CN107479093A - A kind of micro-seismic event denoising and clustering method based on potential function - Google Patents

A kind of micro-seismic event denoising and clustering method based on potential function Download PDF

Info

Publication number
CN107479093A
CN107479093A CN201710838746.3A CN201710838746A CN107479093A CN 107479093 A CN107479093 A CN 107479093A CN 201710838746 A CN201710838746 A CN 201710838746A CN 107479093 A CN107479093 A CN 107479093A
Authority
CN
China
Prior art keywords
micro
seismic event
value
cluster
mrow
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.)
Granted
Application number
CN201710838746.3A
Other languages
Chinese (zh)
Other versions
CN107479093B (en
Inventor
尚雪义
李夕兵
董陇军
王泽伟
刘栋
周勇勇
刘德彪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Central South University
Original Assignee
Central South University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Central South University filed Critical Central South University
Priority to CN201710838746.3A priority Critical patent/CN107479093B/en
Publication of CN107479093A publication Critical patent/CN107479093A/en
Application granted granted Critical
Publication of CN107479093B publication Critical patent/CN107479093B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. analysis, for interpretation, for correction
    • G01V1/30—Analysis
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. analysis, for interpretation, for correction
    • G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy

Abstract

The invention discloses a kind of micro-seismic event denoising based on potential function and clustering method, comprise the following steps：Micro-seismic event data set to be clustered is imported in Matlab；Using potential function calculate any one occurrence i gesture value andSet threshold valuesRemove noise micro-seismic event；Cluster centre is obtained according to the gesture value and Furthest Neighbor of proposition；Clustered in this, as K means initial cluster center, and then to micro-seismic event after denoising.This method effectively removes the larger micro-seismic event of position error, while reduce the K means technical problems higher to initial cluster center requirement.The method has the characteristics that potential function is various, removes noise event, readily available global optimum.

Description

A kind of micro-seismic event denoising and clustering method based on potential function
Technical field
The invention belongs to Clustering Analysis Technology field, more particularly, to a kind of micro-seismic event denoising based on potential function and Clustering method.
Background technology
Micro-seismic monitoring has at home and abroad obtained extensively in the field such as mine engineering, oil-gas mining, stability of slope and Tunnel Engineering General application.Wherein, micro-seismic event cluster analysis is significant to geology interpretation of structure and microseism Disaster Assessment.Micro-ly Shaking affair clustering can be taking human as division, but it is had a great influence by subjective factor, and is difficult to handle high-volume data in real time.Therefore, The methods of K-means clusters, Agglomerative Hierarchical Clustering, self-organizing map neural network cluster and Gaussian clustering, is applied to microseism In affair clustering analysis.
Agglomerative Hierarchical Clustering each o'clock will merge two immediate clusters as a cluster, each step, and noise spot is easily each Account for cluster.In addition, Agglomerative Hierarchical Clustering is readily obtained strip result.The essence of SOM clusters is a kind of only input layer-hidden layer Neutral net, its effect of visualization is preferable, but it can still update and face after each input data finds a most like class Near node so that its Clustering Effect is poor compared with K-means.Data to be clustered are considered as K Gauss cluster by GMM clusters, but cluster is not For Gaussian Profile when, Clustering Effect is poor.And K-means cluster with its it is simple, practical the features such as be widely applied.
Weatherill＆Burton (2009) is entered using K-means clusters to Aegea regional earthquakes distribution and fault pattern Research is gone.Rehman etc. (2014) explains Pakistan earthquake disaster, risk and geologic structure by K-means clusters. Morales-Esteban etc. (2014) proposes a kind of K-means clustering methods based on adaptive mahalanobis distance, and have studied Croatia and the seismic profile of Iberia Peninsula.Ramdani etc. (2015) demonstrates straight cloth using K-means cluster centres Subduction zone be present in sieve top arc and Andes.Besheli etc. (2015) is clustered to Iranian different zones earthquake using K-means Omen is analyzed.
Above-mentioned K-means clusters research does not optimize selection to initial cluster center, and noise event is not gone Make an uproar.It can be seen that there is great limitation, it is necessary to study a kind of initial with removal noise, optimization in existing seismic events clustering method The automatic clustering method of cluster centre.
The content of the invention
The technical problems to be solved by the invention are to provide a kind of micro-seismic event denoising based on potential function and cluster side Method, the micro-seismic event clustering method have the characteristics that potential function is various, removes noise event, readily available global optimum.
The technical solution of invention is as follows：
A kind of micro-seismic event denoising and clustering method based on potential function, comprise the following steps：
Step 1：Micro-seismic event data set to be clustered is imported in Matlab；
By micro-seismic event data set U to be clustered1Import in Matlab, U1Refer to the attribute of micro-seismic event to be clustered, it is each Three-dimensional coordinate (X of the attribute of individual micro-seismic event from micro-seismic eventi,Yi,Zi), time of origin tiAnd magnitude MiIn choose Arrive, i represents i-th of micro-seismic event, i=1,2 ..., n, and n is the number of micro-seismic event to be clustered；
Step 2：Using potential function calculate micro-seismic event i gesture value and
Wherein,The gesture value that micro-seismic event j is acted on micro-seismic event i is represented, xi,lExpression event i l-th of property value, xj,lExpression event j l Individual property value；P is micro-seismic event attribute dimension, value be 2~5, Ω be apart from acting factor, value be so that be in power value withEntropy obtain minimum value when corresponding Ω；
U1- i represents data set U1Subtract event i, during micro-seismic event attribute unit difference, potential function must normalizing before calculating Change input data.
Step 3：Set the first threshold valuesRemove noise micro-seismic event；
By gesture value and less than the first threshold valuesMicro-seismic event from U1Middle removal, obtain the microseism data set after denoising U2；
By noise data number determine, willArrange from small to large, typically take data corresponding to the 5%~15% AsValue, that is, remove 5%~15% gesture value and minimum event.Noise gets the small value less, and noise takes large values more.
Step 4：The microseism data set U after denoising is obtained according to gesture value and-Furthest Neighbor2Cluster centre；
The gesture value and-Furthest Neighbor screen cluster centre according to following criterion：
(1) the second threshold values is setBy gesture value and it is more thanMicro-seismic event as possible cluster centre collection U3；
Wherein,WithRespectively microseism data set U2In all micro-seismic events gesture value The first quartile and median of sum；
First quartile：By the after all event gesture values and ascending arrangement the 25%th numeral.
Median：By the after all event gesture values and ascending arrangement the 50%th numeral.
(2) the first cluster centre point m is used as using gesture value and maximum micro-seismic event position1；
(3) data set U is taken3-m1In with m1Apart from farthest micro-seismic event position as the second cluster point m2；
(4) data set U is calculated3-m1-m2In each micro-seismic event position and point m1、m2Distance, take each microseism thing Part position and point m1、m2Between small distance as data set V3, then take V3Micro-seismic event position corresponding to middle maximum range value Put as the 3rd cluster point m3；
(5) data set U is calculated3Remove each micro-seismic event position after current all cluster points and current each cluster The distance between point, the small distance between each micro-seismic event position and each cluster point is taken as data set Vk, then take in V Micro-seismic event position corresponding to maximum range value is as kth cluster point mk；
(6) judge whether current cluster point number is K, if so, then obtaining all initial cluster center points, otherwise, is repeated (5), K is that the cluster centre of setting is counted out；
Step 5：The cluster centre point m obtained with step 41,m2,…,mKAs K-means initial cluster center, to going Micro-seismic event clusters after making an uproar, and exports cluster result.
Beneficial effect
A kind of micro-seismic event denoising and clustering method based on potential function provided by the invention, mainly solve typical K- Not the problem of means clustering algorithms do not remove noise micro-seismic event, initial cluster center randomly selects.This method includes following step Suddenly：Micro-seismic event data set to be clustered is imported in Matlab；Using potential function calculate any one occurrence i gesture value and Set the first threshold valuesRemove noise micro-seismic event；Cluster centre is obtained according to the gesture value of proposition and-Furthest Neighbor；In this, as K-means initial cluster center, and then micro-seismic event after denoising is clustered.This method effectively removes position error compared with Big micro-seismic event, while reduce the K-means technical problems higher to initial cluster center requirement.The method has gesture Function is various, removes the features such as noise event, readily available global optimum.
Brief description of the drawings
Fig. 1 is the method for the invention flow chart.
Fig. 2 is microseism data coordinates (Y, X) figure to be clustered.
Fig. 3 is the gesture value and isodensity line chart of microseism data to be clustered.
Fig. 4 is microseism data coordinates and initial cluster center figure after denoising.
Fig. 5 is patent cluster figure compared with typical K-means Clustering Effects.
Embodiment
Below in conjunction with accompanying drawing 1~5, the invention will be further described.
The method of the invention for micro-seismic event typical case K-means clusters do not consider to remove noise micro-seismic event, The problem of being had a great influence by initial clustering point, propose a kind of micro-seismic event denoising and clustering method based on potential function.The party Method by potential function try to achieve each micro-seismic event gesture value and, recycle gesture value and threshold values to remove noise spot, utilize proposition Gesture value and-Furthest Neighbor obtain cluster centre, in this, as K-means initial cluster centers, obtain more excellent Clustering Effect.
As shown in figure 1, a kind of micro-seismic event denoising and clustering method based on potential function, comprise the following steps：
Step 1：Micro-seismic event data set to be clustered is imported in Matlab；
By micro-seismic event data set U to be clustered1Import in Matlab, U1Refer to the attribute of micro-seismic event to be clustered, it is each Three-dimensional coordinate (X of the attribute of individual micro-seismic event from micro-seismic eventi,Yi,Zi), time of origin tiAnd magnitude MiIn choose Arrive, i represents i-th of micro-seismic event, i=1,2 ..., n, and n is the number of micro-seismic event to be clustered；
Step 2：Using potential function calculate micro-seismic event i gesture value and
Wherein,The gesture value that micro-seismic event j is acted on micro-seismic event i is represented, xi,lExpression event i l-th of property value, xj,lExpression event j l Individual property value；P is micro-seismic event attribute dimension, value be 2~5, Ω be apart from acting factor, value be so that be in power value withEntropy obtain minimum value when corresponding Ω；
U1- i represents data set U1Subtract event i, during micro-seismic event attribute unit difference, potential function must normalizing before calculating Change input data.
Step 3：Set the first threshold valuesRemove noise micro-seismic event；
By gesture value and less than the first threshold valuesMicro-seismic event from U1Middle removal, obtain the microseism data set after denoising U2；
Determined by the number of noise data.WillArrange from small to large, typically take data corresponding to the 5%~15% AsValue, that is, remove 5%~15% gesture value and minimum event.Noise gets the small value less, and noise takes large values more.
Step 4：The microseism data set U after denoising is obtained according to gesture value and-Furthest Neighbor2Cluster centre；
The gesture value and-Furthest Neighbor screen cluster centre according to following criterion：
(1) threshold values is setBy gesture value and it is more thanMicro-seismic event as possible cluster centre collection U3；
Wherein,WithRespectively microseism data set U2In all micro-seismic events gesture value The first quartile and median of sum；
First quartile：By the after all event gesture values and ascending arrangement the 25%th numeral.
Median：By the after all event gesture values and ascending arrangement the 50%th numeral.
(2) the first cluster centre point m is used as using gesture value and maximum micro-seismic event position1；
(3) data set U is taken3-m1In with m1Apart from farthest micro-seismic event position as the second cluster point m2；
(4) data set U is calculated3-m1-m2In each micro-seismic event position and point m1、m2Distance, take each microseism thing Part position and point m1、m2Between small distance as data set V3, then take V3Micro-seismic event position corresponding to middle maximum range value Put as the 3rd cluster point m3；
(5) data set U is calculated3Remove each micro-seismic event position after current all cluster points and current each cluster The distance between point, the small distance between each micro-seismic event position and each cluster point is taken as data set Vk, then take in V Micro-seismic event position corresponding to maximum range value is as kth cluster point mk；
(6) judge whether current cluster point number is K, if so, then obtaining all initial cluster center points, otherwise, is repeated (5), K is that the cluster centre of setting is counted out；
Step 5：The cluster centre point m obtained with step 41,m2,…,mKAs K-means initial cluster center, to going Micro-seismic event clusters after making an uproar, and exports cluster result.
Embodiment
The 1210 microseism data monitored in April, 2014 using Kaiyang phosphorus ore by the use of sand bar mining area IMS microseismic systems as K-means cluster datas based on potential function, micro-seismic event attribute take coordinate (Y, X).Implementation process is as follows：
Fig. 2 is microseism data coordinates (Y, X) figure to be clustered.Known by Fig. 2：Micro-seismic event position has with sensor arrangement Have that good correlation-sensor proximity micro-seismic event is more, but poor spaced point (noise spot) is positioned in the presence of part.This A little noise spots can pollute cluster result and subsequent analysis, therefore must remove these noise spots before clustering.
Fig. 3 is the gesture value and isodensity line chart of microseism data to be clustered.Wherein, gesture value and by formulaWithIt is calculated, isodensity line chart is by Natural in software Surfer Neighbor interpolation obtains.Comparison diagram 3 and Fig. 2 know：In Fig. 2 micro-seismic event concentrate position have larger gesture value and, and The scattered position of micro-seismic event have less gesture value and.Thus, gesture value and threshold values can be setRemove noise event.
Fig. 4 is microseism data coordinates and initial cluster center figure after denoising.Setting removes 10% event, thenFor gesture Value and from small to large data corresponding to arrangement the 10%th,ForKnown by figure, this method eliminates noise spot well.Initially Cluster centre is widely distributed, in the absence of differing close point, and gesture value corresponding to initial cluster center and relatively large.Thus, Initial cluster center should be preferable.
Fig. 5 is figure compared with patent cluster clusters (generating initial cluster center at random) effect with typical K-means.Evaluation Effectiveness indicator for conventional Euclidean distance accumulation andWhen clusters number is identical, Euclidean distance tires out Long-pending and smaller, then Clustering Effect is better.Wherein, K is clusters number, and p is the dimension of event attribute, CkFor k-th of clustering cluster.To Numerical value shown in lower arrow is the difference of typical K-means clustering targets and the K-means clustering targets based on potential function.By scheming Know, the Euclidean distance of the K-means clusters based on potential function is accumulated and is respectively less than or clustered equal to typical K-means, it is seen that one Cluster of the kind based on gesture value and-Furthest Neighbor provides a kind of good analysis method for micro-seismic event cluster.
Embodiments of the invention are the foregoing is only, are not intended to limit the invention, it is all in spirit of the invention and former Within then, change, equivalent substitution, improvement etc., should be included in the scope of the protection.

Claims (4)

1. a kind of micro-seismic event denoising and clustering method based on potential function, it is characterised in that comprise the following steps：
Step 1：Micro-seismic event data set to be clustered is imported in Matlab；
By micro-seismic event data set U to be clustered1Import in Matlab, U1Refer to the attribute of micro-seismic event to be clustered, each is micro- Three-dimensional coordinate (X of the attribute of seismic events from micro-seismic eventi,Yi,Zi), time of origin tiAnd magnitude MiMiddle selection obtains, i I-th of micro-seismic event, i=1,2 ..., n are represented, n is the number of micro-seismic event to be clustered；
Step 2：Using potential function calculate micro-seismic event i gesture value and
Wherein,The gesture value that micro-seismic event j is acted on micro-seismic event i is represented, xi,lExpression event i l-th of property value, xj,lExpression event j l Individual property value；P is micro-seismic event attribute dimension, value be 2~5, Ω be apart from acting factor, value be so that be in power value withEntropy obtain minimum value when corresponding Ω；
Step 3：Set the first threshold valuesRemove noise micro-seismic event；
By gesture value and less than the first threshold valuesMicro-seismic event from U1Middle removal, obtain the microseism data set U after denoising2；
Step 4：The microseism data set U after denoising is obtained according to gesture value and-Furthest Neighbor2Cluster centre；
The gesture value and-Furthest Neighbor screen cluster centre according to following criterion：
(1) the second threshold values is setBy gesture value and it is more thanMicro-seismic event as possible cluster centre collection U3；
Wherein, WithRespectively microseism data set U2In all micro-seismic events gesture value sum One quartile and median；
(2) the first cluster centre point m is used as using gesture value and maximum micro-seismic event position1；
(3) data set U is taken3-m1In with m1Apart from farthest micro-seismic event position as the second cluster point m2；
(4) data set U is calculated3-m1-m2In each micro-seismic event position and point m1、m2Distance, take each micro-seismic event position Put and point m1、m2Between small distance as data set V3, then take V3Make micro-seismic event position corresponding to middle maximum range value For the 3rd cluster point m3；
(5) data set U is calculated3Remove each micro-seismic event position after current all cluster points and current each cluster point Between distance, take the small distance between each micro-seismic event position and each cluster point as data set Vk, then take maximum in V Micro-seismic event position corresponding to distance value is as kth cluster point mk；
(6) judge whether current cluster point number is K, if so, then obtaining all initial cluster center points, otherwise, is repeated (5), K Counted out for the cluster centre of setting；
Step 5：The cluster centre point m obtained with step 41,m2,…,mKAs K-means initial cluster center, after denoising Micro-seismic event clusters, and exports cluster result.
2. according to the method for claim 1, it is characterised in that gesture value and removal noise micro-seismic event, gesture value and-distance Method screens micro-seismic event initial cluster center, and
3. according to the method for claim 1, it is characterised in that the gesture value andEntropy PE calculated according to below equation：
4. according to the method described in claim any one of 1-3, it is characterised in that calculatingBefore, the microseism to participating in calculating After the property value of each generic attribute of event is first normalized, normalized value is recycled to calculateNormalized formula is such as Under：
<mrow> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>*</mo> <mo>=</mo> <mfrac> <mrow> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>-</mo> <mfrac> <mn>1</mn> <mi>n</mi> </mfrac> <mo>*</mo> <munderover> <mi>&amp;Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>x</mi> <mi>i</mi> </msub> </mrow> <msqrt> <mrow> <munderover> <mi>&amp;Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>-</mo> <mfrac> <mn>1</mn> <mi>n</mi> </mfrac> <mo>*</mo> <munderover> <mi>&amp;Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>/</mo> <mrow> <mo>(</mo> <mi>n</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow> </msqrt> </mfrac> </mrow>
Wherein, xi* it is the data after normalization, xiFor the property value of any generic attribute of i-th of micro-seismic event.
CN201710838746.3A 2017-09-18 2017-09-18 A kind of micro-seismic event denoising and clustering method based on potential function Active CN107479093B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710838746.3A CN107479093B (en) 2017-09-18 2017-09-18 A kind of micro-seismic event denoising and clustering method based on potential function

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710838746.3A CN107479093B (en) 2017-09-18 2017-09-18 A kind of micro-seismic event denoising and clustering method based on potential function

Publications (2)

Publication Number Publication Date
CN107479093A true CN107479093A (en) 2017-12-15
CN107479093B CN107479093B (en) 2018-12-21

Family

ID=60584424

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710838746.3A Active CN107479093B (en) 2017-09-18 2017-09-18 A kind of micro-seismic event denoising and clustering method based on potential function

Country Status (1)

Country Link
CN (1) CN107479093B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108107475A (en) * 2018-03-05 2018-06-01 吉林大学 A kind of borehole microseismic denoising method based on experience wavelet transformation and multi-threshold function
CN110146918A (en) * 2019-06-23 2019-08-20 广东石油化工学院 Based on the microseismic event detection method and system for dividing group
CN111158045A (en) * 2020-01-06 2020-05-15 中国石油化工股份有限公司 Reservoir transformation microseism event scattered point clustering analysis method and system
CN114382544A (en) * 2022-01-12 2022-04-22 中国矿业大学 Quantitative analysis method for fracture characteristics of working face overlying strata
CN114563826A (en) * 2022-01-25 2022-05-31 中国矿业大学 Microseism sparse table network positioning method based on deep learning fusion drive
CN114779330A (en) * 2022-04-26 2022-07-22 中国矿业大学 Mining working face main fracture azimuth analysis and prediction method based on microseismic monitoring

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070255500A1 (en) * 2006-04-28 2007-11-01 Saudi Aramco Sound enabling computerized system for real time reservoir model calibration using field surveillance data
US20070255545A1 (en) * 2006-04-28 2007-11-01 Pita Jorge A Automated event monitoring system for online reservoir simulation
CN103760597A (en) * 2013-12-27 2014-04-30 淮南万泰电子股份有限公司 Automatic mine fault identification method
US20160146957A1 (en) * 2014-11-26 2016-05-26 Cgg Services Sa Systems and methods for distributed seismic recording and analysis

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070255500A1 (en) * 2006-04-28 2007-11-01 Saudi Aramco Sound enabling computerized system for real time reservoir model calibration using field surveillance data
US20070255545A1 (en) * 2006-04-28 2007-11-01 Pita Jorge A Automated event monitoring system for online reservoir simulation
CN103760597A (en) * 2013-12-27 2014-04-30 淮南万泰电子股份有限公司 Automatic mine fault identification method
US20160146957A1 (en) * 2014-11-26 2016-05-26 Cgg Services Sa Systems and methods for distributed seismic recording and analysis

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108107475A (en) * 2018-03-05 2018-06-01 吉林大学 A kind of borehole microseismic denoising method based on experience wavelet transformation and multi-threshold function
CN108107475B (en) * 2018-03-05 2019-07-19 吉林大学 Borehole microseismic denoising method based on experience wavelet transformation and multi-threshold function
CN110146918A (en) * 2019-06-23 2019-08-20 广东石油化工学院 Based on the microseismic event detection method and system for dividing group
CN110146918B (en) * 2019-06-23 2020-11-10 广东石油化工学院 Grouping-based microseismic event detection method and system
CN111158045A (en) * 2020-01-06 2020-05-15 中国石油化工股份有限公司 Reservoir transformation microseism event scattered point clustering analysis method and system
CN111158045B (en) * 2020-01-06 2022-02-22 中国石油化工股份有限公司 Reservoir transformation microseism event scattered point clustering analysis method and system
CN114382544A (en) * 2022-01-12 2022-04-22 中国矿业大学 Quantitative analysis method for fracture characteristics of working face overlying strata
CN114382544B (en) * 2022-01-12 2022-11-01 中国矿业大学 Quantitative analysis method for fracture characteristics of working face overlying strata
CN114563826A (en) * 2022-01-25 2022-05-31 中国矿业大学 Microseism sparse table network positioning method based on deep learning fusion drive
CN114563826B (en) * 2022-01-25 2023-03-03 中国矿业大学 Microseismic sparse table network positioning method based on deep learning fusion drive
CN114779330A (en) * 2022-04-26 2022-07-22 中国矿业大学 Mining working face main fracture azimuth analysis and prediction method based on microseismic monitoring
CN114779330B (en) * 2022-04-26 2022-12-27 中国矿业大学 Mining working face main fracture azimuth analysis and prediction method based on microseismic monitoring

Also Published As

Publication number Publication date
CN107479093B (en) 2018-12-21

Similar Documents

Publication Publication Date Title
CN107479093A (en) A kind of micro-seismic event denoising and clustering method based on potential function
CN110362949B (en) Landslide sensitivity evaluation method based on neural network
CN102222365A (en) Method for reconstructing curved surface of complex space
CN104950334A (en) Method and device for predicting reservoir distribution
CN105319585B (en) A kind of method hidden using thin-layers interference amplitude recovery identification of hydrocarbon
CN105738952B (en) A kind of horizontal well region reservoir rock phase modeling method
CN113886917B (en) CNN-LSTM model-based railway line region ground settlement prediction early warning method
CN106570422A (en) Realization method of dynamic distribution of differential privacy noise
CN105574642A (en) Smart grid big data-based electricity price execution checking method
CN104865614A (en) Complicated reservoir fluid identification method based on variable skeleton parameter
CN103176221A (en) Mineralizing abnormal feature identification method based on different geological unit background values and contrast values
CN103412333A (en) Static correction base level determining method
CN104297800B (en) A kind of from phased prestack inversion method
CN103647665B (en) Network traffics tracing analysis method and apparatus
CN103914558A (en) Method for mining space-time aggregation patterns of meteorological elements on basis of space-time statistics
CN104500141B (en) Prominent method for early warning based on ore pressure monitor behavioral characteristics
CN105005649A (en) Potential underground water distribution mapping method
CN104459783B (en) Method and device for recognizing structural trap
CN104991280B (en) A kind of perforated zone system of selection and device based on micro throat structure index
CN109410225A (en) Trees dividing method based on the analysis of multi-layer tree structure
Neves et al. Climatic and geologic controls on the piezometry of the Querença-Silves karst aquifer, Algarve (Portugal)
CN104834811A (en) Sea wave height analysis method
CN105676288B (en) A kind of sand body stacked relation method of discrimination based on reflectance factor analysis
CN106530109A (en) Oilfield development appraisal well decision method based on information value
CN104502962A (en) Method and device for designing shot points

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