CN114087021B - Rock burst multi-parameter dynamic trend early warning method - Google Patents
Rock burst multi-parameter dynamic trend early warning method Download PDFInfo
- Publication number
- CN114087021B CN114087021B CN202111258391.3A CN202111258391A CN114087021B CN 114087021 B CN114087021 B CN 114087021B CN 202111258391 A CN202111258391 A CN 202111258391A CN 114087021 B CN114087021 B CN 114087021B
- Authority
- CN
- China
- Prior art keywords
- early warning
- index
- rock burst
- trend
- time
- 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
- 239000011435 rock Substances 0.000 title claims abstract description 97
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012544 monitoring process Methods 0.000 claims abstract description 55
- 239000002243 precursor Substances 0.000 claims abstract description 33
- 230000002159 abnormal effect Effects 0.000 claims abstract description 26
- 238000005457 optimization Methods 0.000 claims abstract description 14
- 230000004927 fusion Effects 0.000 claims abstract description 13
- 238000007689 inspection Methods 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims abstract description 9
- 230000008859 change Effects 0.000 claims description 43
- 230000000694 effects Effects 0.000 claims description 24
- 230000035939 shock Effects 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000003247 decreasing effect Effects 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 5
- 238000010998 test method Methods 0.000 claims description 4
- 238000012512 characterization method Methods 0.000 claims description 3
- 230000000977 initiatory effect Effects 0.000 claims description 3
- 230000001788 irregular Effects 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 2
- 239000003245 coal Substances 0.000 abstract description 9
- 238000011081 inoculation Methods 0.000 abstract description 3
- 230000002265 prevention Effects 0.000 abstract description 3
- 238000009412 basement excavation Methods 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 5
- 238000005065 mining Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 125000004435 hydrogen atom Chemical class [H]* 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 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
- 238000013139 quantization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21F—SAFETY DEVICES, TRANSPORT, FILLING-UP, RESCUE, VENTILATION, OR DRAINING IN OR OF MINES OR TUNNELS
- E21F17/00—Methods or devices for use in mines or tunnels, not covered elsewhere
- E21F17/18—Special adaptations of signalling or alarm devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mining & Mineral Resources (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Geochemistry & Mineralogy (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a rock burst multi-parameter dynamic trend early warning method, and belongs to the technical field of underground excavation engineering and coal rock dynamic disaster early warning. The method comprises the following steps: describing the variation trend of a 'time-space-strong' multi-element precursor early warning index in the process of the inoculation evolution of the rock burst by utilizing a Mann-Kendall trend inspection method according to real-time monitoring data of a field microseismic monitoring system, and early warning by taking whether each index variation trend accords with the representation rule of the rock burst precursor; evaluating the early warning efficiency of each index; performing index optimization based on an early warning efficiency maximization principle; performing multi-index fusion by taking the early warning efficiency corresponding to the optimized index as weight to obtain an impact risk comprehensive abnormal index; comparing the impact risk comprehensive abnormal index with a corresponding quantitative grading standard to determine the impact risk grade; and (4) periodically carrying out index optimization and weight updating under the drive of field actual monitoring data. By adopting the method and the device, an efficient and accurate decision basis can be provided for the prevention and treatment of the underground rock burst.
Description
Technical Field
The invention relates to the technical field of underground excavation engineering and coal rock dynamic disaster early warning, in particular to a rock burst multi-parameter dynamic trend early warning method.
Background
Rock burst is one of the main dynamic disasters affecting underground coal mine production, and serious consequences such as casualties, major property loss and the like are often caused. In recent years, with the gradual depletion of shallow mineral resources, coal mining is continuously expanded to the deep part of the earth, the occurrence conditions of coal seam structures of stopes and surrounding rocks around roadways are gradually complicated, the dynamic response characteristics in the coal rock bodies are stronger, the occurrence frequency of rock burst accidents is rapidly increased, and accurate and efficient monitoring and early warning means for the disasters are the key for guaranteeing underground safe production.
At present, the on-line monitoring means commonly used in the underground comprises microseismic, earthquake sound, electromagnetic radiation, ground stress and the like, wherein a microseismic monitoring system can monitor the breakage of coal rock mass in a large-range area of a mine, and is widely applied to monitoring and early warning of rock burst, and related personnel can achieve the purpose of early warning by analyzing the change trend of indexes such as energy, frequency and the like of microseismic events. However, the judgment of the change trend of the early warning indexes of the impact risk precursor depends on manual experience, the efficiency is low, the reliability is low, and meanwhile, due to the fact that dimensions of the early warning indexes in the process of the inoculation evolution of the impact rock pressure are different, the phenomenon that results conflict with each other may occur, and the misjudgment of relevant personnel on the actual dangerous state is caused. Therefore, it is necessary to provide an early warning method capable of automatically and efficiently judging the time-series variation trend of the early warning index of the impact risk precursor and providing comprehensive, single and quantitative impact risk levels.
Disclosure of Invention
The embodiment of the invention provides a rock burst multi-parameter dynamic trend early warning method, which can carry out efficient and accurate monitoring and early warning on rock burst. The technical scheme is as follows:
the embodiment of the invention provides a rock burst multi-parameter dynamic trend early warning method, which comprises the following steps:
s101, according to real-time monitoring data of a field microseismic monitoring system, describing the variation trend of a 'time-space-strong' multi-element precursor early warning index in the process of the initiation and evolution of the rock burst by using a Mann-Kendall trend inspection method, and carrying out early warning by taking whether the variation trend of each index meets the representation rule of the rock burst precursor or not as an early warning criterion;
s102, evaluating the early warning efficiency of each index by using a confusion matrix;
s103, performing index optimization based on an early warning effectiveness maximization principle;
s104, performing multi-index fusion by taking the early warning efficiency corresponding to the optimized index as weight to obtain an impact risk comprehensive abnormal index;
s105, comparing the comprehensive impact risk abnormal index with a corresponding quantitative grading standard to determine the impact risk grade;
and S106, periodically carrying out index optimization and weight updating under the drive of field actual monitoring data.
Further, the step of describing a time-space-strong multi-element precursor early warning index change trend in the process of the development of the rock burst by using a Mann-Kendall trend inspection method according to real-time monitoring data of the on-site microseismic monitoring system, and the early warning by taking whether each index change trend meets the rock burst precursor characterization rule as an early warning criterion comprises the following steps:
acquiring real-time monitoring data of a field microseismic monitoring system, and preprocessing original monitoring data by using a specific time window and a slip step length to obtain a time sequence of a time-space-strong multi-precursor early warning index of rock burst danger; wherein, the time-space-strength multi-element precursor early warning indexes of the rock burst danger comprise:
reflecting the daily total frequency, frequency deviation value, average total frequency, lack of earthquake, A (b) value, microseismic activity scale, algorithm complexity, P (b) value and time information entropy of a time dimension;
reflecting the microseismic activity degree and the seismic focus concentration degree of the space dimension;
reflecting daily maximum energy, daily total energy, daily average energy, energy deviation value, average total energy, total fault area and b value of the intensity dimension;
the daily maximum energy, daily total frequency, daily average energy, energy deviation value, frequency deviation value, average total energy, microseismic activity degree, lack of shock, A (b) value, total fault area, average total frequency, microseismic activity scale and algorithm complexity belong to forward early warning indexes, namely the higher the value is, the greater the impact risk is; the seismic source concentration degree, the b value, the P (b) value and the time information entropy belong to negative early warning indexes, namely the lower the value is, the greater the impact risk is;
and judging the real-time change trend of each early warning index by using a Mann-Kendall trend inspection method, and early warning by taking whether the change trend of each index meets the rock burst precursor representation rule as an early warning criterion.
Further, the preprocessing the original monitoring data with a specific time window and a specific sliding step length means converting the original monitoring data from an irregular time sequence to a regular time sequence, and includes:
defining a sliding time window of length Deltat, monitoring of the acquisitionThe time series of data is divided into n data sets of length Δ T and each corresponds to a time instant at the end of the time window, i.e. where T is i Time of day dataset X i [x 1 ,x 2 ,x 3 ,...,x k ]，k≤t,0<i is less than or equal to n, where Δ T is the length of the sliding time window, T i The time corresponding to the end of the ith time window;
calculating X i Early warning index value y corresponding to all samples in i The early warning index value y i Arranging in sequence to obtain a converted regular time sequence marked as Y [ Y ] 1 ,y 2 ,y 3 ,...,y n ]。
Further, the step of judging the real-time change trend of each early warning index by using a Mann-Kendall trend test method comprises the following steps:
the early warning index is regulated to be in a time sequence Y [ Y ] 1 ,y 2 ,y 3 ,...,y n ]Dividing the data into m data sets with time window length delta a and respectively corresponding to time A at the end of the time window i I.e. A i The dataset of the time of day is Y i [y 1 ,y 2 ,y 3 ,...,y q ]，10≤q≤a,0<i is less than or equal to m, and Y is calculated i Test statistic S:
wherein sgn (·) represents a sign function,
q represents Y i [y 1 ,y 2 ,y 3 ,...,y q ]Length of data set, y p Denotes the pth data, p 1,2,3.. q-1, y j Represents the jth data, j ═ p +1,2,3.. q;
according to the obtained Y i Of the test statistic S, determining Y i Test standard amount of (3) Z:
when Z is>At 0, T i The early warning indexes at all times have a growing trend; when Z is<At 0, T i The early warning index at the moment has a decreasing trend; when Z is 0, T i The early warning indexes at the moment have no obvious variation trend.
Further, in this embodiment, the performing an early warning with the criterion of whether the change trend of each index meets the representation rule of the rock burst precursor includes:
and comparing the real-time change trend of each index with the rock burst precursor representation rule, performing early warning if the positive early warning index has an increasing trend, performing early warning if the negative early warning index has a decreasing trend, and not performing early warning if other change trends.
Further, the early warning effectiveness is expressed as:
wherein, EFF represents the early warning efficiency; recall represents the Recall rate of a call,
indicating that the pre-warning is a shock hazard and that a rock burst event has actually occurred,
indicating that the early warning is no impact danger but a rock burst event actually occurs; precision represents the rate of Precision at which,
wherein
Indicates that the early warning is that there is a shock hazard and that a shock actually occursThe occurrence of a ground pressure event,
indicating that the early warning is a shock hazard but that no rock burst event actually occurs.
Further, the preferably performing the index based on the early warning effectiveness maximization principle includes:
and arranging the early warning effect values of all the early warning indexes from large to small, and screening indexes n% of which are ranked before for the next data fusion, wherein n is a positive integer, and the number of the indexes is rounded upwards.
Further, the obtaining of the impact risk comprehensive anomaly index by performing multi-index fusion by using the early warning effectiveness corresponding to the preferred index as a weight includes:
and inputting the early warning efficiency corresponding to the optimized indexes into the comprehensive early warning model of rock burst as weight, and performing multi-index fusion on the comprehensive early warning model of rock burst by using a comprehensive abnormal index method to obtain a comprehensive abnormal index of the impact risk:
wherein Q represents a shock hazard comprehensive abnormality index; e represents a natural base number; n represents the total number of preferred indicators; EFF k The early warning efficiency corresponding to the kth index; w k(+/-) Representing the abnormal membership of the kth positive/negative early warning index,
for the forward early warning index, the value of the abnormal membership degree is as follows:
for negative early warning indexes, the value of the abnormal membership degree is as follows:
further, the impact risk classes include: a no-impact-risk state, a weak-impact-risk state, a medium-impact-risk state, and a strong-impact-risk state.
Further, the performing of index optimization and weight update under the drive of the field actual monitoring data in the scheduled period includes:
and (4) performing index optimization periodically under the drive of field actual monitoring data, and inputting the early warning efficiency corresponding to the optimized index into the comprehensive early warning model of rock burst as weight so as to realize self-feedback updating of the comprehensive early warning model of rock burst.
The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:
1) the real-time change trend of the early warning index can be automatically judged, and rock burst early warning is carried out by utilizing the change trend of the early warning index;
2) the comprehensive impact danger abnormal index is used for quantitatively describing the impact ground pressure danger, so that the phenomena of high false alarm/missing alarm rate caused by a single early warning index and early warning grade conflict of different early warning indexes are avoided;
3) the method is driven by on-site actual monitoring data to perform periodic self-feedback updating, has strong expandability and adaptability, adapts to complex and variable working conditions in the well, is favorable for performing efficient and accurate monitoring and early warning on the rock burst, and provides efficient and accurate decision basis for prevention and treatment of the rock burst in the well.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a rock burst multi-parameter dynamic trend early warning method provided in an embodiment of the present invention;
fig. 2 is a detailed flowchart schematic diagram of a rock burst multi-parameter dynamic trend early warning method provided by an embodiment of the invention;
FIG. 3 is a schematic diagram illustrating a preprocessing process performed on raw monitoring data according to the present embodiment;
FIG. 4 shows the total daily frequency F in this embodiment sum A time sequence change curve of the early warning index;
FIG. 5 shows the minor deviation D of the frequency in this embodiment F A time sequence change curve of the early warning index;
FIG. 6 is a graph showing the average total frequency in the present embodiment
A time sequence change curve of the early warning index;
FIG. 7 shows the lack of vibration M in this embodiment m A time sequence change curve of the early warning index;
FIG. 8 is a time-series variation curve of the value of A (b) in the present embodiment;
FIG. 9 is a time series curve of the microseismic activity scale F pre-alarm indicator of the present embodiment;
FIG. 10 is a time sequence variation curve of the algorithm complexity AC early warning indicator in the present embodiment;
FIG. 11 is a time-series variation curve of the pre-warning indicator of the value P (b) in this embodiment;
FIG. 12 is the time information entropy Q in the present embodiment t A time sequence change curve of the early warning index;
FIG. 13 shows the microseismic activity S of the present embodiment D A time sequence change curve of the early warning index;
FIG. 14 is a time series curve of the seismic focus level λ warning indicator in the present embodiment;
FIG. 15 shows the daily maximum energy E in the present embodiment max A time sequence change curve of the early warning index;
FIG. 16 shows the total daily energy E in this example sum A time sequence change curve of the early warning index;
FIG. 17 shows the daily average energy E in this example avg A time sequence change curve of the early warning index;
FIG. 18 shows the energy deviation D in this embodiment E A time sequence change curve of the early warning index;
FIG. 19 shows the exampleMean total energy in the examples
A time sequence change curve of the early warning index;
FIG. 20 is a time-series curve of the warning indicator for the total area A (t) of the interruption layer in this embodiment;
FIG. 21 is a time-series variation curve of the b-value warning indicator in the present embodiment;
fig. 22 is a time-series change curve of the impact risk comprehensive abnormality index Q in the present embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
As shown in fig. 1, an embodiment of the present invention provides a rock burst multi-parameter dynamic trend early warning method, including:
s101, according to real-time monitoring data of an on-site microseismic monitoring system, describing a time-space-strong multi-element precursor early warning index change trend in an impact ground pressure inoculation evolution process by using a Mann-Kendall (Mann-Kendall) trend inspection method, and performing early warning by taking whether each index change trend meets an impact ground pressure precursor characterization rule as an early warning criterion; the method specifically comprises the following steps:
a1, acquiring real-time monitoring data of a field microseismic monitoring system, preprocessing original monitoring data by a specific time window and a slip step length, and obtaining a time sequence of a time-space-strong multi-precursor early warning index of rock burst danger; wherein, the time-space-strong multivariate precursor early warning indexes of the rock burst danger comprise but are not limited to:
total daily frequency F reflecting the time dimension sum Frequency deviation value D F Average total frequency
Lack of vibration M m A (b) value, microseismic activity scale F, algorithm complexity AC, P (b) value, and time information entropy Q t (ii) a Wherein the A (b) value and the P (b) value are one of basic laws of seismologyThe derivation index for the test constant b in the "G-R" relationship (Gutenberg-Rickett equation);
wherein, the total daily frequency F sum The value of (d) is calculated by:
wherein n represents the total frequency of microseismic events within the first 24h (hours);
deviation value of frequency D F The value of (d) is calculated by:
wherein, F t Indicating the frequency of microseismic events within the first 24h,
represents the average daily frequency of microseismic events within the first 30d (days);
average total frequency
The value of (d) is calculated by:
wherein T represents the time window length, F (t) Representing the microseismic frequency at the time t;
lack of vibration M m The value of (d) is calculated by:
wherein m is an energy levelTotal number of grading; lgE i Is the ith gear energy level; n is a radical of hydrogen i The actual microseismic number of the ith gear level; the value of A (b) is calculated by the following formula:
wherein N is the total number of microseisms, M i Is the microseismic event energy level;
the value of the microseismic activity scale F is calculated by:
F 0 ＝10 6.11+1.09M
wherein T represents the time window length, M i Is the microseismic event energy level;
the value of the algorithm complexity AC is calculated by the following formula:
wherein n represents the number of changes in microseismic energy level within a time window, M max Representing the maximum energy level of the microseisms, M min Representing the microseismic minimum energy level;
the value of P (b) is calculated by the following formula:
wherein N is the total number of microseisms, M i Is the microseismic event energy level;
time information entropy Q t The value of (d) is calculated by:
wherein n is the frequency of microseismic events within the time window, t i The ith mine earthquake occurrence time;
microseismic activity S reflecting spatial dimensions D And a seismic focus concentration degree λ; wherein,
microseismic activity S D The value of (d) is calculated by:
wherein N is the total number of microseisms, E i Is the microseismic event energy, M is the microseismic maximum energy level;
the value of the seismic focus concentration λ is calculated by the following formula:
wherein λ is 1 、λ 2 、λ 3 Forming a characteristic root of a covariance matrix for microseismic event coordinates (x, y, z) within a time window;
daily maximum energy E reflecting the intensity dimension max Total daily energy E sum Daily average energy E avg Energy deviation value D E Average total energy
Total fault area A (t) and b values; wherein,
maximum daily energy E max The value of (d) is calculated by:
E max ＝max(E 1 ,E 2 ,...,E n )
wherein E is i The ith microseismic event within the first 24 h;
total daily energy E sum The value of (d) is calculated by:
wherein n represents the total frequency of microseismic events within the first 24h, E i Representing the individual microseismic event energies;
average daily energy E avg The value of (d) is calculated by:
wherein n represents the total frequency of microseismic events within the first 24h, E i Representing the individual microseismic event energies;
deviation of energy value D E The value of (d) is calculated by:
wherein E is t Representing the total energy of the microseismic events within the first 24h,
represents the average daily energy of the microseismic events in the first 30 d;
mean total energy
The value of (d) is calculated by:
wherein T represents the time window length, E (t) Representing the microseismic energy at time t;
the value of the total fault area A (t) is calculated by the following formula:
wherein, k0 is a time windowThe lower limit of the energy level of the internal microseisms, k is the energy level of each microseism, and N (k) is the number of microseismic events with the energy level of k in the time window;
the value of b is calculated by the following formula:
wherein m is the total number of the energy level grading; lgE i Is the ith gear energy level; n is a radical of i Is the actual microseismic number of the ith gear level.
The early warning indexes can construct a precursor early warning index library, wherein the daily maximum energy, daily total frequency, daily average energy, energy deviation value, frequency deviation value, average total energy, microseismic activity degree, lack of shock, A (b) value, total fault area, average total frequency, microseismic activity scale and algorithm complexity in the early warning indexes belong to forward early warning indexes, namely the higher the value is, the greater the impact risk is; the seismic source concentration degree, the b value, the P (b) value and the time information entropy belong to negative early warning indexes, namely, the lower the value is, the larger the impact risk is.
In this embodiment, the preprocessing the original monitoring data with a specific time window and a specific sliding step length means converting the original monitoring data from an irregular time sequence to a regular time sequence, and includes:
defining a sliding time window with the length of delta T, dividing the acquired monitoring data time sequence into n data sets with the length of delta T and respectively corresponding to the time at the end of the time window, namely T i The data set at time is denoted X i [x 1 ,x 2 ,x 3 ,...,x k ]，k≤t,0<i is less than or equal to n, where Δ T is the length of the sliding time window, T i The time corresponding to the end of the ith time window;
calculating X i Early warning index value y corresponding to all samples in i The early warning index value y i Arranging in sequence to obtain a converted regular time sequence marked as Y [ Y ] 1 ,y 2 ,y 3 ,...,y n ]As shown in fig. 3.
A2, judging the real-time change trend of each early warning index by using a Mann-Kendall trend test method, and carrying out early warning by taking whether the change trend of each index meets the rock burst precursor representation rule as an early warning criterion, wherein the method specifically comprises the following steps:
the early warning index is regulated to be in a time sequence Y [ Y ] 1 ,y 2 ,y 3 ,...,y n ]Dividing the data into m data sets with time window length delta a and respectively corresponding to time A at the end of the time window i I.e. A i The dataset of the time of day is Y i [y 1 ,y 2 ,y 3 ,...,y q ]，10≤q≤a,0<i is less than or equal to m, and Y is calculated i Test statistic S:
wherein sgn (·) represents a sign function,
q represents Y i [y 1 ,y 2 ,y 3 ,...,y q ]Length of data set, y p Denotes the pth data, p 1,2,3.. q-1, y j Represents the jth data, j ═ p +1,2,3.. q;
according to the obtained Y i Of the test statistic S, determining Y i Test standard amount of (3) Z:
when Z is>At 0, T i The early warning indexes at all times have a growing trend; when Z is<At 0, T i The early warning index at the moment has a decreasing trend; when Z is 0, T i The early warning indexes at the moment have no obvious variation trend. Repeating the calculation until obtaining the corresponding early warning index change trend at all timesPotential;
and comparing the real-time change trend of each index with the rock burst precursor representation rule, performing early warning if the positive early warning index has an increasing trend, performing early warning if the negative early warning index has a decreasing trend, and not performing early warning if other change trends.
S102, evaluating the early warning efficiency of each index by using a confusion matrix;
in this embodiment, the confusion matrix is specifically the following 2 × 2 matrix:
in this embodiment, the early warning performance of the early warning index obtained according to the confusion matrix is:
wherein, EFF represents the early warning efficiency; recall represents the Recall rate of the call,
wherein
Indicating that the pre-warning is a shock hazard and that a rock burst event has actually occurred,
indicating that the early warning is no impact danger but a rock burst event actually occurs; precision represents the rate of Precision at which,
wherein
Indicating that the pre-warning is a shock hazard and that a rock burst event has actually occurred,
indicating that the early warning is a shock hazard but that no rock burst event actually occurs. .
In the embodiment, the trend range of the EFF is 0-1, and the closer the value is to 1, the better the early warning effect is.
S103, performing index optimization based on an early warning effectiveness maximization principle;
in this embodiment, the early warning effect values of all the early warning indicators are arranged from large to small, and the indicators n% before ranking are screened for the next data fusion, where n is a positive integer and the number of the indicators is rounded up.
S104, performing multi-index fusion by taking the early warning efficiency corresponding to the optimized index as weight to obtain an impact risk comprehensive abnormal index;
in this embodiment, the early warning performance corresponding to the preferred index is input into the comprehensive early warning model of rock burst as a weight, and the comprehensive early warning model of rock burst performs multi-index fusion by using a comprehensive anomaly index method to obtain a comprehensive anomaly index of the impact risk:
wherein Q represents a shock hazard comprehensive abnormality index; e represents a natural base number equal to about 2.718; n represents the total number of preferred indicators; EFF k The early warning efficiency corresponding to the kth index; w k(+/-) The abnormal membership degree of the kth positive/negative early warning index is represented and is 0-1, and the specific calculation mode is as follows:
for the forward early warning index, the value of the abnormal membership degree is as follows:
for negative early warning indexes, the value of the abnormal membership degree is as follows:
and S105, comparing the impact risk comprehensive abnormal index with the corresponding quantitative grading standard to determine the impact risk level.
In this embodiment, the impact risk quantization classification standard specifically includes:
when Q is more than or equal to 0 and less than or equal to 0.25, the state is in a non-impact dangerous state;
when Q is more than 0.25 and less than or equal to 0.5, the state is a weak impact dangerous state;
when Q is more than 0.5 and less than or equal to 0.75, the state is a medium impact dangerous state;
when Q is more than 0.75 and less than or equal to 1, the state is a strong impact dangerous state.
And S106, periodically carrying out index optimization and weight updating under the drive of field actual monitoring data.
In this embodiment, the indexes are optimized periodically under the drive of the actual field monitoring data, and the early warning performance corresponding to the optimized indexes is input into the comprehensive early warning model of rock burst as a weight, so as to realize the self-feedback update of the comprehensive early warning model of rock burst.
In this embodiment, for example, the steps S101 to S104 may be repeated at intervals of 1 month or more (or under the conditions of replacement of working face, great change of geological conditions, occurrence of high-energy mine earthquake or rock burst event, and the like) on site by using all existing historical monitoring data, and the adaptive indexes are preferably selected from the precursor early warning index library again while the corresponding early warning performance is used as weight to be input into the rock burst comprehensive early warning model, so as to achieve the purpose of improving the adaptability by self-feedback updating of the rock burst comprehensive early warning model.
In summary, the early warning method for the multi-parameter dynamic trend of rock burst according to the embodiment of the invention has at least the following beneficial effects:
1) the real-time change trend of the early warning index can be automatically judged, and rock burst early warning is carried out by utilizing the change trend of the early warning index;
2) the comprehensive abnormal indexes of the impact dangers are used for quantitatively describing the impact ground pressure dangers, so that the phenomena of high false alarm/missing report rate caused by a single early warning index and early warning grade conflicts of different early warning indexes are avoided;
3) the method has the advantages that the periodic self-feedback updating is carried out under the driving of the field actual monitoring data, the expandability and the adaptability are strong, the complicated and changeable working conditions in the underground are adapted, the high-efficiency and accurate monitoring and early warning of the rock burst are facilitated, and the high-efficiency and accurate decision basis is provided for the prevention and the treatment of the rock burst in the underground.
For better understanding of the present invention, the rock burst multi-parameter dynamic trend early warning method provided by the present embodiment is further described with reference to specific application scenarios:
in this embodiment, taking a certain working face of a certain coal mine as an example, the microseismic raw monitoring data and related information during the mining period of the working face (2 month 1 in 2018 to 2 month 1 in 2019, wherein the mining period is stopped due to the face-on inspection in 2018 to 10 month 19 in 2018, and no monitoring data is collected: the method comprises the following steps of generating 15 rock burst events in the working face mining process, and early warning the 15 rock burst events, wherein the specific steps are as follows:
acquiring real-time monitoring data of a field microseismic monitoring system, preprocessing original monitoring data by taking 15 days as a time window and 1 day as a sliding step length, and calculating to obtain a time sequence of time-space-strong multi-element precursor early warning indexes of rock burst danger, wherein each early warning index comprises: the total daily frequency, the frequency deviation value, the average total frequency, the lack of earthquake, the A (b) value, the microseismic activity scale, the algorithm complexity, the P (b) value, the time information entropy, the microseismic activity degree, the seismic source concentration degree, the daily maximum energy, the daily total energy, the daily average energy, the energy deviation value, the average total energy, the total fault area and the b value are 18 indexes, and the time sequence change curve of the calculation result is shown in fig. 4-21. Wherein the forward early warning indicators include: daily maximum energy, daily total frequency, daily average energy, energy deviation value, frequency deviation value, average total energy, microseismic activity, lack of seism, A (b) value, total fault area, average total frequency, microseismic activity scale and algorithm complexity; negative early warning indexes include: the degree of focus, b value, P (b) value, time information entropy.
The real-time change trend of each early warning index is judged by using a Mann-Kendall trend test method, and the trend judgment result of each early warning index at the moment of occurrence of the rock burst event for 15 times is shown in Table 1:
TABLE 1 Trend determination results of various early warning indicators before rock burst
Note: "/" indicates no significant trend in the index.
And combining the change trend of each index with the rock burst precursor representation rule to perform early warning, wherein if the index is early warned within 5d before the rock burst event, the early warning on the rock burst danger is correct.
The early warning performance of each index is evaluated by using a confusion matrix, and the calculation result is shown in table 2:
table 2 evaluation of rock burst warning index warning effectiveness
Index optimization is carried out based on the principle of maximization of early warning efficiency, and as can be seen from table 2, the early warning efficiencies of the early warning indexes are arranged in the following sequence:
according to the early warning efficiency maximization principle, preferably, early warning efficiencies corresponding to indexes within 40% of the early warning efficiency ranking in the precursor early warning index library are input into the rock burst comprehensive early warning model, and E is selected according to the calculation result in the table 2 max ,D F ,E avg ,E sum ,D E ,F sum And 8 indexes such as AC, P (b), etc.
The time sequence variation curve of the impact risk comprehensive abnormal index Q obtained by carrying out multi-index fusion by using the comprehensive abnormal index method is shown in FIG. 22. Different values of Q correspond to different impact hazard levels, as shown in table 3.
TABLE 3 impact ground pressure hazard classification
| Comprehensive abnormal index Q of coal rock dynamic disaster danger | Hazard class | State of |
| 0≤Q＜0.25 | Class I | Without risk of impact |
| 0.25≤Q＜0.5 | Class II | Danger of weak impact |
| 0.5≤Q＜0.75 | Class III | Danger of medium impact |
| 0.75≤Q≤1 | Grade IV | Danger of high impact |
E is selected as the early warning efficiency of the impact danger comprehensive abnormity index Q max ,D F ,E avg ,E sum ,D E ,F sum And 8 indexes such as AC, P (b), and the like reach 0.563, which is higher than the early warning efficiency of any single early warning index. Then when the site is on every 1 month or more (or the working face is changed, the geological conditions are greatly changed, and the high-energy mine earthquake occurs orAnd in case of rock burst events and the like, the optimization and weight determination of the indexes are carried out again so as to adapt to complicated working condition changes on site.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. A rock burst multi-parameter dynamic trend early warning method is characterized by comprising the following steps:
s101, according to real-time monitoring data of a field microseismic monitoring system, describing the variation trend of a 'time-space-strong' multi-element precursor early warning index in the process of the initiation and evolution of the rock burst by using a Mann-Kendall trend inspection method, and carrying out early warning by taking whether the variation trend of each index meets the representation rule of the rock burst precursor or not as an early warning criterion;
s102, evaluating the early warning efficiency of each index by using a confusion matrix;
s103, performing index optimization based on an early warning effectiveness maximization principle;
s104, performing multi-index fusion by taking the early warning efficiency corresponding to the optimized index as weight to obtain an impact risk comprehensive abnormal index;
s105, comparing the comprehensive impact risk abnormal index with a corresponding quantitative grading standard to determine the impact risk grade;
s106, periodically carrying out index optimization and weight updating under the drive of on-site actual monitoring data;
the method comprises the following steps of describing the variation trend of a 'time-space-strong' multi-element precursor early warning index in the process of the initiation and evolution of the rock burst by utilizing a Mann-Kendall trend inspection method according to real-time monitoring data of a field microseismic monitoring system, and carrying out early warning by taking whether the variation trend of each index meets the representation rule of the rock burst precursor as an early warning criterion:
acquiring real-time monitoring data of a field microseismic monitoring system, and preprocessing original monitoring data by using a specific time window and a slip step length to obtain a time sequence of a time-space-strong multi-precursor early warning index of rock burst danger; wherein, the time-space-strength multi-element precursor early warning indexes of the rock burst danger comprise:
reflecting the daily total frequency, frequency deviation value, average total frequency, lack of earthquake, A (b) value, microseismic activity scale, algorithm complexity, P (b) value and time information entropy of a time dimension; wherein the values of A (b) and P (b) are derived from the empirical constant b in the Goodenberg-Rickett equation, one of the basic laws of seismology, and the values of A (b) are expressed as:
wherein N is total number of microseisms, M i Is the microseismic event energy level;
the value of P (b) is expressed as:
reflecting daily maximum energy, daily total energy, daily average energy, energy deviation value, average total energy, total fault area and b value of the intensity dimension;
the daily maximum energy, daily total frequency, daily average energy, energy deviation value, frequency deviation value, average total energy, microseismic activity degree, lack of shock, A (b) value, total fault area, average total frequency, microseismic activity scale and algorithm complexity belong to forward early warning indexes, namely the higher the value is, the greater the impact risk is; the seismic source concentration degree, the b value, the P (b) value and the time information entropy belong to negative early warning indexes, namely the lower the value is, the greater the impact risk is;
judging the real-time change trend of each early warning index by using a Mann-Kendall trend inspection method, and early warning by taking whether the change trend of each index meets the rock burst precursor representation rule as an early warning criterion;
the preprocessing of the original monitoring data by the specific time window and the slip step length refers to converting the original monitoring data from an irregular time sequence to a regular time sequence, and comprises the following steps:
defining a sliding time window with the length delta T, dividing the acquired monitoring data time sequence into n data sets with the length delta T and respectively corresponding to the time moments at the end of the time window, namely T i The data set at time is denoted X i [x 1 ,x 2 ,x 3 ,...,x k ]，k≤t,0<i is less than or equal to n, wherein, delta T is the length of the sliding time window, T i The time corresponding to the end of the ith time window;
calculating X i Early warning index value y corresponding to all samples in i The early warning index value y i Arranging in sequence to obtain a converted regular time sequence marked as Y [ Y ] 1 ,y 2 ,y 3 ,...,y n ]；
The method for judging the real-time change trend of each early warning index by using a Mann-Kendall trend test method comprises the following steps:
the early warning index is regulated to be in a time sequence Y [ Y ] 1 ,y 2 ,y 3 ,...,y n ]Dividing the data into m data sets with time window length delta a and respectively corresponding to the time A at the end of the time window i I.e. A i The dataset of the time of day is Y i [y 1 ,y 2 ,y 3 ,...,y q ]，10≤q≤a,0<i is less than or equal to m, and Y is calculated i Test statistic S:
wherein sgn (·) represents a sign function,
q represents Y i [y 1 ,y 2 ,y 3 ,...,y q ]Length of data set, y p Denotes the pth data, p 1,2,3.. q-1, y j Represents the jth data, j ═ p +1,2,3.. q;
according to the obtained Y i Of the test statistic S, determining Y i Test standard quantity Z：
When Z is>At 0, T i The early warning indexes at all times have a growing trend; when Z is<At 0, T i The early warning index at the moment has a decreasing trend; when Z is 0, T i The early warning indexes at the moment have no obvious variation trend.
2. The method for early warning of the multi-parameter dynamic trend of rock burst according to claim 1, wherein the early warning by taking whether the change trend of each index meets the precursor characterization rule of rock burst as an early warning criterion comprises the following steps:
and comparing the real-time change trend of each index with the rock burst precursor representation rule, performing early warning if the positive early warning index has an increasing trend, performing early warning if the negative early warning index has a decreasing trend, and not performing early warning if other change trends.
3. The rock burst multi-parameter dynamic trend warning method according to claim 1, wherein the warning effectiveness is expressed as:
wherein, EFF represents the early warning efficiency; recall represents the Recall rate of the call,
indicating that the warning is an impactDangerous and actual occurrence of a rock burst event,
indicating that the early warning is no impact danger but a rock burst event actually occurs; precision indicates the rate of Precision at which the Precision,
wherein
Indicating that the pre-warning is a shock hazard and that a rock burst event has actually occurred,
indicating that the early warning is a shock hazard but that no rock burst event actually occurs.
4. The rock burst multi-parameter dynamic trend early warning method according to claim 1, wherein the performing indexes based on an early warning effectiveness maximization principle preferably comprises:
and arranging the early warning effect values of all the early warning indexes from large to small, and screening indexes n% of which are ranked before for the next data fusion, wherein n is a positive integer, and the number of the indexes is rounded upwards.
5. The rock burst multi-parameter dynamic trend early warning method according to claim 1, wherein the obtaining of the comprehensive impact risk abnormality index by performing multi-index fusion with early warning effectiveness corresponding to the preferred index as a weight comprises:
and inputting the early warning efficiency corresponding to the optimized indexes into the comprehensive early warning model of rock burst as weight, and performing multi-index fusion on the comprehensive early warning model of rock burst by using a comprehensive abnormal index method to obtain a comprehensive abnormal index of the impact risk:
wherein Q represents a shock hazard comprehensive abnormality index; e represents a natural base number; n represents the total number of preferred indicators; EFF k The early warning efficiency corresponding to the kth index; w k(+/-) Representing the abnormal membership of the kth positive/negative early warning index,
for the forward early warning index, the value of the abnormal membership degree is as follows:
for negative early warning indexes, the value of the abnormal membership degree is as follows:
6. the rock burst multi-parameter dynamic trend warning method according to claim 5, wherein the impact risk level comprises: a no-impact-risk state, a weak-impact-risk state, a medium-impact-risk state, and a strong-impact-risk state.
7. The early warning method for the multiparameter dynamic trend of rock burst as claimed in claim 1, wherein said periodically performing index optimization and weight update under the drive of on-site actual monitoring data comprises:
and (4) performing index optimization periodically under the drive of field actual monitoring data, and inputting the early warning efficiency corresponding to the optimized index into the comprehensive early warning model of rock burst as weight so as to realize self-feedback updating of the comprehensive early warning model of rock burst.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111258391.3A CN114087021B (en) | 2021-10-27 | 2021-10-27 | Rock burst multi-parameter dynamic trend early warning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111258391.3A CN114087021B (en) | 2021-10-27 | 2021-10-27 | Rock burst multi-parameter dynamic trend early warning method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114087021A CN114087021A (en) | 2022-02-25 |
| CN114087021B true CN114087021B (en) | 2022-08-02 |
Family
ID=80297936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111258391.3A Active CN114087021B (en) | 2021-10-27 | 2021-10-27 | Rock burst multi-parameter dynamic trend early warning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114087021B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114889546A (en) * | 2022-03-28 | 2022-08-12 | 郑州炜盛电子科技有限公司 | Living body detection method and device based on carbon dioxide sensor |
| CN115239840A (en) * | 2022-07-21 | 2022-10-25 | 临沂矿业集团菏泽煤电有限公司郭屯煤矿 | Impact early warning method based on microseismic activity process time sequence analysis |
| CN116048235B (en) * | 2023-03-29 | 2023-06-16 | 南京群顶科技股份有限公司 | Temperature-sensing future trend detection method based on bidirectional GRU and Mankendel method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU1788289C (en) * | 1990-10-22 | 1993-01-15 | Институт Физики И Механики Горных Пород Ан Киргсср | Method for prediction of rock bursts in mine workings |
| CN101582191B (en) * | 2009-06-24 | 2010-11-10 | 上海添成电子科技有限公司 | Method for realizing micro-seismic monitoring and real-time early warning to mine power disaster |
| CN102644482B (en) * | 2012-05-18 | 2014-04-02 | 河南大有能源股份有限公司 | Rock burst predicting and warning method |
| CN109447837B (en) * | 2018-11-15 | 2022-05-20 | 国家能源投资集团有限责任公司 | Danger assessment method for rock burst in mining area |
| CN110779574B (en) * | 2019-10-30 | 2020-11-13 | 北京科技大学 | Coal rock dynamic disaster multi-system multi-parameter integrated comprehensive early warning method and system |
| CN112324506B (en) * | 2020-11-20 | 2024-05-14 | 上海大屯能源股份有限公司江苏分公司 | Dynamic early warning method for preventing rock burst of coal mine based on microseism |
-
2021
- 2021-10-27 CN CN202111258391.3A patent/CN114087021B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114087021A (en) | 2022-02-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114087021B (en) | Rock burst multi-parameter dynamic trend early warning method | |
| CN107977749B (en) | Prediction analysis method for oil field development index | |
| CN103291364B (en) | Microseismic multidimensional information comprehensive time sequence early warning method for rock burst | |
| Kabiesz et al. | Application of rule-based models for seismic hazard prediction in coal mines. | |
| Liu et al. | Microseismicity-based method for the dynamic estimation of the potential rockburst scale during tunnel excavation | |
| CN104153814A (en) | Micro-seismic multi-parameter early warning method for rock burst | |
| CN105607127A (en) | Microquake multi-parameter early warning method of rock burst in high-stress concentration area | |
| Li et al. | Automatic recognition of effective and interference signals based on machine learning: A case study of acoustic emission and electromagnetic radiation | |
| CN114294062B (en) | Time-space dynamic comprehensive early warning method for rock burst | |
| CN113253344B (en) | Method for realizing pressure raising early warning of underground gas storage based on microseism monitoring technology | |
| Liang et al. | Prediction model for coal-gas outburst using the genetic projection pursuit method | |
| CN113673119A (en) | Dynamic and static coupling evaluation method for coal mine rock burst danger based on Bayes method | |
| Ma et al. | Rockburst prediction model using machine learning based on microseismic parameters of Qinling water conveyance tunnel | |
| CN114895352B (en) | Rock mass instability prediction method and device based on microseismic monitoring | |
| CN117152918A (en) | Dangerous dynamic early warning method applied to regional landslide disaster | |
| Xue et al. | Integrated rockburst hazard estimation methodology based on spatially smoothed seismicity model and Mann-Kendall trend test | |
| Pytel et al. | Universal approach for risk identification and evaluation in underground facilities | |
| CN113266421B (en) | Comprehensive early warning method for full-dangerous period time and space of rock burst | |
| Vallejos | Analysis of seismicity in mines and development of re-entry protocols | |
| CN114692410B (en) | Mine dynamic disaster prediction method based on ground stress field inversion numerical range-to-average analysis | |
| CN116070907A (en) | Karst collapse susceptibility assessment method and system based on analytic hierarchy process | |
| Basnet et al. | Developing an explainable rockburst risk prediction method using monitored microseismicity based on interpretable machine learning approach | |
| CN113374525A (en) | Coal and gas outburst danger area identification comprehensive early warning method based on multi-parameter data fusion | |
| Hosseini et al. | Application of fuzzy logic for determining of coal mine mechanization | |
| Liu et al. | Quantitative risk assessment for deep tunnel failure based on clustering and logistic regression at the Ashele copper mine, China |
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 |