CN108919353A - A kind of automatic classification of microseism waveform first arrival-time picks up and preferred method - Google Patents
A kind of automatic classification of microseism waveform first arrival-time picks up and preferred method Download PDFInfo
- Publication number
- CN108919353A CN108919353A CN201810720984.9A CN201810720984A CN108919353A CN 108919353 A CN108919353 A CN 108919353A CN 201810720984 A CN201810720984 A CN 201810720984A CN 108919353 A CN108919353 A CN 108919353A
- Authority
- CN
- China
- Prior art keywords
- arrival
- channel
- time
- time difference
- window
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 9
- 238000007621 cluster analysis Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 238000009331 sowing Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 description 8
- 238000009826 distribution Methods 0.000 description 7
- 238000005457 optimization Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000001914 filtration Methods 0.000 description 6
- 238000010835 comparative analysis Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 241001269238 Data Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 239000004615 ingredient Substances 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 210000003746 feather Anatomy 0.000 description 1
- 244000144992 flock Species 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 238000005304 joining Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000010358 mechanical oscillation Effects 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 238000007619 statistical method Methods 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
- G01V1/303—Analysis for determining velocity profiles or travel times
- G01V1/305—Travel times
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/40—Transforming data representation
- G01V2210/41—Arrival times, e.g. of P or S wave or first break
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)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
Picked up the invention discloses a kind of automatic classification of microseism waveform first arrival-time and preferred method comprising level Four step, i.e. denoising-then fast search-then precision pick-then judge with it is preferred；The Butterworth Butterworth band logical 10-200Hz filter carried using MATLAB carries out denoising to original signal；The peak point in signal is quickly judged using peak value determining method, then using peak point or so 500 sampled points of each passage, STA/LTA method is recycled tentatively to establish first arrival-time position；First arrival-time precision pick has selected two methods, is AIC algorithm and MER algorithm respectively, establishes the first arrival-time position of microseism waveform；Effective channel judges to sketch with preferred process as three steps：Clustering seeks reference channel, non-valid channel is rejected in time difference judgement, envelope curve judgement proposes error path.The present invention can quickly, accurately pick up the first arrival-time of microseismic signals, and can judge automatically and reject interference or error larger passage.
Description
Technical field
The present invention relates to structural applications, in particular to the automatic classification of a kind of microseism waveform first arrival-time pick up with it is excellent
Choosing method.
Background technique
Microseism technology be monitoring, the exploitation of mineral resources of early warning deep, geotechnical engineering induce bump, rock burst hazard it is effective
How means, further promote that microseism technology is quick, accurate analysis ability, such as microseism automatic positioning, unattended quick early warning,
Firstly the need of this important prerequisite of the accurately pickup of solution microseismic signals first arrival-time.Quickly, accurate automatic Picking first arrival-time
It is the key that the microseism data realized quickly handles, promotes calamity emergency processing.
Mine microquake monitoring or ground field micro seismic monitoring are different from seismic monitoring, and since monitoring range is small, P, S wave are difficult to
It distinguishes, single component sensors are in the majority, therefore it is most important how to carry out the pickup of P wave.And microseism data is more vulnerable to work condition environment
Influence, signal component is complicated, frequency tends to high frequency, and the Various types of data quantity triggered is big, in addition scene to microseism data at
Timeliness and positioning accuracy request are managed, this brings great difficulty to the processing and analysis of field engineering personnel.On the one hand, different
In conventional mechanical oscillation signal, the interference signal of engineering site is many kinds of, feature is different, not only there is site operation equipment shadow
It rings, Human disturbance, there are also the transmission interferences of microseismic system itself, meanwhile, the heterogeneity of propagation medium also brings along interference, these
The presence of background interference, so that microseismic signals component content is complicated, it is not single, periodic.Theoretical method is believed in ideal
Number or high s/n ratio signal in the better effects that can go, but have in Low SNR signal and acquire a certain degree of difficulty, accuracy of identification needs
It improves, domestic and international expert has carried out numerous studies thus, as shown in Figure 1, for common preliminary wave then pick-up method comparative analysis.
Summary of the invention
To solve the problems, such as above-mentioned background technique, the purpose of the present invention is to provide a kind of microseism waveform first arrivals to arrive
When automatic classification pick up and preferred method, to reach the first arrival-time that can quickly, accurately pick up microseismic signals, and can be automatic
Judgement and the purpose for rejecting interference or error larger passage.
In order to achieve the above objectives, technical scheme is as follows：
A kind of automatic classification of microseism waveform first arrival-time picks up and preferred method comprising level Four step, i.e., at denoising
Reason-then fast search-then precision pick-then judge with it is preferred；
The first order：Denoising is filtered using the Butterworth Butterworth band logical 10-200Hz that MATLAB is carried
Device carries out denoising to original signal；
The second level：Then fast search quickly judges the peak point in signal using peak value determining method, then utilizes peak value
Point or so 500 sampled points of each passage, recycle STA/LTA method tentatively to establish first arrival-time position；
Third pole：Then precision pick, first arrival-time precision pick have selected two methods, are AIC algorithm and MER respectively
Algorithm establishes the first arrival-time position of microseism waveform；
The fourth stage：Then with preferably, effective channel judges to sketch with preferred process as three steps for judgement：Clustering
Seek reference channel, non-valid channel is rejected in time difference judgement, envelope curve judgement proposes error path.
Preferably, the then fast search is specially：It is searched in entire 5000ms signal first with the when window of 500ms
Rope, when window be not overlapped, for the first time when window starting point of the end sampled point as second of search window, quickly searched in entire signal
Rope；While search, the sum of amplitude absolute value in computation window, and be averaged, obtain a series of search characteristics values；Place
After complete signal of reason, the maximum value position of characteristic value is sought, as center, window two sides extend when secondary to this respectively, obtain
One 1500ms it is big when window.
Preferably, the clustering seeks reference channel and carries out space cluster analysis to arrival time difference, i.e. k=2 selects number
According to one kind more than individual as main cluster, and best reference point will be asserted apart from the smallest point A with the poly- heart in the cluster；Time difference judgement
Reject non-valid channel in choose reference channel A as reference channel, calculate separately other each channels and A channel space length and
The practical then time difference, can calculating the theoretical maximum in A and any channel by space length and calibration velocity of wave, then the time difference, comparison are managed
By the size of the then time difference and the practical then time difference, judgment threshold is selected, the big channel of duration error can be rejected, obtained just
Walk the result of screening；Envelope curve judgement proposes that error path selects to trigger at first from front portion primary election result, i.e., then
The smallest channel C0For reference channel, by this point then on the basis of, seek other each channel Csi(i is except zero each logical
Road number) with the practical arrival time difference in this channel, radius value R is sought with the velocity of wave that arrival time difference and calibration big gun obtaini, then with radius
RiCircle is drawn, for the microseismic event in internal field, spatial position is necessarily in circle or on circle, and outfield event can pass through
Primary Location seeks focus microseism and each station distance Ui, as sum (Ui) it is minimum when, as optimum position is then counter to push away Ri。
Preferably, the channel for participating in calculating is screened in the way of preliminary establishment hypocentral location, computation model is adopted
Location Calculation is carried out with simplex method, the positioning result that primary Calculation goes out is O ' (xf, yf, zf, tf), the then focal point and each
Propagation time between standing can be expressed as：
When to calculate best focus away from method be actually to seek the minimum value of the sum of arrival time difference λ, i.e.,
Wherein N is station quantity, and Ri is the distance between hypocentral location and the station, i.e. radius.
Through the above technical solutions, a kind of microseism waveform first arrival-time provided by the invention automatic classification pick up with preferably
Method, can quickly, accurately pick up the first arrival-time of microseismic signals, and can judge automatically and reject interference or error is larger logical
Road.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described.
Fig. 1 is to commonly use preliminary wave then pick-up method comparative analysis figure in the prior art；
Fig. 2 is that AIC of the present invention calculating preliminary wave is then schemed；
Fig. 3 is first arrival-time of the present invention classification pickup and preferred flow charts；
Fig. 4 is the station of the present invention and hypocentral location floor map；
Fig. 5 is microseism waveform and its time-frequency figure comparison diagram before and after bandpass filtering of the present invention；
Fig. 6 is each channel waveform of the typical microseismic event of the present invention and then picks up result figure；
Fig. 7 is that the present invention then scheme by cluster and establishing for reference channel；
Fig. 8 is the theoretical maximum of the present invention then time difference, practical time difference comparison diagram；
Fig. 9 is the envelope curve figure of internal field event distribution of the present invention；
Figure 10 is automatic Picking result schematic diagram of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description.
A kind of automatic classification of microseism waveform first arrival-time provided by the invention picks up and preferred method, as shown in figure 3, its
Including level Four step, i.e., denoising-then fast search-then precision pick-then judge with preferably；
The first order：Denoising, it is contemplated that this feature of the frequency range Relatively centralized of mine microquake signal utilizes
MATLAB included Butterworth Butterworth band logical 10-200Hz filter carries out denoising to original signal；
The second level：Then fast search quickly judges the peak point in signal using peak value determining method, then utilizes peak value
Point or so 500 sampled points of each passage, recycle STA/LTA method tentatively to establish first arrival-time position, specially：First with
The when window of 500ms is searched in entire 5000ms signal, when window be not overlapped, for the first time when window end sampled point as searching for the second time
The starting point of rope window is quickly searched in entire signal；While search, the sum of amplitude absolute value in computation window, and
It is averaged, obtains a series of search characteristics values；After complete signal of processing, the maximum value position of characteristic value is sought, as
Center, window two sides extend when secondary to this respectively, obtain a 1500ms it is big when window, in Fig. 2 (a), shown in (b), red back
Window when scape box is locating for tentatively established initial time, red dotted line is the curve of search characteristics value in Fig. 2 (b)；
Third pole：Then precision pick, first arrival-time precision pick have selected two methods, are AIC algorithm and MER respectively
Algorithm further establishes the first arrival-time position of microseism waveform, and as shown in Fig. 2 (c), red locations are that the above method picks up
Obtained first arrival-time and end time position；
The fourth stage：Then with preferably, effective channel judges to sketch with preferred process as three steps for judgement：Clustering
Seek reference channel, non-valid channel is rejected in time difference judgement, envelope curve judgement proposes that error path, the clustering are sought joining
Space cluster analysis is carried out to arrival time difference according to channel, i.e. k=2 selects one kind more than data individual as main cluster, and will be in the cluster
Best reference point is asserted apart from the smallest point A with the poly- heart；Time difference judgement, which is rejected, chooses reference channel A as ginseng in non-valid channel
According to channel, calculate separately other each channels and A channel space length and the practical then time difference, by space length and calibration velocity of wave
The theoretical maximum in A and any channel then time difference, the size of the comparison theoretical then time difference and the practical then time difference, choosing can be calculated
Determine judgment threshold, the big channel of duration error can be rejected, obtain the result of preliminary screening；Envelope curve judgement proposes error
Channel selects to trigger at first from front portion primary election result, i.e. then the smallest channel C0For reference channel, put with this
On the basis of then, other each channel Cs are soughtiThe practical arrival time difference of (i is each channel number except zero) with this channel, with
The velocity of wave that arrival time difference and calibration big gun obtain seeks radius value Ri, then drawn and justified with radius Ri, for the microseismic event in internal field,
Spatial position is necessarily in circle or on circle, and outfield event can seek focus microseism and each station by Primary Location
Distance Ui, as sum (Ui) it is minimum when, as optimum position is then counter to push away Ri, to ginseng in the way of preliminary establishment hypocentral location
It is screened with the channel of calculating, computation model carries out location Calculation using simplex method, and the positioning result that primary Calculation goes out is
O ' (xf, yf, zf, tf), then the propagation time between the focal point and each station can be expressed as：
When to calculate best focus away from method be actually to seek the minimum value of the sum of arrival time difference λ, i.e.,
Wherein N is station quantity, and Ri is the distance between hypocentral location and the station, i.e. radius.
Its judgment method can be sketched：If the then error between focal point and the station is smaller, show that the station participates in
The reliability of location Calculation is higher；, whereas if the then error between calculated focal point and the station is bigger, show this
The reliability stood is lower.
Under propagation medium processing condition, using station position as the center of circle, station distance at focus is that radius draws circle, then respectively
The circular curve of the station polymerize at focus, as shown in figure 4, blue △ is the monitoring station, red * is focus, due to propagation medium
Unevenly, velocity of wave changes, which exists only among theory.But consider that above-mentioned factor influences, give certain Δ R more than needed,
Radius R+ Δ R is fabricated, the validity of duration is determined with this.
Example checking computations：
In order to verify the above method, a microseismic event 2011-10-07T13 being monitored with Hebei mine microseismic system:
For 42: 17, above-mentioned algorithm is verified.Experimental place is the mine stope, and air way, fortune lane are total to cloth in working face
12 microseismic sensors are equipped with, parameter information carries out real-time dynamic monitoring referring to table 1, for working the working face extraction.
The live station arrangement parameter of table 1
Bandpass filtering Bandfilter
Field data statistical analysis shows that the frequency range of mine microquake signal is influenced by many factors, such as transmission distance
From, enclosing lithologies, focus itself etc., the presence of these ingredients carrys out difficulty to then pick-up tape, while also affecting the essence of pickup
Degree.Therefore, before being picked up when proceeding to, it is necessary to carry out denoising to microseism data first, bandpass filtering is believed in earthquake
It is especially the most famous with Butterworth filter (Butterworth) among these using relatively broad in number denoising.Utilize MATLAB
Worked out Bandfilter bandpass filter, it is contemplated that above-mentioned factor, be arranged useful signal frequency distribution range be 10~
200Hz is handled microseismic signals using filter.Fig. 5 show the forward and backward effect of typical microseismic signals bandpass filtering,
As can be seen that the main component and Energy distribution of signal are in 100~200Hz range.Original signal is in high-frequency region before filtering
There are also sparse distributions, and high-frequency region ingredient has been eliminated after denoising.Be obtained by calculation denoising after signal Signal to Noise Ratio (SNR)=
35.6126db, Signal-to-Noise is increased dramatically after showing denoising.
First arrival-time picks up
Fig. 6 is the waveform diagram and first break pickup result in each channel of the event.It is denoted in figure in 12 channels of secondary event
" when m- amplitude " curve, the first arrival-time that red straight line is picked up by context of methods.Station1~12 respectively indicate
12 monitoring sensors.As can be seen from Fig., under the conditions of certain signal-to-noise ratio, context of methods can be arrived with quick pick-up first arrival
When, such as No. 11 channel of Station1,2,3,5,6,7,8,9 and in figure, and Station4,10,12 fail accurately to pick up then.
Noise is relatively low in No. 4 channels among these, and it is 0.003s that this paper algorithm, which picks up then, indicates that the secondary pickup can not be accurately obtained
When, this illustrates the pickup that noise compares so as to see who is superior when seriously affecting；Without effective waviness in 10 and No. 12 channels, show above-mentioned
It is not triggered in event generation time in two channels.Above-mentioned channel pick up then and amplitude is as shown in table 2 below.In channel 10 and 12
Fail effectively to trigger, is 1s by the first arrival-time that model is calculated.
The first arrival-time of waveform in each channel of table 2
From Fig. 6 and table 2 as can be seen that the height of signal-to-noise ratio is then affected to preliminary pickup, signal-to-noise ratio is higher, picks up
It is then about accurate, conversely, the first arrival of signal-to-noise ratio or abnormal signal is difficult accurately to pick up or judge.If Fig. 6 plants 4,6,10,12, this
Be can directly be observed by naked eyes, if but then picking algorithm return value, and without judgement and preferably, this will be to tight
Ghost image rings subsequent location Calculation precision.After first arrival-time picks up, respectively arrives duration and will be brought among location model and carry out
Focus calculates.In upper figure, in addition to No. 11 channel of Station1,2,3,5,6,7,8,9 and, have by being visually tentatively judged as
Effect, the other three channel is invalid.
Effective channel is preferred
The judgement and selection in effective channel, actually select each initial time, reject invalid or error
Biggish first arrival value, its purpose is to guarantee the precision of location Calculation.
(1) selection of reference channel
Since sensor is in same working face, arranged by interval 30~50m spacing, therefore, in platform network arrangements
For microseismic event, the time interval of trigger sensor is little.If carrying out cluster calculation to these triggering moments, k- is selected
Means clustering algorithm (is calculated) by k=2, can reject the big channel of application condition first.The selection of reference channel is by poly-
What alanysis method was realized, as shown in table 3, Fig. 7,12 initial times are gathered for two classes, red " * " number is cluster matter in Fig. 7
The heart (the poly- heart).
Each website first arrival-time cluster result of table 3
From Fig. 7 (a) as can be seen that in addition to 10,12 channels, rest channels are one kind.Due in A cluster element far more than
B cluster, therefore, the A class for selecting initial time more continues to study, and with (all in the cluster to arrive from the heart of birdsing of the same feather flock together in this cluster
When average value) nearest point (station) is used as object of reference, reject B class.When two cluster number of elements maintain an equal level, may be selected two clusters
Center as the poly- heart, seek at a distance from each station.Each element is shown in Fig. 7 (b) at a distance from the poly- heart (A cluster), it can
To find out, the corresponding distance of 8# is minimum, and distance is 7.57 × 10-4, therefore, select No. 8 as the reference station.
(2) rejecting of non-valid channel
After carrying out first step processing, it is also necessary to solve the problems, such as that first arrival-time value is wrong or error is larger.Fig. 8 is
The reality of Station 8 and other stations then time difference Δ TA (Actual Value), theoretical maximum then time difference Δ TT
(Theoretical Max Value) comparison diagram.
It can be seen from the figure that then time difference Δ TT is far smaller than reality then to the theoretical maximum in 4,10,12 3 channels
Time difference Δ TA.The relationship of the time difference and the practical then time difference when being up to for description theory, given threshold λ=Δ TA/ Δ TT, when
Then judge the channel for duration error larger passage when λ > 10.Then the relevant parameter calculated result of the time difference is as shown in table 4.
The then time difference in table 4 channel 8 and other channels
It can be seen that the exception of the λ value in channel 4,10 and 12 from upper table calculated result, be far longer than theoretical maximum, this
Be do not meet it is actual.Therefore, these three channels is useless road to duration, and the positioning in later period should not be participated in by then picking up result
It calculates, should reject these three channels first arrives duration.In addition to rejecting then abnormal passage, it is also contemplated that duration error is larger
Channel.
(3) rejecting in the big channel of error
Reject the 4th, 10, behind 12 channels, totally 9 channels of residue 1~3,5~9 and 11 recycle envelope curve judgement
Method optimizes selection to remaining channel.Before drawing envelope curve, need to calculate the radius of envelope curve.Radius value Ri can be by
True arrival time difference Δ TT propagates velocity of wave multiplied by microseism wave and acquires, and calculated result is as shown in table 5.Envelope curve figure as shown in figure 9,
It can be seen from the figure that focus (red " * " labelled notation) when each envelope curve is distributed in shake closely, mostly intersects shape two-by-two.
5 envelope curve maximum radius value of table
Situation shown in comparison diagram 4 is combined with the theoretical analysis of it is found that the public domain of envelope intersection is hypocenter distributing
Region.This conclusion is equally confirmed in Fig. 9, in addition to channel 6, focus is wrapped in interior by other envelopes.It can be seen that logical
Road 6 is not suitable for participation location Calculation to duration.From the point of view of waveform angle, the doubtful electric pulse waveform of waveform in channel 6, this is also straight
It connects and confirms the channel to the undesirable of duration.
Above-mentioned analysis the result shows that, the first arrival-time in 4,10, No. 12 channels can not detect accurately, No. 6 channels are then missed
Difference is larger, should not participate in the location Calculation of the event, as non-valid channel processing.1~3,5,7~9 and No. 11 channel is
Therefore the location Calculation that above-mentioned channel participates in the event is finally chosen in effective channel.
Comparative analysis
Be picked up using first arrival-time of the method that manually picks up to above-mentioned event, and comparative analysis manually pick up with automatically
The result of pickup.By comparison, manually with the pickup precision of automated process and time-consuming comparison as shown in following table table 6.By surveying,
Time-consuming 3min is picked up to the first arrival-time of above-mentioned 12 channel waveform, and seeks only needing 0.039s using calculating.
Table 6 manually picks up and picks up Comparative result with context of methods
Live data analysis and application
In order to which the practical application effect to above-mentioned hierarchical detection and picking algorithm is judged, mine scene microseism thing is chosen
It picks up and calculates when part (265 groups of Wave datas) proceeds to, the calculating environment used is：10 professional version of operating system Windows,
Processor Interl (R) Core (TM) i7-7700K, CPU4.20GHz, memory 16GB.
Then picking algorithm is examined
For the accuracy of quantitative contrast several method, above-mentioned 265 groups of data are carried out manually to pick up (total time-consuming
0.828s), and to manually pick up then as reference, this method is compared and analyzed with it, the results are shown in Figure 10.Its
In, Figure 10 (a) is first arrival automatic Picking result and the distribution relation for manually picking up result, and Figure 10 (b) is that then error distribution is straight
Side's figure (only showing result in 50ms error range).
It can be seen that from two figures：Using above-mentioned proposed pick-up method, pickup result is close with result is picked up by hand,
A large amount of automatic Picking results (red No. *), which are fallen into, to be manually picked up within Blue circles；It can also be seen by final error analysis
This conclusion out, the distributive law such as error in the section -5~5ms reach 71.7% (190 groups), the distribution between -10~10ms
Rate reaches 83.4% (221 groups), and the distributive law between -20~20ms reaches 90.57% (240 groups).This result demonstrates again that
Above-mentioned proposed then pick-up method pickup precision with higher, can replace manual picking method substantially.
Therefore, generally speaking, the classification pick-up method that the application is proposed improves the essence of automatic Picking to a certain extent
Exactness can substantially improve this phenomenon manually picked up.
Comparison before and after positioning result
In order to verify influence of the CHANNEL OPTIMIZATION to final location Calculation result, using aforementioned calibration big gun as reference, comparison is true
Value and the difference for optimizing forward and backward calculated value, the superiority-inferiority of hierarchical detection proposed in text and pick-up method is judged with this.It is selected
The localization method selected is four or four integrated positioning methods, and the result finally calculated is as shown in table 7.Before optimization, all channels is selected to arrive
Shi Jinhang is calculated；After optimization, the channel of optimum option is selected then to participate in location Calculation.
As can be seen from the table, the selection in different channels has larger impact to final calculation result, shakes before optimization with true
Source error is about 21.12m, and error is decreased to 9.41m after optimizing.Before optimization, pick up in a manual manner then with effective channel
Have two big defects, first is that time-consuming, second is that can not biggish channel different to then value difference screen, this will seriously affect microseism
The processing speed of data；And by optimization, it can be with quick pick-up then, and the channel for participating in calculating preferably, ensure that
The reasonability for participating in calculating channel, reduces calculating error.
The positioning result comparison of the optimization of table 7 front and back
Using above-mentioned proposed method, solves the problems, such as the pickup of mine microquake wave first arrival-time to a certain extent, and be
Microseism location Calculation provides premise.
It is verified, is drawn the following conclusions by theory analysis and case history：
(1) bandpass filtering can effectively remove the interference component in mine microquake signal, improve the signal-to-noise ratio of microseismic signals,
It lays the foundation for the pickup of first arrival-time.
(2) quickly being judged using peak-peak and STA/LTA can be with the initial time of fast searching microseismic signals, after this is
Continuous precision pick provides help, and greatly reduces calculation amount and the time of data processing.
(3) it improves AIC method first break picking then, solves that multiple wave crests occurs in characteristic value or trough is difficult to establish first arrival
Problem, meanwhile, in conjunction with STA/LTA algorithm, improves and then pick up computation rate.Two methods complement each other, and effectively increase
The accuracy and speed then picked up.
(4) 265 groups of scene Wave datas are utilized, the comparative analysis to artificial pickup and two class method of automatic Picking, as a result
Show that picking up result mean error using the above method reaches 71.7% (190 groups) in the distributive law in the section -5~5ms, -10
Distributive law between~10ms reaches 83.4% (221 groups), and the distributive law between -20~20ms reaches 90.57% (240 groups),
In addition, the application proposed method algorithm time-consuming is 0.828s, significantly faster than manually pick up.
A kind of automatic classification of microseism waveform first arrival-time disclosed by the invention picks up and preferred method, can be quick, quasi-
The first arrival-time of microseismic signals is really picked up, and can judge automatically and reject interference or error larger passage.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (4)
1. a kind of automatic classification of microseism waveform first arrival-time picks up and preferred method, which is characterized in that it includes level Four step,
I.e. denoising-then fast search-then precision pick-then judge with preferably；
The first order：Denoising, the Butterworth Butterworth band logical 10-200Hz filter carried using MATLAB are right
Original signal carries out denoising；
The second level：Then fast search quickly judges the peak point in signal using peak value determining method, then left using peak point
500 sampled points of right each passage, recycle STA/LTA method tentatively to establish first arrival-time position；
Third pole：Then precision pick, first arrival-time precision pick have selected two methods, are that AIC algorithm and MER are calculated respectively
Method establishes the first arrival-time position of microseism waveform；
The fourth stage：Then with preferably, effective channel judges to sketch with preferred process as three steps for judgement：Clustering is sought
Reference channel, time difference judgement reject non-valid channel, envelope curve judgement proposes error path.
2. a kind of automatic classification of microseism waveform first arrival-time according to claim 1 picks up and preferred method, feature
It is, the then fast search is specially：Searched in entire 5000ms signal first with the when window of 500ms, when window not
Overlapping, for the first time when window starting point of the end sampled point as second of search window, quickly searched in entire signal；It is searching for
While, the sum of amplitude absolute value in computation window, and be averaged, obtain a series of search characteristics values；Processing complete
After signal, the maximum value position of characteristic value is sought, as center, window two sides extend when secondary to this respectively, obtain one
1500ms it is big when window.
3. a kind of automatic classification of microseism waveform first arrival-time according to claim 2 picks up and preferred method, feature
It is, the clustering seeks reference channel and carries out space cluster analysis to arrival time difference, i.e. k=2 is selected more than data individual
One kind is used as main cluster, and will be asserted best reference point apart from the smallest point A with the poly- heart in the cluster；Time difference judgement is rejected invalid logical
Reference channel A is chosen in road as reference channel, the space length and reality for calculating separately other each channels and A channel are to constantly
Difference can calculate the theoretical maximum in A and any channel then time difference, the comparison theoretical then time difference by space length and calibration velocity of wave
With the size of the practical then time difference, judgment threshold is selected, the big channel of duration error can be rejected, obtain the knot of preliminary screening
Fruit；Envelope curve judgement proposes that error path selects to trigger at first from front portion primary election result, i.e. then the smallest channel
C0For reference channel, by this point then on the basis of, seek other each channel Csi(i be each channel number) except zero and this
The practical arrival time difference in a channel seeks radius value R with the velocity of wave that arrival time difference and calibration big gun obtaini, then with radius RiCircle is drawn, for
For the microseismic event in internal field, spatial position is necessarily in circle or on circle, and outfield event can be asked by Primary Location
Take focus microseism and each station distance Ui, as sum (Ui) it is minimum when, as optimum position is then counter to push away Ri。
4. a kind of automatic classification of microseism waveform first arrival-time according to claim 3 picks up and preferred method, feature
It is, the channel for participating in calculating is screened in the way of preliminary establishment hypocentral location, computation model uses simplex method
Location Calculation is carried out, the positioning result that primary Calculation goes out is O ' (xf, yf, zf, tf), then the biography between the focal point and each station
It can be expressed as between sowing time：
When to calculate best focus away from method be actually to seek the minimum value of the sum of arrival time difference λ, i.e.,
Wherein N is station quantity, and Ri is the distance between hypocentral location and the station, i.e. radius.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810720984.9A CN108919353B (en) | 2018-07-03 | 2018-07-03 | Automatic grading pickup and optimization method for first arrival time of microseismic waveform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810720984.9A CN108919353B (en) | 2018-07-03 | 2018-07-03 | Automatic grading pickup and optimization method for first arrival time of microseismic waveform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108919353A true CN108919353A (en) | 2018-11-30 |
| CN108919353B CN108919353B (en) | 2020-03-24 |
Family
ID=64424378
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810720984.9A Active CN108919353B (en) | 2018-07-03 | 2018-07-03 | Automatic grading pickup and optimization method for first arrival time of microseismic waveform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108919353B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110286408A (en) * | 2019-06-28 | 2019-09-27 | 广东石油化工学院 | A kind of microseismic event detection method and system |
| CN110398775A (en) * | 2019-08-23 | 2019-11-01 | 山东大学 | Tunnel is dashed forward discharge disaster microseismic event signal fluctuation first break pickup method and system |
| CN110441819A (en) * | 2019-08-06 | 2019-11-12 | 五季数据科技(北京)有限公司 | A kind of seismic first break automatic pick method based on mean shift clustering |
| CN110542920A (en) * | 2019-09-03 | 2019-12-06 | 北京云庐科技有限公司 | Seismic data processing method and system |
| CN111175810A (en) * | 2019-07-05 | 2020-05-19 | 中南大学 | Microseismic signal arrival time picking method, device, equipment and storage medium |
| CN112180433A (en) * | 2019-07-04 | 2021-01-05 | 中国石油天然气集团有限公司 | Seismic first-motion wave pickup method and device |
| CN112526602A (en) * | 2020-11-16 | 2021-03-19 | 重庆大学 | P wave arrival time picking method based on long and short time windows and AR model variance surge effect |
| CN113820663A (en) * | 2021-08-02 | 2021-12-21 | 中南大学 | Robust microseismic/acoustic emission source positioning method and system |
| CN115963548A (en) * | 2023-01-16 | 2023-04-14 | 中国矿业大学 | Mine microseismic P-wave arrival time picking model construction method based on inverse deductive learning |
| CN117872464A (en) * | 2024-01-15 | 2024-04-12 | 淮南师范学院 | Accurate time-to-time automatic pickup method based on wave propagation attenuation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995290A (en) * | 2014-06-03 | 2014-08-20 | 山东科技大学 | High-precision automatic microseism P wave seismic phase first arrival pickup method |
| US20150039235A1 (en) * | 2011-12-08 | 2015-02-05 | Halliburton Energy Services, Inc. | Permeability prediction systems and methods using quadratic discriminant analysis |
| US9103935B2 (en) * | 2013-02-04 | 2015-08-11 | King Fahd University Of Petroleum And Minerals | Method of first arrival picking of seismic refraction data |
| CN106646598A (en) * | 2016-12-27 | 2017-05-10 | 吉林大学 | FAST-AIC-algorithm micro-seismic signal collecting method |
-
2018
- 2018-07-03 CN CN201810720984.9A patent/CN108919353B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150039235A1 (en) * | 2011-12-08 | 2015-02-05 | Halliburton Energy Services, Inc. | Permeability prediction systems and methods using quadratic discriminant analysis |
| US9103935B2 (en) * | 2013-02-04 | 2015-08-11 | King Fahd University Of Petroleum And Minerals | Method of first arrival picking of seismic refraction data |
| CN103995290A (en) * | 2014-06-03 | 2014-08-20 | 山东科技大学 | High-precision automatic microseism P wave seismic phase first arrival pickup method |
| CN106646598A (en) * | 2016-12-27 | 2017-05-10 | 吉林大学 | FAST-AIC-algorithm micro-seismic signal collecting method |
Non-Patent Citations (2)
| Title |
|---|
| 崔云洁 等: "一种微震震相到时自动拾取方法", 《山东科技大学学报》 * |
| 朱权洁 等: "微震波自动拾取与多通道联合定位优化", 《煤炭学报》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110286408A (en) * | 2019-06-28 | 2019-09-27 | 广东石油化工学院 | A kind of microseismic event detection method and system |
| CN112180433B (en) * | 2019-07-04 | 2023-08-22 | 中国石油天然气集团有限公司 | Method and device for picking up first arrival wave of earthquake |
| CN112180433A (en) * | 2019-07-04 | 2021-01-05 | 中国石油天然气集团有限公司 | Seismic first-motion wave pickup method and device |
| CN111175810A (en) * | 2019-07-05 | 2020-05-19 | 中南大学 | Microseismic signal arrival time picking method, device, equipment and storage medium |
| CN110441819A (en) * | 2019-08-06 | 2019-11-12 | 五季数据科技(北京)有限公司 | A kind of seismic first break automatic pick method based on mean shift clustering |
| CN110398775A (en) * | 2019-08-23 | 2019-11-01 | 山东大学 | Tunnel is dashed forward discharge disaster microseismic event signal fluctuation first break pickup method and system |
| CN110542920A (en) * | 2019-09-03 | 2019-12-06 | 北京云庐科技有限公司 | Seismic data processing method and system |
| CN110542920B (en) * | 2019-09-03 | 2021-06-22 | 北京云庐科技有限公司 | Seismic data processing method and system |
| CN112526602A (en) * | 2020-11-16 | 2021-03-19 | 重庆大学 | P wave arrival time picking method based on long and short time windows and AR model variance surge effect |
| CN112526602B (en) * | 2020-11-16 | 2023-10-20 | 重庆大学 | P-wave arrival time pickup method based on long and short time windows and AR model variance surge effect |
| CN113820663A (en) * | 2021-08-02 | 2021-12-21 | 中南大学 | Robust microseismic/acoustic emission source positioning method and system |
| CN113820663B (en) * | 2021-08-02 | 2024-04-16 | 中南大学 | Robust microseism/acoustic emission source positioning method and system |
| CN115963548A (en) * | 2023-01-16 | 2023-04-14 | 中国矿业大学 | Mine microseismic P-wave arrival time picking model construction method based on inverse deductive learning |
| CN115963548B (en) * | 2023-01-16 | 2024-01-23 | 中国矿业大学 | Mine microseismic P wave arrival time pickup model construction method based on deduction learning |
| CN117872464A (en) * | 2024-01-15 | 2024-04-12 | 淮南师范学院 | Accurate time-to-time automatic pickup method based on wave propagation attenuation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108919353B (en) | 2020-03-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108919353A (en) | A kind of automatic classification of microseism waveform first arrival-time picks up and preferred method | |
| CN110458055B (en) | Obstacle detection method and system | |
| CN104459779B (en) | A kind of method and apparatus of the automatic identification in Unusual Seismic road | |
| CN102183773B (en) | Method for filtering ineffective drift signal of global positioning system (GPS) equipment | |
| CN108037770A (en) | Unmanned plane power transmission line polling system and method based on artificial intelligence | |
| CN107656247A (en) | Three-dimensional surveillance radar Target dots processing method based on AESA technical system | |
| CN107219554B (en) | The automatic obtaining method of the Value of residual static correction of land seismic data | |
| CN104714225B (en) | Dynamic programming tracking-before-detection method based on generalized likelihood ratios | |
| CN109934811A (en) | A kind of optical element surface defect inspection method based on deep learning | |
| CN105261136B (en) | The method and device of weather interference is shielded in a kind of fiber-optic monitoring warning system | |
| CN108919193A (en) | A kind of parameter agile radar signal sorting method excavated based on sequence fragment | |
| CN108226869A (en) | A kind of detection method of the radar signal of PRF staggering | |
| CN105445699B (en) | The distance measuring method and system that a kind of non-market value eliminates | |
| CN104122587A (en) | Abnormity first appearance recognition method and system based on offset domain | |
| CN107490808A (en) | A kind of method for building up of high reliability seismic prospecting observation system | |
| CN106199561A (en) | The processing method and processing device of impulse wave | |
| CN104570076A (en) | Automatic seismic wave first-arrival picking method based on dichotomy | |
| CN106872973A (en) | The method monitored using CINRAD/SA weather radars and track forest fires echo | |
| CN105093319A (en) | Ground micro-seismic static correction method based on three-dimensional seismic data | |
| CN104977615B (en) | A kind of multiple ripple drawing method of deep water OBC data based on modeling statistics pickup | |
| CN102508271B (en) | Navigation satellite signal capturing method based on peak value location comparison | |
| CN106199705B (en) | The detection method of seismic acquisition data | |
| CN107132515A (en) | A kind of point mark screening technique constrained based on multidimensional information | |
| CN105675909A (en) | Burst firing bullet signal identification method for indoor target channel light screen velocity measurement apparatus | |
| CN104977602A (en) | Control method and apparatus for earthquake data acquisition construction |
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 |