CN115291277B (en) | Reservoir configuration interpretation method based on multi-attribute intelligent fusion under less well condition | |
| CN111852460B (en) | Logging curve normalization method based on empirical mode decomposition | |
| CN117235628B (en) | Well logging curve prediction method and system based on hybrid Bayesian deep network | |
| Gholanlo | Analysis of permeability based on petrophysical logs: comparison between heuristic numerical and analytical methods | |
| CN117407841B (en) | Shale layer seam prediction method based on optimization integration algorithm | |
| CN111783825B (en) | Logging lithology recognition method based on convolutional neural network learning | |
| Aponte et al. | Enhanced history matching process by incorporation of saturation logs as model selection criteria | |
| US20230333277A1 (en) | Lithofacies Guided Core Description Using Unsupervised Machine Learning | |
| Madani et al. | Improvement of fracture network modeling in fractured reservoirs using conditioning and geostatistical method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |