阅读说明：本技术 基于声纹识别技术的岩石破裂模式分类识别方法 (Rock fracture mode classification and identification method based on voiceprint identification technology ) 是由 姚旭龙 张艳博 刘祥鑫 梁鹏 孙林 于 2021-08-05 设计创作，主要内容包括：本发明公开了基于声纹识别技术的岩石破裂模式分类识别方法,包括以下步骤：S1在待监测岩石周围设置声发射系统；S2构建声发射声谱分析模型和理想信号生成方法与分类特征判别标准；S3通过声发射系统中每个通道的声发射监测传感器依次采集信号,信号输入声谱分析模型计算获取声谱特征图和时频强度分布特征图；S4根据构建的信号分类判别标准,综合岩石破裂事件对应的声谱特征图和时频强度分布特征图完成判断岩石破裂的分类；本发明不仅能够实现脆性材料破裂任一声发射信号包含的破裂类型识别,还能确定破裂的时间、强度等分布情况。(The invention discloses a rock fracture mode classification and identification method based on a voiceprint identification technology, which comprises the following steps: s1 arranging an acoustic emission system around the rock to be monitored; s2, constructing an acoustic emission sound spectrum analysis model, an ideal signal generation method and a classification characteristic judgment standard; s3, sequentially acquiring signals through the acoustic emission monitoring sensor of each channel in the acoustic emission system, and inputting the signals into an acoustic spectrum analysis model to calculate and obtain an acoustic spectrum characteristic diagram and a time-frequency intensity distribution characteristic diagram; s4, according to the constructed signal classification judgment standard, the classification of rock fracture is judged by integrating the acoustic spectrum characteristic diagram and the time-frequency intensity distribution characteristic diagram corresponding to the rock fracture event; the method can not only realize the fracture type identification contained in any acoustic emission signal of the fracture of the brittle material, but also determine the distribution conditions of the time, the strength and the like of the fracture.)

1. The rock fracture mode classification and identification method based on the voiceprint recognition technology is characterized by comprising the following steps of:

s1, arranging an acoustic emission system around the rock to be monitored;

s2, constructing an acoustic emission acoustic spectrum analysis model, an ideal signal generation method and a classification characteristic judgment standard;

s3, sequentially acquiring signals through the acoustic emission monitoring sensor of each channel in the acoustic emission system, and inputting the signals into an acoustic spectrum analysis model to calculate and obtain an acoustic spectrum characteristic diagram and a time-frequency intensity distribution characteristic diagram;

and S4, according to the constructed signal classification judgment standard, integrating the acoustic spectrum characteristic diagram and the time-frequency intensity distribution characteristic diagram corresponding to the rock fracture event to finish the classification of the rock fracture.

2. The method for classifying and identifying the rock fracture pattern based on the voiceprint recognition technology as claimed in claim 1, wherein the ideal signal generation method and the classification feature discrimination criteria in S2 include tension fracture, shear fracture and tension-shear composite fracture.

3. The method for identifying rock cracking pattern classification based on the voiceprint recognition technology as claimed in claim 2, wherein the ideal signal generation method of rock tension cracking and shear cracking in S2 is as follows:

where A represents the amplitude of the signal, ω is the angular frequency, and h is the damping factor.

4. The method for classifying and identifying the rock breaking pattern based on the voiceprint recognition technology as claimed in claim 1, wherein the acoustic emission sonogram analysis model in the S2 is as follows:

wherein h (m) is a window function, a Hamming window is selected,in order to be the frequency of the radio,

let S (n, ω) ═ STFT_x(n,ω)|²And representing the intensity of frequency components corresponding to the time n and the frequency omega, and further forming a spectrogram of the acoustic emission signal x (m).

5. The method for classifying and identifying a rock breaking pattern based on the voiceprint recognition technology as claimed in claim 1, wherein the acoustic emission system in S1 is composed of a matrix of correlation type acoustic emission monitoring sensors.

Technical Field

The invention belongs to the technical field of rock fracture damage detection, and particularly relates to a rock fracture mode classification and identification method based on a voiceprint identification technology.

Background

Rock fracture instability is always a key problem in rock mechanics research at home and abroad, and a scientific method is adopted to research an internal fracture evolution law in the rock fracture instability process, so that the method has important significance in understanding the internal fracture mechanism of the rock and monitoring and early warning.

Different rock fracture modes generate different types of cracks, the cracks are basic expression forms of rock fracture, and the initiation, propagation and penetration of the cracks form a rock fracture instability process. Accurate determination of the basic type of rock fracture cracking is fundamental to a proper understanding of the rock fracture destabilization process. Monitoring of the destabilizing process of fracturing is difficult due to the opacity of the rock material. The existing detection technology mainly comprises acoustic emission monitoring and ultrasonic detection, and the ultrasonic and acoustic emission nondestructive detection technology is most widely applied in relevant fields such as geotechnical engineering and the like. The method for distinguishing the crack type in the rock breaking process by using the characteristics of the acoustic emission monitoring signal becomes a general standard method in the industry (JCMS-III B5706(2003) Japan). The characteristics of rock cracking acoustic emission signals are deeply researched, a new method is developed, the crack type judgment accuracy is improved, and the method has important scientific research value and engineering application value.

Disclosure of Invention

The invention aims to provide a rock fracture mode classification and identification method based on a voiceprint identification technology, aiming at the defects, not only can the fracture type identification included in any acoustic emission signal of the brittle material fracture be realized, but also the distribution situation of the fracture time, the fracture intensity and the like can be determined.

The technical scheme of the invention is as follows: the rock fracture mode classification and identification method based on the voiceprint identification technology comprises the following steps:

s1, arranging an acoustic emission system around the rock to be monitored;

s2, constructing an acoustic emission acoustic spectrum analysis model, an ideal signal generation method and a classification characteristic judgment standard;

s3, sequentially acquiring signals through the acoustic emission monitoring sensor of each channel in the acoustic emission system, and inputting the signals into an acoustic spectrum analysis model to calculate and obtain an acoustic spectrum characteristic diagram and a time-frequency intensity distribution characteristic diagram;

and S4, according to the constructed signal classification judgment standard, integrating the acoustic spectrum characteristic diagram and the time-frequency intensity distribution characteristic diagram corresponding to the rock fracture event to finish the classification of the rock fracture.

Preferably, the ideal signal generation method and the classification feature discrimination criteria in S2 include tension fracture, shear fracture, and tension-shear composite fracture.

Preferably, the ideal signal generation method for rock tension fracture and shear fracture in S2 is:

ω＝2πf

where A represents the amplitude of the signal, ω is the angular frequency, and h is the damping factor.

Preferably, the acoustic emission spectrum analysis model in S2 is:

wherein h (m) is a window function, a Hamming window is selected,in order to be the frequency of the radio,

let S (n, ω) ═ STFT_x(n,ω)|²And representing the intensity of frequency components corresponding to the time n and the frequency omega, and further forming a spectrogram of the acoustic emission signal x (m).

Preferably, the acoustic emission system in S1 is comprised of a matrix of correlation-type acoustic emission monitoring sensors.

The rock fracture mode classification and identification method based on the voiceprint identification technology has the beneficial effects that:

(1) the method can identify the type of the corresponding fracture mode by using a single acoustic emission signal, does not need to determine the classification boundary of the fracture mode through all signals in the rock fracture process, overcomes the dependency on the overall process information, and can lay a foundation for the real-time identification of the rock fracture mode.

(2) The method can realize classification and identification of the fracture events corresponding to the multi-fracture acoustic emission superposition composite signal, effectively solves the problem that a JCMS-III B5706 universal standard method proposed in Japan cannot identify a plurality of fracture types in one acoustic emission signal, improves the accuracy of fracture type identification, and provides technical support for correctly understanding the rock fracture evolution mechanism.

(3) The method realizes the identification of a plurality of acoustic emission components of the fracture event in an acoustic emission signal, greatly improves the time-out accuracy in the acoustic emission positioning technology, effectively solves the problem of low acoustic emission positioning accuracy, and can lay a foundation for rock fracture evolution mechanism research and high-accuracy positioning early warning of on-site disasters.

Drawings

Fig. 1 is a flowchart of a rock fracture pattern classification recognition method based on a voiceprint recognition technology according to an embodiment of the invention.

Fig. 2 is a schematic diagram illustrating a matrix configuration of a correlation acoustic emission monitoring sensor according to an embodiment of the present invention.

FIG. 3 is a schematic diagram illustrating the generation of an ideal signal for tensile fracture according to an embodiment of the present invention.

FIG. 4 is a diagram illustrating the generation of ideal shear fracture signals according to an embodiment of the present invention.

FIG. 5 is a display diagram of the acoustic emission spectrum feature analysis recognition criteria of burst mode provided by the embodiment of the present invention.

Fig. 6 is a graph showing an actual acoustic emission waveform of a rock fracture according to an embodiment of the present invention.

Fig. 7 shows an actual acoustic emission spectrogram of rock fracture provided by an embodiment of the present invention.

Fig. 8 is a diagram illustrating a characteristic distribution of actual acoustic emission time-frequency intensity of rock fracture according to an embodiment of the present invention.

Fig. 9 shows the intention of highlight search result of audio spectrum analysis according to the embodiment of the present invention.

Fig. 10 is a schematic diagram illustrating a fracture pattern analysis result according to an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It is to be understood that the embodiments shown and described in the drawings are merely exemplary and are intended to illustrate the principles and spirit of the invention, not to limit the scope of the invention.

The embodiment of the invention provides a rock fracture acoustic emission and damage imaging integrated monitoring method, which comprises the following steps of S1-S4 as shown in FIG. 1:

s1, arranging an acoustic emission system around the rock to be monitored.

In the embodiment of the invention, the acoustic emission system is composed of a correlation type acoustic emission monitoring sensor matrix. The specific arrangement of the correlation type acoustic emission monitoring sensor matrix is shown in fig. 2.

And S2, constructing an acoustic emission acoustic spectrum analysis model, an ideal signal generation method and a classification characteristic discrimination standard.

Acoustic emission signal is constructed based on voiceprint recognition speech spectrum analysis principleThe sound spectrum analysis mathematical model is as follows:

where h (m) is a window function, a Hamming window is selected,is the frequency.

Let S (n, ω) ═ STFT_x(n,ω)|²And representing the intensity of frequency components corresponding to the time n and the frequency omega, and further forming a spectrogram of the acoustic emission signal x (m).

And constructing an ideal acoustic emission signal of the rock fracture by utilizing a compounding mode of an exponential signal and a sinusoidal signal, wherein a signal compounding function is as follows:

ω＝2πf

where A represents the amplitude of the signal, ω is the angular frequency, and h is the adjustment factor.

Constructing a tension fracture signal: a 8, f 50kHz, h₀＝32，h₁The sampling length is 2048, yielding a signal diagram as in fig. 3.

Constructing a shear fracture signal: a 5, f 50kHz, h₀＝8，h₁The sample length is 2048, 16, resulting in the signal diagram of fig. 4.

Performing sound spectrum analysis on the constructed signals to establish a classification standard, wherein the characteristic distribution is shown in fig. 5, and the following classification criteria are formed:

basic conditions:

condition 1: there are distinct high energy points O in the spectrogram at certain frequency bands.

Condition 2: a distinct continuous decreasing band is formed around the high energy point O along the time axis over a range.

(1) Single tension rupture pattern recognition criterion

The spectrogram of the cracking acoustic emission signal meets the conditions 1 and 2, and has the following characteristics:

the continuous tapered strip in condition 2 extends slightly or not in the counterclockwise direction, but has a longer extension in the clockwise direction with respect to the counterclockwise direction, and is shaped like a "comet" of a counterclockwise impact as a whole;

time-frequency intensity contour map reverse time extension length L drawn by spectrogram₁And a clockwise extension length L₂The ratio is less than 0.5;

meanwhile, the acoustic emission signal satisfying the conditions 1 and 2 is a tensile fracture (see fig. 5 a).

(2) Single shear crack pattern recognition criterion

The spectrogram of the cracking acoustic emission signal meets the conditions 1 and 2, and has the following characteristics:

the continuous tapered strip in condition 2 has similar extension in both the counterclockwise direction and the clockwise direction, or the counterclockwise extension is stronger than the clockwise extension;

the ratio of the time-frequency intensity contour map reverse-time extension length L1 to the forward-time extension length drawn by the spectrogram is more than or equal to 1, and the external contour of the upper contour is smooth;

meanwhile, the acoustic emission signal satisfying the conditions 1 and 2 is a shear fracture (see fig. 5 b).

(3) Tension-shear composite signal identification criterion of different frequency bands

And (3) the spectrogram of the acoustic emission signal meets the conditions 1 and 2, and respectively meets the single stretch-break pattern recognition criterion or the single shear-break pattern recognition criterion in different frequency band ranges, and the signal is formed by overlapping and compounding a stretch-break signal and a shear-break signal (see fig. 5 c).

(4) Composite signal identification criterion for overlapping shear signals by taking tension signals as main bodies under same frequency band

And the spectrogram of the acoustic emission signal meets the conditions 1 and 2, the spectrogram of the signal meets the spectrogram characteristics of the shear signal, but the time-frequency intensity contour line only presents the contour line characteristics of the tension signal, and the signal is considered to be a superposed composite signal with tension fracture as a main part and shear as an auxiliary part (see fig. 5 d).

(5) Composite signal identification criterion for superposing tension signals by taking shear signals as main bodies under same frequency band

The spectrogram of the acoustic emission signal meets the conditions 1 and 2, the spectrogram of the signal meets the spectrogram characteristics of the shear signal, the ratio of the time-frequency intensity isoline reverse-time extension length L1 to the forward-time extension length is more than or equal to 1, but the upper contour is not flat, has obvious sharp points, is shaped like a rhombus integrally, and the signal is considered to be a superimposed composite signal with shear fracture as the main part and tension as the auxiliary part (see figure 5e)

S3, sequentially collecting signals (the acoustic emission signals are shown in figure 6) through the acoustic emission monitoring sensor of each channel in the acoustic emission system, inputting the signals into the acoustic spectrum analysis model, and calculating to obtain an acoustic spectrum characteristic diagram (figure 7) and a time-frequency intensity distribution characteristic diagram (figure 8).

And S4, according to the constructed signal classification discrimination standard, integrating the acoustic spectrum characteristic and the time-frequency intensity characteristic corresponding to the rock fracture event to finish rock fracture classification.

The following describes in detail the analysis process and effect of the rock fracture pattern classification and identification method based on the voiceprint identification technology, with a specific experimental example.

The rock test piece shown in fig. 2 is loaded, and during the rock loading and breaking process, an acoustic emission monitoring system is used for monitoring and acquiring acoustic emission waveform data (shown as fig. 6).

The acquired acoustic emission waveform data (here, the waveform data of fig. 6 is taken as an example) is subjected to acoustic spectrum analysis, and a spectrogram (fig. 7) and a time-frequency intensity distribution map (fig. 8) are respectively acquired.

The highlight points are searched in fig. 7, so that a plurality of highlight points 1, 2, 3, 4, and 5 shown in fig. 9 can be obtained, where, taking the highlight point No. 3 as an example, the feature area corresponding to the highlight point No. 3 is obtained at the same position of the time-frequency intensity distribution diagram in fig. 8, as shown in fig. 10. Referring to the classification features and the identification criteria of fig. 5, it can be found that the point 3 corresponds to a tension fracture, and similarly, fracture patterns corresponding to other highlight points can be analyzed.

It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.