CN-122017965-A - Coal rock mass fracture precursor signal identification method, device, medium and equipment

CN122017965ACN 122017965 ACN122017965 ACN 122017965ACN-122017965-A

Abstract

The application discloses a method, a device, a medium and equipment for identifying a coal rock mass fracture precursor signal, wherein the method comprises the steps of carrying out noise reduction treatment on an original voltage signal corresponding to an underground sound signal and/or a vibration signal to obtain a noise-reduced signal; the method comprises the steps of carrying out time-frequency analysis on a noise-reduced signal to obtain a time spectrum, converting the time spectrum into an energy spectrum, constructing a three-dimensional map based on the time spectrum and the energy spectrum, calculating frequency barycentric coordinates, three-dimensional energy entropy and frequency band energy ratio in three-dimensional energy based on the three-dimensional map, constructing a characteristic time sequence according to time sequence of the frequency barycentric coordinates, the three-dimensional energy entropy and the frequency band energy ratio in the three-dimensional energy, carrying out smoothing treatment, analyzing the characteristic sequence of the frequency barycentric coordinates, the characteristic sequence of the energy entropy and the characteristic sequence of the frequency band energy and the corresponding preset precursor characteristic sequence, and obtaining early warning levels according to analysis results. The application improves the accuracy and reliability of precursor signal identification and early warning.

Inventors

ZHANG CHUANJIU
LI XUANLIANG
LU ZHEN
CHEN JIE
PU YUANYUAN
SHI KAIWEN
RUI YICHAO
ZHOU ZELIN
JIA YULIANG
YANG JUN
ZHANG YUNFEI

Assignees

国能神东煤炭集团有限责任公司
中国神华能源股份有限公司神东煤炭分公司
重庆大学

Dates

Publication Date: 20260512
Application Date: 20251208

Claims (10)

1. A method for identifying a coal rock mass fracture precursor signal, comprising: Acquiring an original voltage signal corresponding to an underground sound signal and/or a vibration signal, and performing noise reduction treatment on the original voltage signal to obtain a noise-reduced signal; Performing wavelet transformation on the noise-reduced signal to obtain a wavelet transformation time spectrum, converting the wavelet transformation time spectrum into an energy spectrum, and constructing a three-dimensional map based on the wavelet transformation time spectrum and the energy spectrum; calculating frequency barycentric coordinates in three-dimensional energy based on the three-dimensional map, and calculating three-dimensional energy entropy and frequency band energy ratio based on the three-dimensional map; Respectively constructing a characteristic time sequence according to the frequency barycentric coordinates, the three-dimensional energy entropy and the frequency band energy ratio in the three-dimensional energy in time sequence, and performing smoothing treatment to obtain a smoothed frequency barycentric coordinate characteristic sequence, an energy entropy characteristic sequence and a frequency band energy characteristic sequence; and analyzing the smoothed frequency barycentric coordinate characteristic sequence, the energy entropy characteristic sequence and the frequency band energy bit characteristic sequence with the corresponding preset precursor characteristic sequences, and obtaining an early warning level according to an analysis result.
2. The method for identifying a coal rock mass fracture precursor signal according to claim 1, wherein the noise reduction processing is performed on the original voltage signal to obtain a noise-reduced signal, and the method comprises the following steps: Decomposing the original voltage signal by adopting an adaptive noise set empirical mode decomposition method to obtain a plurality of eigenmode function components and residual items, and screening effective signals of the eigenmode function components to obtain effective component signals and ineffective component signals; respectively carrying out quantization processing on the effective component signal and the ineffective component signal by adopting a semi-soft threshold function; and reconstructing the residual item, the quantized effective component signal and the quantized ineffective component signal to obtain a noise-reduced signal.
3. The method for identifying the coal rock mass fracture precursor signal according to claim 2, wherein the screening of the effective signal for the eigenmode function component to obtain the effective component signal and the ineffective component signal comprises the following steps: Calculating the correlation coefficient and variance contribution rate of each eigen-mode function component and the original voltage signal respectively; and taking the eigenmode function components with the correlation coefficients larger than a preset coefficient threshold value and the variance contribution rate larger than the preset contribution rate as effective component signals, and taking the rest eigenmode function components as ineffective component signals.
4. The method for identifying a coal rock mass fracture precursor signal according to claim 2, wherein the step of performing wavelet transform on the noise-reduced signal to obtain a wavelet-transformed frequency spectrum, and converting the wavelet-transformed frequency spectrum into an energy spectrum comprises the steps of: performing continuous wavelet transformation on the noise-reduced signal to obtain wavelet coefficients, and calculating the instantaneous frequency of the wavelet coefficients; Calculating based on the wavelet coefficient and the instantaneous frequency of the wavelet coefficient, and mapping the wavelet coefficient from a scale-time plane to a frequency-time plane to obtain a frequency spectrum during wavelet transformation; and taking the square of the absolute value of the complex time spectrum in the wavelet transformation time spectrum as the energy spectrum corresponding to the wavelet transformation time spectrum.
5. The method for identifying the coal rock mass fracture precursor signals according to claim 1, wherein the following method is adopted to calculate the frequency barycentric coordinates in three-dimensional energy: Where P (t m ,f k ) is the energy intensity at time t m and frequency f k , f c is the frequency barycentric coordinates in three-dimensional energy, M is the total number of time discrete points, and K is the total number of frequency discrete points.
6. The method for identifying a coal rock mass fracture precursor signal according to claim 1, wherein the calculating a three-dimensional energy entropy and a frequency band energy ratio based on the three-dimensional map comprises: based on the three-dimensional map, calculating the probability of three-dimensional energy distribution to obtain the probability of energy intensity distribution on time-frequency points; Calculating based on the probability of the energy intensity distribution on the time-frequency points to obtain three-dimensional energy entropy; dividing a time spectrum into a first frequency band and a second frequency band based on a preset frequency division threshold, and respectively calculating the low-frequency total energy of the first frequency band and the high-frequency total energy of the second frequency band based on the energy intensity in the three-dimensional map; the ratio of the high-frequency total energy to the low-frequency total energy is taken as a band energy ratio.
7. The method for identifying a fracture precursor signal of a coal rock mass according to claim 1, wherein the analyzing the smoothed frequency barycentric coordinate feature sequence, the energy entropy feature sequence, and the frequency band energy bit feature sequence with the respective corresponding preset precursor feature sequences, and obtaining the pre-warning level according to the analysis result comprises: Calculating a first distance between the smoothed frequency barycentric coordinate feature sequence and a corresponding preset precursor feature sequence, determining a trend of the frequency barycentric coordinate in a preset period based on the smoothed frequency barycentric coordinate feature sequence, and determining an early warning value of the frequency barycentric coordinate according to the trend and the first distance; Calculating a second distance between the smoothed energy entropy characteristic sequence and a corresponding preset precursor characteristic sequence, determining a descending mode of the energy entropy based on the smoothed energy entropy characteristic sequence, and determining an early warning value of the energy entropy according to the descending mode and the second distance; Calculating a third distance between the smoothed frequency band energy bit sequence and a corresponding preset precursor feature sequence, determining an inflection point mode of the frequency band energy ratio based on the smoothed frequency band energy bit sequence, and determining an early warning value of the frequency band energy ratio according to the inflection point mode and the third distance; and calculating to obtain a total early warning level based on the early warning value of the frequency barycentric coordinates, the early warning value of the energy entropy, the early warning value of the frequency band energy ratio and the weight corresponding to the frequency band energy ratio.
8. A coal rock mass fracture precursor signal identification device, comprising: The noise reduction module is used for acquiring an original voltage signal corresponding to the underground sound signal and/or the vibration signal, and carrying out noise reduction treatment on the original voltage signal to obtain a noise-reduced signal; the three-dimensional spectrum construction module is used for carrying out wavelet transformation on the noise-reduced signal to obtain a wavelet transformation time spectrum, converting the wavelet transformation time spectrum into an energy spectrum and constructing a three-dimensional spectrum based on the wavelet transformation time spectrum and the energy spectrum; the characteristic value calculation module is used for calculating the frequency barycentric coordinates in the three-dimensional energy based on the three-dimensional map, and calculating the three-dimensional energy entropy and the frequency band energy ratio based on the three-dimensional map; the characteristic sequence acquisition module is used for constructing a characteristic time sequence according to the frequency barycentric coordinates, the three-dimensional energy entropy and the frequency band energy ratio in the three-dimensional energy in time sequence and performing smoothing processing to obtain a smoothed frequency barycentric coordinate characteristic sequence, an energy entropy characteristic sequence and a frequency band energy bit characteristic sequence; And the analysis module is used for analyzing the smoothed frequency barycentric coordinate characteristic sequence, the energy entropy characteristic sequence and the frequency band energy bit sequence with the corresponding preset precursor characteristic sequences, and obtaining the early warning level according to the analysis result.
9. A storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the coal rock mass fracture precursor signal identification method of any one of claims 1-7.
10. A computer device comprising a processor, a memory, a communication interface and a communication bus, said processor, said memory and said communication interface completing communication with each other through said communication bus; The memory is configured to store at least one executable instruction that causes the processor to perform operations corresponding to the coal rock mass fracture precursor signal identification method of any one of claims 1-7.

Description

Coal rock mass fracture precursor signal identification method, device, medium and equipment Technical Field The invention relates to the technical field of geotechnical engineering, in particular to a method, a device, a medium and equipment for identifying a coal rock mass fracture precursor signal. Background Destabilization and rupture of coal rock mass (such as rock burst, coal and gas protrusion, roof caving, etc.) are serious disasters in coal mine safety production. Numerous theories and practices indicate that coal and rock masses undergo internal microcrack initiation, propagation, and penetration processes prior to destabilization, with concomitant generation of a large number of Acoustic Emission (AE) or Microseismic (MS) signals. The signals contain abundant cracking precursor information, so that analysis of the monitored signals to identify the cracking precursor is a key for realizing accurate disaster early warning. Existing coal and rock fracture precursor signal identification methods mainly rely on single parameters or simple joint analysis of signal time domain parameters (such as ringing count, energy, duration) or frequency domain parameters (such as dominant frequency, spectrum center of gravity). However, these methods have the following significant problems and disadvantages: 1. The precursor feature extraction precision is low, and the single dimension analysis of the time domain or the frequency domain cannot comprehensively describe the non-stationary and nonlinear acoustic emission/microseismic signal features. The time domain analysis loses the frequency structure information, the frequency domain analysis (such as FFT) loses the time evolution information, the essential characteristics of precursor signal transient and abrupt change are difficult to capture, and the extraction precision is low. Because of the weak energy of the burst precursor signal, the frequency component is complex, and there is much overlap with the background noise band. The simple time-frequency analysis is difficult to construct a high-resolution and high-definition characteristic spectrum, so that the sensitivity to the precursor tiny change is insufficient, and the extraction precision is low. 2. The anti-interference capability is weak, underground environmental noise (such as mechanical vibration and electromagnetic interference) is complex, and the traditional method is difficult to effectively distinguish effective signals and noise generated by real rock mass fracture, so that false alarm or missing alarm is easily caused. Therefore, a high-precision precursor signal identification method with high noise immunity is highly demanded. Disclosure of Invention In view of the above, the invention provides a coal rock mass fracture precursor signal identification method, which mainly aims to solve the problems of low identification accuracy and weak anti-interference capability of the existing coal rock mass fracture precursor signal identification method. According to one aspect of the present application, there is provided a coal rock mass fracture precursor signal identification method, the method comprising: Acquiring an original voltage signal corresponding to an underground sound signal and/or a vibration signal, and performing noise reduction treatment on the original voltage signal to obtain a noise-reduced signal; Performing wavelet transformation on the noise-reduced signal to obtain a wavelet transformation time spectrum, converting the wavelet transformation time spectrum into an energy spectrum, and constructing a three-dimensional map based on the wavelet transformation time spectrum and the energy spectrum; calculating frequency barycentric coordinates in three-dimensional energy based on the three-dimensional map, and calculating three-dimensional energy entropy and frequency band energy ratio based on the three-dimensional map; Respectively constructing a characteristic time sequence according to the frequency barycentric coordinates, the three-dimensional energy entropy and the frequency band energy ratio in the three-dimensional energy in time sequence, and performing smoothing treatment to obtain a smoothed frequency barycentric coordinate characteristic sequence, an energy entropy characteristic sequence and a frequency band energy characteristic sequence; and analyzing the smoothed frequency barycentric coordinate characteristic sequence, the energy entropy characteristic sequence and the frequency band energy bit characteristic sequence with the corresponding preset precursor characteristic sequences, and obtaining an early warning level according to an analysis result. Optionally, the noise reduction processing is performed on the original voltage signal to obtain a noise reduced signal, which includes: Decomposing the original voltage signal by adopting an adaptive noise set empirical mode decomposition method to obtain a plurality of eigenmode function components and residua