CN106772606A - A kind of new method that effective identification is carried out to ground fissure - Google Patents
A kind of new method that effective identification is carried out to ground fissure Download PDFInfo
- Publication number
- CN106772606A CN106772606A CN201710006085.8A CN201710006085A CN106772606A CN 106772606 A CN106772606 A CN 106772606A CN 201710006085 A CN201710006085 A CN 201710006085A CN 106772606 A CN106772606 A CN 106772606A
- Authority
- CN
- China
- Prior art keywords
- training
- gas chimney
- ground fissure
- chimney body
- seismic
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 37
- 230000007935 neutral effect Effects 0.000 claims abstract description 13
- 238000005070 sampling Methods 0.000 claims abstract description 9
- 230000002708 enhancing effect Effects 0.000 claims abstract description 8
- 238000001914 filtration Methods 0.000 claims abstract description 7
- 238000012360 testing method Methods 0.000 claims description 6
- 230000002547 anomalous effect Effects 0.000 claims description 2
- 239000003086 colorant Substances 0.000 claims description 2
- 239000000284 extract Substances 0.000 claims description 2
- 238000010943 off-gassing Methods 0.000 claims description 2
- 210000005036 nerve Anatomy 0.000 claims 1
- 210000004218 nerve net Anatomy 0.000 claims 1
- 238000013528 artificial neural network Methods 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005553 drilling Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000005336 cracking Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000003673 groundwater Substances 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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. for interpretation or for event detection
- G01V1/30—Analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/60—Analysis
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The present invention provides a kind of new method being identified to ground fissure, belong to field of geophysical exploration, dip filtering treatment is done to original earthquake data first and obtains ground fissure enhancing filter banks modulation data volume, artificial pickup Gas chimney and non-Gas chimney body training sampling point, then various seismic properties and project training neutral net are extracted at two groups of pickup point sets, satisfied Application of Neural Network will be trained to obtain Gas chimney body in whole seismic data cube.Eventually through the method for the overlapping display of the gentle chimney stack of geological data, the purpose for recognizing and explaining ground fissure is reached, the present invention is improve using the precision of seismic data interpretation ground fissure.
Description
Technical field
The invention belongs to engineering exploration field, and in particular to a kind of effective ways of crack identification over the ground.
Background technology
The abbreviation of " ground fissure " ground fractures.It is top stratum, the soil body in natural cause (crustal motion, effect of water etc.)
Or under human factor (draw water, irrigate, excavating) effect, cracking is produced, and the crack of certain length and width is formed on ground
A kind of macroscopical earth's surface breakoff phenomenon.The geological disaster phenomenon that ground fissure is given birth to as a kind of table, generally deposits in world many countries
Cause ground and underground structure to ftracture in, ground fissure, destroy road surface, bad break groundwater supply, gas pipeline jeopardize cultural relics and historic sites
Safety, causes huge economic losses, brings many inconvenience to resident living, therefore take effective detection method to find out ground fissure
Distribution Characteristics are significant.Due to the disguised and uncertainty of Ground Fractures ' Development, therefore its detection difficulty is larger, if single
Pure use drilling method, a stratum for point is only can be appreciated that due to one borehole, and not only Collate checing cycle is long, costly, Er Qiebu
The drilling hole amount put and reconnoitre that scope is all limited, it is difficult to meet the purpose for investigating thoroughly ground fissure overall space distribution characteristics.Compared to it
Under, that this method has is convenient, economical, quick and visual results the characteristics of, have not on the spatial distribution characteristic for solving ground fissure
Alternative superiority, therefore be a preferably selection using the fine detection that party's law technology carries out ground fissure.
The content of the invention
The ground fissure enhancing filter banks modulation data volume that dip filtering treatment is obtained is done on original earthquake data.It is to use inclination angle
Inclination angle directional filtering is done in control to geological data, improves the lateral continuity of lineups, reduces the random perturbation produced during treatment,
Neutral net is improved using the enhancing filter banks modulation data physical efficiency for including inclination angle and azimuth information extract optimal seismic properties
Accuracy.
The utilization filter banks modulation data volume is calculated and the property set of single seismic channel or many seismic channels is determined, including shaken
Window, change of pitch angle, similitude etc. are when width, phase, relevant, similar, energy, frequency, curvature etc., wherein fiducial time, energy
Important attribute.
Artificial pickup Gas chimney body and non-Gas chimney body training sampling point collection.Be enhancing filter banks modulation data volume and it is relevant,
Carried out on analogue, with reference to some single attribute volumes sensitive to ground fissure in pick process, with artificial experience and areal geology
Based on situation, two groups of training sampling point collection of Gas chimney body and non-Gas chimney body are picked up.
Neutral net is designed for the seismic properties sampling point that Gas chimney area and non-Gas chimney area are extracted.It will be randomly assigned
Attribute data starts physical training condition to training group and test group, and training and adjustment repeatedly are carried out to the network structure, instruction
Practicing implementation status can be tracked during the training period, and be represented with two kinds of indexes：Normal root mean square curve and root mean square error rate
Curve, RMS error rate curves represent total mistake of training group and test group, respectively from 1 (maximum mistake) to 0 (most mistake
By mistake), two curves should all drop during the training period, and when test curve, high expression network is excessively adapted to again, and training should be at this
Stopped before going too far before generation；A typical RMS value is considered as reasonable in 0.8 scope, and 0.8~0.6 is, 0.6~0.4 is
Very well, it is just fabulous less than 0.4；Additionally, weight of each attribute in current training gives different colours and represents in network node：
When the performance in training group reaches minimal error (curve extreme lower position), the network training of optimized results can stop, this
When can terminate training, the ground cleave that " the Gas chimney body " that the neutral net is predicted can more clearly reflect in seismic data cube
Seam spread.
Satisfied neutral net will be trained to be generalized to whole filter banks modulation data volume obtain Gas chimney body and export.Referring to will
The best attributes that can most reflect seismic anomalous volume obtained by neutral net training are applied to whole seismic data cube, obtain gas
Chimney stack.
Display is conciliate outgassing chimney stack and recognizes ground fissure.Be by Gas chimney section with it is relevant, similitude Profile Correlation can
To find out, Gas chimney section not only highlights the vertical feature of Gas chimney body, fracture system, at the same effectively suppressed again it is main by
Low coherence that noise and low wave impedance etc. cause but the feature of non-Gas chimney, Gas chimney section overlap display, energy with seismic profile
It is enough that quickly and accurately karst collapse col umn is explained.
Brief description of the drawings
Fig. 1：It is workflow diagram of the invention.
Fig. 2：It is Gas chimney body section and common seismic section overlapping display figure.
Fig. 3：Scheme for seismic profile shows.
Specific embodiment
Technical scheme and technique effect are described in detail with reference to specific embodiment, but the present invention
Scope be not restricted to specific examples below.
The work area is located at Qi County of Shanxi, first resulting seismic migration data is done with dip filtering treatment and obtains ground fissure
Enhancing filter banks modulation data volume, artificial pickup Gas chimney and non-Gas chimney body training sampling point, then carry at two groups of pickup point sets
Various seismic properties and project training neutral net are taken, the attribute of extraction has：Amplitude, frequency, phase, energy, relevant, similitude
And curvature, satisfied Application of Neural Network will be trained to obtain Gas chimney body in whole seismic data cube.Eventually through geological data
The method of the overlapping display of gentle chimney stack, the purpose that realization is explained to the research area ground fissure.Fig. 2 is cutd open for Gas chimney body
Face and common seismic section overlapping display figure, the position indicated in figure is ground fissure predicted position.Fig. 3 is the final ground explained
Crack achievement, finds by with the actual ground fissure information contrast that discloses, and predicts the outcome and is fitted like a glove with actual result.
Above-described embodiment is not limitation of the present invention, and this area related those of ordinary skill will be apparent to the skilled artisan that not
Departing from can also have multi-texturing or replacement in the situation of basic conception of the invention and scope, all equivalent technical schemes
Should be comprising within the scope of the present invention.Scope of patent protection of the invention is defined by the restriction of claim.
Claims (7)
1. a kind of new method being identified to ground fissure, it is characterised in that including：
The step of dip filtering being done to original geological data and processes；
The step of property set of single seismic channel or many seismic channels is calculated and determined using filter banks modulation data volume；
The step of artificial pickup Gas chimney body and non-Gas chimney body training sampling point；
The step of attribute and one neutral net of project training being extracted at artificial pickup Gas chimney body and non-Gas chimney body；
The step of satisfied neutral net will be trained being generalized to whole filter banks modulation data volume and obtain Gas chimney body and export；
The step of display is conciliate outgassing chimney stack and recognizes ground fissure with this.
2. a kind of new method being identified to ground fissure according to claim 1, it is characterised in that to described original
The ground fissure enhancing filter banks modulation data volume that dip filtering treatment is obtained is done on geological data, is to geological data with pitch angle control
Do inclination angle directional filtering, improve the lateral continuity of lineups, reduce the random perturbation produced during treatment, using including inclination angle
Enhancing filter banks modulation data physical efficiency with azimuth information improves the accuracy that neutral net extracts optimal seismic properties.
3. a kind of new method being identified to ground fissure according to claim 1, it is characterised in that to described using filter
Slope of wave surface data volume calculates and determines the property set of single seismic channel or many seismic channels, including amplitude, phase, relevant, similar, energy
Window, change of pitch angle, similitude etc. are important attributes when amount, frequency, curvature etc., wherein fiducial time, energy.
4. a kind of new method being identified to ground fissure according to claim 1, it is characterised in that artificially picked up to described
Take Gas chimney body and non-Gas chimney body training sampling point collection is carried out on enhancing filter banks modulation data volume and relevant, analogue,
With reference to some single attribute volumes sensitive to ground fissure, based on artificial experience and areal geology situation, pickup in pick process
Two groups of training sampling point collection of Gas chimney body and non-Gas chimney body.
5. a kind of new method being identified to ground fissure according to claim 1, it is characterised in that to described artificial
Attribute is extracted at pickup Gas chimney body and non-Gas chimney body and the step of one neutral net of project training, for Gas chimney area and
The seismic properties sampling point design neutral net that non-Gas chimney area is extracted, it will be randomly assigned attribute data to training group and test
Group, and start physical training condition, training and adjustment repeatedly are carried out to the network structure, training implementation status during the training period may be used
It is tracked, and is represented with two kinds of indexes：Normal root mean square curve and root mean square error rate curves, RMS error rate curves are represented
Total mistake of training group and test group, respectively from 1 (maximum mistake) to 0 (minimal error), two curves are during the training period all
Should drop, when test curve, high expression network is excessively adapted to again, and training should stop before going too far before this generation；Typical one
Individual RMS value is considered as reasonable in 0.8 scope, and 0.8~0.6 is, 0.6~0.4 is fine, just fabulous less than 0.4；This
Outward, weight of each attribute in current training gives different colours and represents in network node：When the performance in training group reaches
During minimal error (curve extreme lower position), the network training of optimized results can stop, and can now terminate training, the nerve net
The ground fissure spread that " the Gas chimney body " that network is predicted can more clearly reflect in seismic data cube.
6. a kind of new method being identified to ground fissure according to claim 1, it is characterised in that to it is described will training
The step of satisfied neutral net is generalized to whole filter banks modulation data volume and obtains Gas chimney body and export, referring to will be by nerve
The best attributes that can most reflect seismic anomalous volume that online training is obtained are applied to whole seismic data cube, obtain Gas chimney body.
7. a kind of new method being identified to ground fissure according to claim 1, it is characterised in that to the display and
The step of explaining Gas chimney body and recognize ground fissure, be by Gas chimney section with relevant, similitude Profile Correlation as can be seen that gas
Chimney section not only highlights the vertical feature of Gas chimney body, fracture system, while effectively having suppressed main by noise and low again
Low coherence that wave impedance etc. causes but the feature of non-Gas chimney, Gas chimney section overlap display with seismic profile, can quickly, standard
Really karst collapse col umn is explained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710006085.8A CN106772606A (en) | 2017-01-05 | 2017-01-05 | A kind of new method that effective identification is carried out to ground fissure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710006085.8A CN106772606A (en) | 2017-01-05 | 2017-01-05 | A kind of new method that effective identification is carried out to ground fissure |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106772606A true CN106772606A (en) | 2017-05-31 |
Family
ID=58949525
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710006085.8A Pending CN106772606A (en) | 2017-01-05 | 2017-01-05 | A kind of new method that effective identification is carried out to ground fissure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106772606A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108279435A (en) * | 2017-12-18 | 2018-07-13 | 中国石油天然气股份有限公司 | Method and device for determining fault section |
| CN109581485A (en) * | 2018-12-04 | 2019-04-05 | 成都捷科思石油天然气技术发展有限公司 | A method of carrying out automatic slit detection directly on pre-stack depth migration seismic data |
| CN110952978A (en) * | 2019-12-20 | 2020-04-03 | 西南石油大学 | Drilling leakage fracture width prediction method based on neural network data mining |
| CN111178320A (en) * | 2020-01-07 | 2020-05-19 | 中国矿业大学(北京) | Geological abnormal body recognition method and model training method and device thereof |
| CN111626092A (en) * | 2020-03-26 | 2020-09-04 | 陕西陕北矿业韩家湾煤炭有限公司 | Unmanned aerial vehicle image ground crack identification and extraction method based on machine learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030200030A1 (en) * | 1998-09-11 | 2003-10-23 | Den Norske Stats Oljeselskap A.S. | Method of processing acquired seismic data |
| WO2011062807A1 (en) * | 2009-11-20 | 2011-05-26 | Exxonmobil Upstream Research Company | Method for geophysical and geological interpretation of seismic volumes in the domains of depth, time, and age |
| CN104614762A (en) * | 2014-12-17 | 2015-05-13 | 中国石油天然气股份有限公司 | Loose sandstone gas reservoir boundary determining method and device |
| CN105277979A (en) * | 2015-10-16 | 2016-01-27 | 中国石油天然气集团公司 | Seismic attribute optimization method and device |
-
2017
- 2017-01-05 CN CN201710006085.8A patent/CN106772606A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030200030A1 (en) * | 1998-09-11 | 2003-10-23 | Den Norske Stats Oljeselskap A.S. | Method of processing acquired seismic data |
| WO2011062807A1 (en) * | 2009-11-20 | 2011-05-26 | Exxonmobil Upstream Research Company | Method for geophysical and geological interpretation of seismic volumes in the domains of depth, time, and age |
| CN104614762A (en) * | 2014-12-17 | 2015-05-13 | 中国石油天然气股份有限公司 | Loose sandstone gas reservoir boundary determining method and device |
| CN105277979A (en) * | 2015-10-16 | 2016-01-27 | 中国石油天然气集团公司 | Seismic attribute optimization method and device |
Non-Patent Citations (4)
| Title |
|---|
| 刘伟 等: ""基于倾角数据体的神经网络气烟囱识别"", 《石油地球物理勘探》 * |
| 易远元 等: "《海洋深水钻井浅层地质灾害识别技术及案例分析》", 30 June 2012, 石油工业出版社 * |
| 杨瑞召 等: ""倾角导向体控制的气烟囱识别技术及其在海拉尔盆地贝尔凹陷中的应用"", 《现代地质》 * |
| 高磊 等: ""渤海A油田气烟囱识别及其研究意义"", 《物探与化探》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108279435A (en) * | 2017-12-18 | 2018-07-13 | 中国石油天然气股份有限公司 | Method and device for determining fault section |
| CN108279435B (en) * | 2017-12-18 | 2020-02-14 | 中国石油天然气股份有限公司 | Method and device for determining fault section |
| CN109581485A (en) * | 2018-12-04 | 2019-04-05 | 成都捷科思石油天然气技术发展有限公司 | A method of carrying out automatic slit detection directly on pre-stack depth migration seismic data |
| CN110952978A (en) * | 2019-12-20 | 2020-04-03 | 西南石油大学 | Drilling leakage fracture width prediction method based on neural network data mining |
| CN111178320A (en) * | 2020-01-07 | 2020-05-19 | 中国矿业大学(北京) | Geological abnormal body recognition method and model training method and device thereof |
| CN111178320B (en) * | 2020-01-07 | 2020-11-17 | 中国矿业大学(北京) | Geological abnormal body recognition method and model training method and device thereof |
| CN111626092A (en) * | 2020-03-26 | 2020-09-04 | 陕西陕北矿业韩家湾煤炭有限公司 | Unmanned aerial vehicle image ground crack identification and extraction method based on machine learning |
| CN111626092B (en) * | 2020-03-26 | 2023-04-07 | 陕西陕北矿业韩家湾煤炭有限公司 | Unmanned aerial vehicle image ground crack identification and extraction method based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106772606A (en) | A kind of new method that effective identification is carried out to ground fissure | |
| Yang et al. | Theories and practices of carbonate reservoirs development in China | |
| CN108119142B (en) | - three-two subregion water-protection coal-mining methods of band of five figures | |
| CN107044277B (en) | Evaluation method for stimulation potential of re-fracturing horizontal wells in low-permeability heterogeneous reservoirs | |
| CN103454686B (en) | Method and system for reservoir prediction based on small-scale sedimentary facies of stratigraphic slice | |
| CN104879103B (en) | Layered water injection effect analysis method | |
| CN105986817B (en) | A method of shale formation engineering dessert for identification | |
| Shi et al. | Snøhvit CO2 storage project: Assessment of CO2 injection performance through history matching of the injection well pressure over a 32-months period | |
| Wang et al. | A numerical simulation study of fracture reorientation with a degradable fiber-diverting agent | |
| CN104989360A (en) | Characterization method of low-permeability reservoir dynamic fracture | |
| CN106437674A (en) | Imitation horizontal well waterflooding well pattern adapting method | |
| Hofmann et al. | Potential for enhanced geothermal systems in low permeability limestones–stimulation strategies for the Western Malm karst (Bavaria) | |
| CN105319585B (en) | A kind of method hidden using thin-layers interference amplitude recovery identification of hydrocarbon | |
| CN109655894A (en) | Carbonate rock ancient stream channel seismic inversion low frequency model construction method and system | |
| CN106014369A (en) | Acquisition method and device for oil and gas reservoir | |
| CN103399345A (en) | Survey method and device for buried-hill crevice distribution | |
| CN111379548B (en) | Method, device and equipment for determining well track and storage medium | |
| Ziarani et al. | Fracture and wellbore spacing optimization in multistage fractured horizontal wellbores: Learnings from our experience on Canadian unconventional resources | |
| Hunt et al. | Geophysical investigations of the Wairakei Field | |
| Li et al. | Geological characterization and numerical modelling of CO2 storage in an aquifer structure offshore Guangdong Province, China | |
| CN106968647A (en) | A kind of preparation method of slit formation Carbonate Reservoir perforation | |
| Sprigg | Petroleum prospects of western parts of Great Australian Artesian Basin | |
| Liu et al. | Improved tight oil productivity through integrated technology deployment on a multipad horizontal-well trial in central China | |
| Yu et al. | Simulation study of novel methods for water reinjection efficiency improvement of a doublet system in guantao sandstone geothermal reservoir | |
| Wessel-Berg et al. | Large scale CO2 storage with water production |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |