CN108319915B - Multi-time-window simplified form identification method for dynamically adjusting rock burst signal threshold - Google Patents
Multi-time-window simplified form identification method for dynamically adjusting rock burst signal threshold Download PDFInfo
- Publication number
- CN108319915B CN108319915B CN201810098506.9A CN201810098506A CN108319915B CN 108319915 B CN108319915 B CN 108319915B CN 201810098506 A CN201810098506 A CN 201810098506A CN 108319915 B CN108319915 B CN 108319915B
- Authority
- CN
- China
- Prior art keywords
- rock burst
- value
- burst signal
- signal
- function
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Abstract
The invention discloses a multi-time window simplified form identification method for dynamically adjusting rock burst signal threshold, which comprises the following steps: acquiring various typical rock burst signals and typical noise signals, and establishing a rock burst signal threshold value dynamic adjustment database; determining necessary parameters and a judgment threshold; automatic identification processing; and (4) periodically carrying out dynamic adjustment on the judgment threshold. The method is suitable for real-time processing of the micro-seismic data, meets the actual requirements of engineering, improves the waveform picking efficiency, reduces the workload of manual identification, and further improves the real-time performance of geological disaster early warning of rock burst, mine earthquake, collapse and the like.
Description
Technical Field
The invention relates to the technical field of microseismic monitoring. In particular to a multi-time window simplified form identification method for dynamically adjusting rock burst signal threshold, which can be widely applied to mineral engineering, hydraulic and hydroelectric engineering, petroleum engineering, geotechnical engineering and underground engineering.
Background
The automatic rock burst signal identification technology is the key of micro seismic source positioning. The existing rock burst signal identification method mainly comprises the following steps: constructing an STA/LTA algorithm of a characteristic function time domain according to the energy and the energy change in the time domain; according to the difference of the waveform characteristics of the rock burst signal and the noise signal, such as a Fisher discrimination method, a fast Fourier transform, a maximum likelihood classification method, a logistic regression and neural network, a morphological fractal dimension, a statistical method, an energy extreme value method and the like. However, most of the methods only have a good filtering effect on the blasting signals, and have little or no effect on the signals of the jumbolter, the electrical signals, the secondary blasting signals and the like which are common in practical engineering, so that the rock burst signals are submerged in a large amount of data, and are difficult to automatically analyze and process through software.
Through practical engineering application, the following three points are found to be considered in the algorithm when the rock burst signal is automatically identified: 1) whether the calculated amount can meet the software and hardware conditions in the real-time processing; 2) how quickly and efficiently the threshold value determines that complex waveforms in the project are satisfied; 3) whether the change of the rock burst signal in the engineering can be adapted.
Therefore, aiming at the problems and considering practical conditions of an algorithm, the invention provides the multi-time window simplified form identification method for dynamically adjusting the rock burst signal threshold, which is suitable for real-time processing of micro-seismic data, meets the actual engineering requirements, improves the waveform pickup efficiency, reduces the manual identification workload, and further enhances the real-time performance of geological disaster early warning of rock burst, mine earthquake, collapse and the like.
Disclosure of Invention
The invention aims to overcome the problems in the prior art, takes practical conditions of an algorithm into consideration, provides a multi-time window simplified form identification method for dynamically adjusting rock burst signal threshold values, is suitable for real-time processing of micro-seismic data, meets the actual requirements of engineering, improves the waveform pickup efficiency, reduces the manual identification workload, and further enhances the real-time performance of geological disaster early warning of rock bursts, mine earthquakes, landslides and the like.
A multi-time window simplified form identification method for dynamically adjusting rock burst signal threshold comprises the following steps:
The rock burst signal generally refers to an elastic wave signal generated by gradual damage in a rock body before the rock burst occurs, and the noise signal generally comprises a burst signal, a jumbolter signal, an electrical signal, a field operation signal and the like. The waveform amplitude and frequency change of each type of rock burst signal/noise signal are similar, and the signal of which the waveform can represent a certain type of rock burst signal/noise signal is a typical signal.
And 2, obtaining the RK function of the typical rock burst signal and the RK function of the typical noise signal.
determining the number a of reference points, and selecting reference point rjReference point rjThe abscissa of (a) is the delay position t0+djThe ordinate is the judgment threshold RjJudgment threshold value RjAt a delay position t0+djRK function value and delay position t of typical rock burst signal of (site)0+djOfDelay position t between RK function values of typical noise signals0+djRK function value and delay position t of typical rock burst signal of (site)0+djThe RK function values of the representative noise signals at are not equal; where a is in {1, 2}, j is the sequence number of the reference point, t0To trigger the moment, djIs the delay length;
in the present invention, the reference point rjMay be 1 or 2, i.e. reference points rjMay be a reference point r1Or as a reference point r1And a reference point r2(ii) a Judging threshold value RjCorresponding to 1 or 2, that is, the judgment threshold R can be set1May be a judgment threshold value R1And a judgment threshold R2。
And 4, identifying the RK function value of the microseismic data to be identified, comprising the following steps of:
step 4.1, reading the generated microseismic data to be identified in real time;
step 4.2, calculating the RK function value of the microseismic data to be identified in real time, wherein the RK function value of the microseismic data to be identified is larger than a trigger threshold R for the first timeqThe corresponding time is the trigger time t0Trigger threshold RqIs a preset value;
step 4.3, calculating the delay position t0+djThe RK function value of the microseismic data to be identified,
at a delay position t0+djIn case the RK function value of a typical rock burst signal is greater than the RK function value of a typical noise signal:
if the position t is delayed0+djThe RK function value of the microseismic data to be identified is larger than a judgment threshold value RjIf so, identifying the microseismic data to be identified read this time as a to-be-identified rockburst signal; otherwise, the microseismic data to be identified read this time is a noise signal;
at a delay position t0+djIn the case where the RK function value of the typical rock burst signal is less than the RK function value of the typical noise signal:
if the position t is delayed0+djThe RK function value of the microseismic data to be identified is less thanJudging threshold value RjIf so, identifying the microseismic data to be identified read this time as a to-be-identified rockburst signal; otherwise, the microseismic data to be identified read this time is a noise signal;
if all delay positions t0+djAnd if the identification results are to-be-identified rockburst signals, the read microseismic data to be identified is the rockburst signal.
In the present invention, the reference point rjMay be 1 or 2, corresponding to delay position t0+djMay be the delay position t0+d1Or may be the delay position t0+d1And a delay position t0+d2. Under the condition that two reference points exist, the identification of delay positions corresponding to the two reference points is both to-be-identified rockburst signals, and the to-be-identified microseismic data read at this time can be determined to be the rockburst signals.
The judgment threshold value R as described abovejThe dynamic adjustment of (2) is realized by the following steps:
step 5.1, establishing a rock burst signal threshold value dynamic adjustment database, storing rock burst signals in the rock burst signal threshold value dynamic adjustment database according to the sequence of the occurrence time, and recording each rock burst signal in the rock burst signal threshold value dynamic adjustment database as X1，X2，…，XpWherein p is the maximum number of rock burst signals in the dynamic adjustment database of rock burst signal threshold, and the rock burst signal with the latest occurrence time is Xp；
In general, dynamically adjusting the maximum number p of rock burst signals in a database by using a rock burst signal threshold value to 400;
if the on-site geological conditions are variable, the maximum number p of rock burst signals in the database can be dynamically adjusted by reducing rock burst signal threshold values in projects such as deep tunnel excavation or large-scale factory building excavation, so that the dynamic adjustment of the threshold values is accelerated;
if the change of the field geological conditions is small, projects such as fixed point long-term monitoring and the like can increase the maximum number p of rock burst signals in the rock burst signal threshold dynamic adjustment database, so that the threshold dynamic adjustment is slowed down.
Step 5.2, dynamically adjusting the rock burst signal threshold valueEach rock burst signal in the database at a delay position t0+djThe array formed by the arrangement of the RK function values is marked as RKj- (k), wherein k is the serial number of the rock burst signal and j is the serial number of the reference point;
at a delay position t0+djIn case the RK function value of a typical rock burst signal is greater than the RK function value of a typical noise signal: dynamically adjusting each rock burst signal in delay position t in database by rock burst signal threshold value0+djThe RK function values are arranged from large to small;
at a delay position t0+djIn the case where the RK function value of the typical rock burst signal is less than the RK function value of the typical noise signal: dynamically adjusting each rock burst signal in delay position t in database by rock burst signal threshold value0+djThe RK function values are arranged from small to large;
step 5.3, determining the selection rate BjSelecting the minimum value satisfying the following formula as the selection rate Bj
Wherein, BzAs a total selection rate, BzThe range value of (A) is 50% -100%;
step 5.4, obtain kjValue, kj＝int(p×Bj) Int is the rounding operation;
if k isjIn < p, then judging threshold value RjIs RKj-(kj) And RKj-(kjAn average value of + 1);
if k isjIf p, the threshold value R is determinedjIs RKj-(kj)；
Step 5.5, acquiring new microseismic data and the number a of rockburst events in the new microseismic data1And the number of noise events b1Are all of the parameters which are known, and are,
according to the judgment threshold value RjAnd (4) automatically identifying and processing the new microseismic data to obtain a rock burst signal and a noise signal and obtain the number a of corresponding rock burst events2And the number of noise events b2(ii) a Arranging newly obtained rock burst signals according to occurrence time and recording as Y1，Y2，…，YqQ is the maximum number of newly obtained rock burst signals, q is less than or equal to p, and the rock burst signal with the latest occurrence time is Yq；
It is generally considered that if the number of rock burst signals/noise signals occurring within a set time is greater than or equal to 4, a rock burst event/noise signal can be determined, and all signals that cannot be an event will be filtered. The set time is generally defined to be 0.5 s.
Step 5.6, calculating the automatic identification accuracy rate E of the new microseismic data rockburst event according to the formula (5)
If the automatic identification accuracy rate E of the new microseismic data rockburst event is larger than the preset automatic identification accuracy rate threshold value R of the rockburst eventbIf not, performing step 5.8; wherein, the rock burst event is automatically identified with the accuracy threshold value RbThe value range of (A) is 50% -100%;
step 5.7, calculating the automatic identification accuracy G of the noise event of the new microseismic data according to the formula (6)
If the automatic identification accuracy G of the noise event of the new microseismic data is larger than the identification accuracy threshold R of the noise eventgThen confirm the judgment threshold RjOtherwise, performing step 5.8; wherein the noise event identification accuracy threshold RgThe value range of (A) is 0-50%;
step 5.8, using newly obtained rock burst signal YsDynamic adjustment of rockburst signal X in database by replacing rockburst signal thresholdsAdding 1 to the value of s, wherein the initial value of s is 1, s belongs to {1, 2 …, (q +1) }, judging whether s is greater than q, if s is greater than q, indicating that the dynamic adjustment of the threshold value fails, setting s as the initial value 1, and returning after the set timeStep 5.1; otherwise, the step 5.2 is returned.
The RK function for typical rock burst signals and RK function for typical noise signals as described above are based on equation (1),
rk (t) ═ r (t) formula (1)
The R function in equation (1) is based on equation (2),
wherein STA (t) is a short time window STA function, LTA (t) is a long time window LTA value, t is time, n is STA short time window length, m is LTA long time window length,
the CF function in equation (2) is based on equation (3),
CF(t)＝Y(t)2-Y (t-1). Y (t +1) formula (3)
Where Y (t) is a function of amplitude.
Compared with the prior art, the invention has the following beneficial effects: aiming at the characteristics of the rock burst signal, the practical conditions of the algorithm are considered, the multi-time window simplified form identification method for dynamically adjusting the rock burst signal threshold is provided, the method is suitable for real-time processing of micro-seismic data, the actual engineering requirements are met, the waveform pickup efficiency is improved, the manual identification workload is reduced, and the real-time performance of geological disaster early warning of rock burst, mine earthquake, collapse and the like is improved.
Drawings
FIG. 1 is a graph of the RK function with the RK function values on the ordinate and time (expressed in number of sample points) on the abscissa;
FIG. 2 shows a reference point r in the example1And a reference point r2A schematic diagram of (a); wherein (a) is a reference point r1Selecting; (b) is a reference point r2Selecting;
FIG. 3 is a graph of identification algorithm validation; wherein (a) is the distribution rule of RK (2040) value density function; (b) the distribution rule of the density function of the RK (2130) value is shown.
Detailed Description
The technical scheme of the invention is further explained in detail by combining the attached drawings as follows:
in order to make the technical means, creation features, working procedures and using methods of the present invention easily understood and appreciated, the present invention will be further described with reference to the following embodiments. The scope of the invention is not limited by the following examples.
The Hongtianshan copper mine is located in Qingyuan Manchu autonomous county in Fushun city of Liaoning province, and is one of typical deep metal mines in China. Dynamic damages such as rock burst, stress collapse and the like at multiple positions of the copper mine deep stope become the primary problems restricting the safe mining of the mine. In the field microseismic monitoring process, a large amount of noise signals including electrical signals, secondary blasting signals, locomotive impact rail signals, jumbolter signals and the like exist in microseismic data due to the fact that stopes and roadways mostly adopt electrical equipment. The present example is illustrated by using microseismic data of a red permeable cuprite deep stope as an example.
A multi-time window simplified form identification method for dynamically adjusting rock burst signal threshold comprises the following steps:
The rock burst signal generally refers to an elastic wave signal generated by gradual damage in a rock body before the rock burst occurs, and the noise signal generally comprises a burst signal, a jumbolter signal, an electrical signal, a field operation signal and the like. The waveform amplitude and frequency change of each type of rock burst signal/noise signal are similar, and the signal of which the waveform can represent a certain type of rock burst signal/noise signal is a typical signal.
In this example, microseismic data is selected from data monitored in real time in a deep stope of Hongtong hillside copper mine within a period of 2016, 6 months and 1 day. From this data, a typical rock burst signal, a typical low amplitude electrical signal, a typical high amplitude electrical signal, a typical short duration electrical signal, a typical long duration electrical signal, a typical secondary burst signal, a typical locomotive impact rail signal, a typical jumbolter signal are selected.
And 2, obtaining the RK function of the typical rock burst signal and the RK function of the typical noise signal.
The RK function of a typical rockburst signal and the RK function of a typical noise signal are based on equation (1)
Rk (t) ═ r (t) formula (1)
Wherein, the R function is an STA/LTA function in the self-contained triggering algorithm of the microseismic monitoring system, and the common calculation mode is based on the formula (2), but not limited to the formula
Wherein STA (t) is a short time window STA function, LTA (t) is a long time window LTA value, t is time, n is STA short time window length, m is LTA long time window length,
the CF function in the formula (2) is a feature function in the self-contained triggering algorithm of the microseismic monitoring system, and the commonly used calculation mode is based on the formula (3), but is not limited to the formula
CF(t)＝Y(t)2-Y (t-1). Y (t +1) formula (3)
Where Y (t) is a function of amplitude.
determining the number a of reference points, and selecting reference point rjReference point rjThe abscissa of (a) is the delay position t0+djThe ordinate is the judgment threshold RjJudgment threshold value RjAt a delay position t0+djRK function value and delay position t of typical rock burst signal of (site)0+djAt a delay position t between the values of the RK functions of the typical noise signal0+djRK function value and delay position t of typical rock burst signal of (site)0+djThe RK function values of the representative noise signals at are not equal; where a is in {1, 2}, j is the sequence number of the reference point, t0To trigger the moment, djIs the delay length.
In the present invention, the reference point rjMay be 1 or 2, i.e. reference points rjMay be a reference point r1Or may be a reference point r1And a reference point r2(ii) a Judging threshold value RjCorresponding to 1 or 2, that is, the judgment threshold R can be set1May be a judgment threshold value R1And a judgment threshold R2。
In the embodiment, the microseismic monitoring system has a trigger algorithm with a trigger time t0At 2000, the STA short time window length is 20, the LTA long time window length is 200, the CF function is based on equation (3), and the R function is based on equation (2). As shown in fig. 2, according to the difference situation of the typical rock burst signal and the typical noise signal, the number a of the reference points is determined to be 2, and the reference point r1Has the coordinate of (2040, 2.5), reference point r2Has the coordinate of (2130, 1), i.e., the delay position t is determined0+d 12040, a threshold value R is judged1Is 2.5, delay position t0+d 22130, the threshold value R is judged2Is 1.
And 4, identifying the RK function value of the microseismic data to be identified, comprising the following steps of:
and 4.1, reading the generated microseismic data to be identified in real time.
Step 4.2, calculating the RK function value of the microseismic data to be identified in real time, wherein the RK function value of the microseismic data to be identified is larger than a trigger threshold R for the first timeqThe corresponding time is the trigger time t0Trigger threshold RqTo establish the values, the parameters of the microseismic monitoring system with its own triggering algorithm can be used.
Step 4.3, calculating the delay position t0+djThe RK function value of the microseismic data to be identified,
at a delay position t0+djIn case the RK function value of a typical rock burst signal is greater than the RK function value of a typical noise signal:
if the position t is delayed0+djThe RK function value of the microseismic data to be identified is larger than a judgment threshold value RjIf so, identifying the microseismic data to be identified read this time as a to-be-identified rockburst signal; otherwise, the microseismic data to be identified read this time is a noise signal;
at a delay position t0+djIn the case where the RK function value of the typical rock burst signal is less than the RK function value of the typical noise signal:
if the position t is delayed0+djThe RK function value of the microseismic data to be identified is smaller than a judgment threshold value RjIf so, identifying the microseismic data to be identified read this time as a to-be-identified rockburst signal; otherwise, the microseismic data to be identified read this time is a noise signal;
if all delay positions t0+djAnd if the identification results are to-be-identified rockburst signals, the read microseismic data to be identified is the rockburst signal.
In the present invention, the reference point rjMay be 1 or 2, corresponding to delay position t0+djMay be the delay position t0+d1Or may be the delay position t0+d1And a delay position t0+d2. Under the condition that two reference points exist, the identification of delay positions corresponding to the two reference points is both to-be-identified rockburst signals, and the to-be-identified microseismic data read at this time can be determined to be the rockburst signals.
In this example, the judgment condition is RK (2040)>2.5 and RK (2130)<1, trigger threshold RqIs 6.
The judgment threshold value R as described abovejThe dynamic adjustment is realized by the following steps:
step 5.1, establishing a rock burst signal threshold value dynamic adjustment database, storing rock burst signals in the rock burst signal threshold value dynamic adjustment database according to the sequence of the occurrence time, and recording each rock burst signal in the rock burst signal threshold value dynamic adjustment database as X1，X2，…，XpWherein p is the maximum number of rock burst signals in the dynamic adjustment database of rock burst signal threshold, and the rock burst signal with the latest occurrence time is Xp。
In general, dynamically adjusting the maximum number p of rock burst signals in a database by using a rock burst signal threshold value to 400;
if the on-site geological conditions are variable, the maximum number p of rock burst signals in the database can be dynamically adjusted by reducing rock burst signal threshold values in projects such as deep tunnel excavation or large-scale factory building excavation, so that the dynamic adjustment of the threshold values is accelerated;
if the change of the field geological conditions is small, projects such as fixed point long-term monitoring and the like can increase the maximum number p of rock burst signals in the rock burst signal threshold dynamic adjustment database, so that the threshold dynamic adjustment is slowed down.
In this example, the maximum number p of rockburst signals in the dynamic adjustment database for the rockburst signal threshold is 400.
Step 5.2, dynamically adjusting the rock burst signal threshold value to each rock burst signal in the database at the delay position t0+djThe array formed by the arrangement of the RK function values is marked as RKj- (k), wherein k is the serial number of the rock burst signal and j is the serial number of the reference point;
at a delay position t0+djIn case the RK function value of a typical rock burst signal is greater than the RK function value of a typical noise signal: dynamically adjusting each rock burst signal in delay position t in database by rock burst signal threshold value0+djThe RK function values are arranged from large to small;
at a delay position t0+djIn the case where the RK function value of the typical rock burst signal is less than the RK function value of the typical noise signal: dynamically adjusting each rock burst signal in delay position t in database by rock burst signal threshold value0+djThe RK function values of are arranged from small to large.
In the example, the rock burst signal threshold value dynamically adjusts each rock burst signal in the database at the delay position t0+d1The function values RK (2040) of the RK are arranged from large to small (namely the array RK1- (k) arranged from large to small), dynamically adjusting the threshold value of the rock burst signal at the delay position t of each rock burst signal in the database0+d2The RK function values RK (2130) are arranged from small to large (namely the array RK2- (k) arranged from small to large).
Step 5.3, determining the selection rate BjSelecting the minimum value satisfying the following formula as the selection rate Bj
Wherein, BzAs a total selection rate, BzThe range of (A) is 50% to 100%.
In this example, the total selection rate BzTaking 95 percent, the selection rate B can be obtained1And B2Both are 97.5%.
Step 5.4, obtain kjValue, kj＝int(p×Bj) Int is the rounding operation;
if k isjIn < p, then judging threshold value RjIs RKj-(kj) And RKj-(kjAn average value of + 1);
if k isjIf p, the threshold value R is determinedjIs RKj-(kj)。
In this example, k1Value sum k2The values are all 390.
Step 5.5, acquiring new microseismic data and the number a of rockburst events in the new microseismic data1And the number of noise events b1Are all of the parameters which are known, and are,
according to the judgment threshold value RjAnd (4) automatically identifying and processing the new microseismic data to obtain a rock burst signal and a noise signal and obtain the number a of corresponding rock burst events2And the number of noise events b2(ii) a Arranging newly obtained rock burst signals according to occurrence time and recording as Y1，Y2，…，YqQ is the maximum number of newly obtained rock burst signals, q is less than or equal to p, and the rock burst signal with the latest occurrence time is Yq。
It is generally considered that if the number of rock burst signals/noise signals occurring within a set time is greater than or equal to 4, a rock burst event/noise signal can be determined, and all signals that cannot be an event will be filtered. The set time is generally defined to be 0.5 s.
Step 5.6, calculating the automatic identification accuracy rate E of the new microseismic data rockburst event according to the formula (5)
If the automatic identification accuracy rate E of the new microseismic data rockburst event is larger than the preset automatic identification accuracy rate threshold value R of the rockburst eventbIf not, performing step 5.8; wherein, the rock burst event is automatically identified with the accuracy threshold value RbThe value range of the compound is 50 to 100 percent。
In this example, the rock burst event is automatically identified by the accuracy threshold RbAnd (4) taking 80 percent.
Step 5.7, calculating the automatic identification accuracy G of the noise event of the new microseismic data according to the formula (6)
If the automatic identification accuracy G of the noise event of the new microseismic data is larger than the identification accuracy threshold R of the noise eventgThen confirm the judgment threshold RjOtherwise, performing step 5.8; wherein the noise event identification accuracy threshold RgThe value range of (A) is 0-50%.
In this example, the noise event identification accuracy threshold RgTaking out 30 percent.
Step 5.8, using newly obtained rock burst signal YsDynamic adjustment of rockburst signal X in database by replacing rockburst signal thresholdsAdding 1 to the value of s, wherein the initial value of s is 1, s belongs to {1, 2 …, (q +1) }, judging whether s is greater than q, if s is greater than q, indicating that the dynamic adjustment of the threshold value fails, setting s as the initial value of 1, and returning to the step 5.1 after the set time; otherwise, the step 5.2 is returned.
In this example, after the threshold is dynamically adjusted, the threshold R is determined1To 1.306, a threshold R is judged2Is 1.367.
In this example, 5109 microseismic data with a time period from 2016 (6 months 2) to 2016 (6 months 12) are selected, wherein 1602 rock burst signals are included, and the others are noise signals and comprise 440 low-amplitude electrical signals, 419 high-amplitude electrical signals, 560 short-duration electrical signals, 440 long-duration electrical signals, 822 locomotive impact rail signals, 487 secondary blasting signals and 339 jumbolter signals, and the RK (2040) value and the RK (2130) value of all the signals in each type are respectively made, as shown in fig. 3. As can be seen from FIG. 3, the density distribution of RK (2040) values of all rock burst signals, all low-amplitude electrical signals, all high-amplitude electrical signals, all short-duration electrical signals and all long-duration electrical signals are obviously different and respectively concentrated at 2-5, 1-2, 1-3 and 1-3; the RK (2130) value density distribution of all rock burst signals, all long continuous electric signals, all locomotive impact rail signals, all secondary burst signals and all jumbolter signals is obviously different and is respectively concentrated in 0-1, 0-3, 1-3, 0-2 and 1-3. And the RK (2040) value and the RK (2130) value are judged according to judgment conditions RK (2040) >1.306 and RK (2130) <1.367, so that most of rock burst signals can be automatically identified from the microseismic data, and most of noise signals are filtered.
As can be analyzed from fig. 3, due to the initial decision threshold R1And an initial judgment threshold R2The dynamic adjustment of the decision threshold is highly advantageous in determining the decision threshold because it depends on the typical situation of the signal and cannot evaluate the recognition effect. Compared with a fixed judgment threshold, the dynamic judgment threshold has a lower omission factor which can be evaluated and controlled, so that higher identification efficiency can be ensured.
Automatically identifying the measured data with the time period of 2016, 6 months, 13 days to 19 days through the corresponding judgment relation RK (2040) >1.306 and RK (2040) value RK (2130) < 1.367. The measured data is determined to contain 212 rock burst events and 5191 noise events, and the identification algorithm automatically identifies 202 rock burst events and 4222 noise events. The automatic identification accuracy rate E of the rock burst event reaches 95.28%, the automatic identification accuracy rate G of the noise event reaches 81.33%, and the workload of 78.32% of manual identification is reduced.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.
Claims (3)
1. A multi-time window simplified form identification method for dynamically adjusting rock burst signal threshold is characterized by comprising the following steps:
step 1, selecting a typical rock burst signal and a typical noise signal;
step 2, obtaining a RK function of a typical rock burst signal and a RK function of a typical noise signal;
step 3, taking time as a horizontal coordinate and a RK function value as a vertical coordinate,
determining the number a of reference points, and selecting reference point rjReference point rjThe abscissa of (a) is the delay position t0+djThe ordinate is the judgment threshold RjJudgment threshold value RjAt a delay position t0+djRK function value and delay position t of typical rock burst signal of (site)0+djAt a delay position t between the values of the RK functions of the typical noise signal0+djRK function value and delay position t of typical rock burst signal of (site)0+djThe RK function values of the representative noise signals at are not equal; where a is in {1, 2}, j is the sequence number of the reference point, t0To trigger the moment, djIs the delay length;
and 4, identifying the RK function value of the microseismic data to be identified, comprising the following steps of:
step 4.1, reading the generated microseismic data to be identified in real time;
step 4.2, calculating the RK function value of the microseismic data to be identified in real time, wherein the RK function value of the microseismic data to be identified is larger than a trigger threshold R for the first timeqThe corresponding time is the trigger time t0Trigger threshold RqIs a preset value;
step 4.3, calculating the delay position t0+djThe RK function value of the microseismic data to be identified,
at a delay position t0+djIn case the RK function value of a typical rock burst signal is greater than the RK function value of a typical noise signal:
if the position t is delayed0+djThe RK function value of the microseismic data to be identified is larger than a judgment threshold value RjIf so, identifying the microseismic data to be identified read this time as a to-be-identified rockburst signal; otherwise, the microseismic data to be identified read this time is a noise signal;
at a delay position t0+djTypical rock ofIn case the RK function value of the burst signal is smaller than the RK function value of a typical noise signal:
if the position t is delayed0+djThe RK function value of the microseismic data to be identified is smaller than a judgment threshold value RjIf so, identifying the microseismic data to be identified read this time as a to-be-identified rockburst signal; otherwise, the microseismic data to be identified read this time is a noise signal;
if all delay positions t0+djAnd if the identification results are to-be-identified rockburst signals, the read microseismic data to be identified is the rockburst signal.
2. The method for identifying the simplified form of the multiple time windows of the dynamic adjustment of the rock burst signal threshold value according to claim 1, wherein the judgment threshold value R isjThe dynamic adjustment is realized by the following steps:
step 5.1, establishing a rock burst signal threshold value dynamic adjustment database, storing rock burst signals in the rock burst signal threshold value dynamic adjustment database according to the sequence of the occurrence time, and recording each rock burst signal in the rock burst signal threshold value dynamic adjustment database as X1，X2，…，XpWherein p is the maximum number of rock burst signals in the dynamic adjustment database of rock burst signal threshold, and the rock burst signal with the latest occurrence time is Xp；
Step 5.2, dynamically adjusting the rock burst signal threshold value to each rock burst signal in the database at the delay position t0+djThe array formed by the arrangement of the RK function values is marked as RKj- (k), wherein k is the serial number of the rock burst signal and j is the serial number of the reference point;
at a delay position t0+djIn case the RK function value of a typical rock burst signal is greater than the RK function value of a typical noise signal: dynamically adjusting each rock burst signal in delay position t in database by rock burst signal threshold value0+djThe RK function values are arranged from large to small;
at a delay position t0+djIn the case where the RK function value of the typical rock burst signal is less than the RK function value of the typical noise signal: dynamic adjustment of rock burst signal thresholdDelay position t of each rock burst signal in whole database0+djThe RK function values are arranged from small to large;
step 5.3, determining the selection rate BjSelecting the minimum value satisfying the following formula as the selection rate Bj
Wherein, BzAs a total selection rate, BzThe range value of (A) is 50% -100%;
step 5.4, obtain kjValue, kj＝int(p×Bj) Int is the rounding operation;
if k isjIn < p, then judging threshold value RjIs RKj-(kj) And RKj-(kjAn average value of + 1);
if k isjIf p, the threshold value R is determinedjIs RKj-(kj)；
Step 5.5, acquiring new microseismic data and the number a of rockburst events in the new microseismic data1And the number of noise events b1Are all of the parameters which are known, and are,
according to the judgment threshold value RjAnd (4) automatically identifying and processing the new microseismic data to obtain a rock burst signal and a noise signal and obtain the number a of corresponding rock burst events2And the number of noise events b2(ii) a Arranging newly obtained rock burst signals according to occurrence time and recording as Y1，Y2，…，YqQ is the maximum number of newly obtained rock burst signals, q is less than or equal to p, and the rock burst signal with the latest occurrence time is Yq；
Step 5.6, calculating the automatic identification accuracy rate E of the new microseismic data rockburst event according to the formula (5)
If the automatic identification accuracy rate E of the new microseismic data rockburst event is larger than the preset automatic identification accuracy rate threshold value R of the rockburst eventbIf not, performing step 5.8; wherein, the rock burst event is automatically identified with the accuracy threshold value RbThe value range of (A) is 50% -100%;
step 5.7, calculating the automatic identification accuracy G of the noise event of the new microseismic data according to the formula (6)
If the automatic identification accuracy G of the noise event of the new microseismic data is larger than the identification accuracy threshold R of the noise eventgThen confirm the judgment threshold RjOtherwise, performing step 5.8; wherein the noise event identification accuracy threshold RgThe value range of (A) is 0-50%;
step 5.8, using newly obtained rock burst signal YsDynamic adjustment of rockburst signal X in database by replacing rockburst signal thresholdsAdding 1 to the value of s, wherein the initial value of s is 1, s belongs to {1, 2 …, (q +1) }, judging whether s is greater than q, if s is greater than q, indicating that the dynamic adjustment of the threshold value fails, setting s as the initial value of 1, and returning to the step 5.1 after the set time; otherwise, the step 5.2 is returned.
3. The method for identifying a simplified form of multiple time windows with dynamically adjusted rock burst signal threshold value according to claim 1, wherein the RK function of typical rock burst signal and the RK function of typical noise signal are based on formula (1),
rk (t) r (t) formula (1)
The R function in equation (1) is based on equation (2),
wherein STA (t) is a short time window STA function, LTA (t) is a long time window LTA value, t is time, n is STA short time window length, m is LTA long time window length,
the CF function in equation (2) is based on equation (3),
CF(t)＝Y(t)2-Y (t-1). Y (t +1) formula (3)
Where Y (t) is a function of amplitude.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810098506.9A CN108319915B (en) | 2018-01-31 | 2018-01-31 | Multi-time-window simplified form identification method for dynamically adjusting rock burst signal threshold |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810098506.9A CN108319915B (en) | 2018-01-31 | 2018-01-31 | Multi-time-window simplified form identification method for dynamically adjusting rock burst signal threshold |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108319915A CN108319915A (en) | 2018-07-24 |
| CN108319915B true CN108319915B (en) | 2020-06-19 |
Family
ID=62891370
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810098506.9A Active CN108319915B (en) | 2018-01-31 | 2018-01-31 | Multi-time-window simplified form identification method for dynamically adjusting rock burst signal threshold |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108319915B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110488352A (en) * | 2019-08-20 | 2019-11-22 | 中国科学院武汉岩土力学研究所 | A kind of determination method of the microseism parameter threshold for rock burst grade early warning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103064111A (en) * | 2012-12-12 | 2013-04-24 | 中国石油天然气集团公司 | Micro seismic event recognition method based on morphological filtering |
| WO2014197464A1 (en) * | 2013-06-03 | 2014-12-11 | Westerngeco Llc | Processing of seismic data |
| CN105527650A (en) * | 2016-02-17 | 2016-04-27 | 中国科学院武汉岩土力学研究所 | Automatic identification algorithm for microseismic signal and p wave first arrival at engineering scale |
| CN105891874A (en) * | 2016-06-30 | 2016-08-24 | 马克 | Mining coal and rock mass water inrush slight shock monitoring method |
-
2018
- 2018-01-31 CN CN201810098506.9A patent/CN108319915B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103064111A (en) * | 2012-12-12 | 2013-04-24 | 中国石油天然气集团公司 | Micro seismic event recognition method based on morphological filtering |
| WO2014197464A1 (en) * | 2013-06-03 | 2014-12-11 | Westerngeco Llc | Processing of seismic data |
| CN105527650A (en) * | 2016-02-17 | 2016-04-27 | 中国科学院武汉岩土力学研究所 | Automatic identification algorithm for microseismic signal and p wave first arrival at engineering scale |
| CN105891874A (en) * | 2016-06-30 | 2016-08-24 | 马克 | Mining coal and rock mass water inrush slight shock monitoring method |
Non-Patent Citations (2)
| Title |
|---|
| Analysis of source characteristics of experimental gas burst and fragmentation explosions generated by rapid decompression of volcanic rocks;Arciniega-Ceballos A, et al.;《JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH》;20150731;第120卷(第7期);5104-5116 * |
| 一种改进的STA/LTA震相自动识别方法;高淑芳等;《世界地震工程》;20080630;第24卷(第2期);37-41 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108319915A (en) | 2018-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110779574B (en) | Coal rock dynamic disaster multi-system multi-parameter integrated comprehensive early warning method and system | |
| CN106353792A (en) | Method suitable for positioning hydraulic fracturing micro-seismic source | |
| CN113050159A (en) | Coal rock hydraulic fracturing crack micro-seismic positioning and propagation mechanism monitoring method | |
| CN113673119B (en) | Coal mine rock burst danger dynamic and static coupling evaluation method based on Bayesian method | |
| CN107203003A (en) | A kind of mine water disaster micro seismic monitoring space-time kmeans cluster method | |
| CN106437843B (en) | coal mine bottom plate water guide channel identification method based on microseismic monitoring | |
| CN104318058A (en) | Mudslide early warning method based on rainfall monitoring | |
| CN108319915B (en) | Multi-time-window simplified form identification method for dynamically adjusting rock burst signal threshold | |
| CN115169948A (en) | Coal face overlying rock separation layer water inrush risk prediction method and safe mining method | |
| CN113962466A (en) | Dynamic prediction method for impact risk of rock burst mine fold area based on microseisms | |
| CN111563653A (en) | Early warning construction method for water-rich broken stratum of underground engineering | |
| CN110376643B (en) | Micro-seismic effect data processing method for jet grouting pile diameter detection | |
| CN112580165A (en) | Prediction method and system for open type TBM passing through unfavorable geological cutter head jamming machine | |
| CN109239775B (en) | Method for tracking and positioning stolen mining of mineral resources | |
| CN113189644B (en) | Microseismic source positioning method and system | |
| CN103575438A (en) | Coal mine advance bearing pressure distribution characteristic monitoring method | |
| Pang et al. | Analysis and prediction of hydraulic support load based on time series data modeling | |
| CN109681181B (en) | Prediction method and device | |
| CN112230275A (en) | Seismic waveform identification method and device and electronic equipment | |
| CN109578075B (en) | Microseismic monitoring and early warning method, device and system for rock burst danger | |
| CN113266421B (en) | Comprehensive early warning method for full-dangerous period time and space of rock burst | |
| CN113139681A (en) | Neural network rock burst prediction method based on time series data | |
| CN115640996A (en) | Evaluation method for water inrush disaster of tunnel in water-rich complex stratum | |
| CN114294062A (en) | Time-space dynamic comprehensive early warning method for rock burst | |
| CN108254781B (en) | A kind of dynamic adjustment of rock burst signal threshold value it is more when window complete form recognition methods |
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 |