阅读说明：本技术 划分岩爆过程中张拉和剪切裂纹的方法 (Method for stretching and shearing cracks in divided rock burst process ) 是由 刘冬桥 孙杰 刘赫赫 张露露 王则丽 于 2021-10-29 设计创作，主要内容包括：本发明提供了一种划分岩爆过程中张拉和剪切裂纹的方法包括通过声发射实验获得的声发射信号计算RA和AF值；划分低RA高AF区,并将其作为张拉裂纹聚集区；划分高RA低AF区,并将其作为剪切裂纹聚集区；将张拉裂纹聚集区和剪切裂纹聚集区的声发射信号作为学习值,输入向量机进行学习,得出初始裂纹信号识别库；以及采用逼近法将剩余的待识别信号分类成张拉裂纹和剪切裂纹的声发射信号,得出最终裂纹信号识别库。该方法能够有效的划分出张拉裂纹和剪切裂纹。(The invention provides a method for dividing tension and shear cracks in a rock burst process, which comprises the steps of calculating RA and AF values through acoustic emission signals obtained by an acoustic emission experiment; dividing a low RA high AF area, and taking the low RA high AF area as a tension crack aggregation area; dividing a high RA low AF area, and taking the area as a shear crack aggregation area; acoustic emission signals of the tension crack gathering area and the shear crack gathering area are used as learning values and input into a vector machine for learning, and an initial crack signal identification library is obtained; and classifying the remaining signals to be identified into acoustic emission signals of the tension cracks and the shear cracks by adopting an approximation method to obtain a final crack signal identification library. The method can effectively divide the tension cracks and the shear cracks.)

1. A method for dividing tension and shear cracks in a rock burst process is characterized by comprising the following steps:

calculating RA and AF values through acoustic emission signals obtained by an acoustic emission experiment;

dividing a low RA high AF area, and taking the low RA high AF area as a tension crack aggregation area;

dividing a high RA low AF area, and taking the area as a shear crack aggregation area;

acoustic emission signals of the tension crack gathering area and the shear crack gathering area are used as learning values and input into a vector machine for learning, and an initial crack signal identification library is obtained; and

and classifying the residual signals to be identified into acoustic emission signals of the tension cracks and the shear cracks by adopting an approximation method to obtain a final crack signal identification library.

2. The method of compartmentalizing tension and shear cracks during a rock burst of claim 1, further comprising: and drawing an RA-AF (random-echo) graph according to the acoustic emission signals of the tension crack accumulation area and the shear crack accumulation area.

3. The method for dividing tension and shear cracks in the rock burst process according to claim 1, wherein rock burst experiment sampling rock is red sandstone, and rock burst is experimentally simulated by adopting a loading mode of loading-single-side sudden unloading-load retention-axial loading.

4. A method of demarcation for tensile and shear cracks during a rock burst as claimed in claim 1 wherein RA is the ratio of rise time to amplitude and AF is the ratio of ring count to duration, wherein ring count is the number of ring pulses of an acoustic emission signal exceeding a threshold in a test, amplitude is the maximum amplitude of the waveform, duration is the time interval taken for the acoustic emission signal to first cross the threshold voltage and finally fall to the threshold voltage, and rise time is the time interval taken for the acoustic emission signal to first cross the threshold voltage and reach maximum amplitude.

5. A method of demarcation for stretch-break and shear-break cracks during a rock burst as claimed in claim 1 wherein RA is the X axis and AF is the Y axis, the zone of stretch-break accumulation being determined by brazilian cleavage experiments.

6. The method for dividing tension and shear cracks in the rock burst process according to claim 5, wherein the value range of the tension crack gathering area is an area with a slope of 73.3008-90 degrees after RA-AF coordinate normalization.

7. A method of compartmentalizing tension and shear cracking during a rock burst as recited by claim 1 wherein RA is X axis and AF is Y axis and shear crack focal zone is determined by direct shear experiments.

8. The method for dividing tension and shear cracks in a rock burst process according to claim 7, wherein the shear crack accumulation area is an area with a slope of 0-20.5560 degrees after RA-AF coordinate normalization.

9. The method for dividing tension and shear cracks in the rock burst process according to claim 1, wherein the acoustic emission signals of the shear crack gathering area are set to be 1, the acoustic emission signals of the tension crack gathering area are set to be 2, the acoustic emission signals of 1 and 2 types are taken as learning values and are brought into a support vector machine for machine learning, and an initial crack signal recognition library is obtained.

10. The method for dividing tension and shear cracks in a rock burst process according to claim 9, wherein the residual signal is set as a signal to be identified 3, and 3 is classified into 1 or 2 by an approximation method to form a final crack signal identification library.

Technical Field

The invention relates to the field of tunnel construction, in particular to a method for stretching and shearing cracks in a divided rock burst process.

Background

With the rapid development of national economy, many deep projects under construction and to be built have large scale, high difficulty and large quantity, and are in the forefront of the world. With the increasing depth of underground engineering, rock burst disasters are increasing under the influence of complex geomechanical conditions of deep three-high-one disturbance (high ground stress, high ground temperature, high osmotic pressure and strong mining and excavation disturbance), and the safe and efficient development of deep resources and the reasonable and effective utilization of underground space in China are seriously threatened.

The rock burst has strong harmfulness, complex generation mechanism and influencing factors, and is difficult to predict before rock destruction, which is always a hot problem for rock destruction research. Through deep understanding of the rock burst, people find that the rock burst undergoes the processes of splitting into plates → cutting into blocks → ejecting rock blocks before occurring; splitting into sheets can be considered as an important prerequisite for the occurrence of rock bursts. The occurrence of splitting into sheets is necessarily due to a series of tensile cracks that develop each other, through-linking. Therefore, accurate division of tensile and shear cracks plays a very critical role in the pre-failure prediction of rock burst.

In traditional rock mechanics, the division of the tension cracks and the shear cracks is based on an empirical method, the shear cracks exist in the rock tension experiment process, the tension cracks exist in the shear experiment process, and due to the problems of inhomogeneity and anisotropy of rock materials, the traditional division method has large manual intervention and inaccurate division.

Formation-propagation-aggregation of rock cracks, producing macrocracks that lead to destabilization of the final rock fracture. Acoustic emission signals can be generated in the rock destruction process, parameters and waveform characteristics can be collected through the monitoring device, and the destruction process is quantified according to the collection results. The indexes are quantified through tension and shear experiments, an effective means is provided for later-stage experiments, a basis is provided for judging the type of the rock burst crack, and the method has important theoretical and practical values.

Disclosure of Invention

The invention aims to provide a method for dividing tension cracks and shear cracks in a rock burst process, which can effectively divide the tension cracks and the shear cracks.

In order to realize the purpose of the invention, the invention adopts the following technical scheme:

according to one aspect of the invention, a method for dividing tension and shear cracks in a rock burst process is provided, which comprises the steps of calculating RA and AF values through acoustic emission signals obtained by an acoustic emission experiment; dividing a low RA high AF area, and taking the low RA high AF area as a tension crack aggregation area; dividing a high RA low AF area, and taking the area as a shear crack aggregation area; acoustic emission signals of the tension crack gathering area and the shear crack gathering area are used as learning values and input into a vector machine for learning, and an initial crack signal identification library is obtained; and classifying the remaining signals to be identified into acoustic emission signals of a tension crack gathering area and a shear crack gathering area by adopting an approximation method to obtain a final crack signal identification library.

According to an embodiment of the invention, the method for dividing tension and shear cracks in a rock burst process further comprises the step of drawing an RA-AF map according to acoustic emission signals of tension crack accumulation areas and shear crack accumulation areas.

According to an embodiment of the invention, the acoustic emission experiment sampling rock is red sandstone, and rock burst is simulated through an experiment in a loading mode of loading, single-sided sudden unloading, load holding and axial loading.

According to an embodiment of the present invention, RA is a ratio of a rise time to an amplitude, and AF is a ratio of a ringing count to a duration, wherein the ringing count refers to a number of ringing pulses of the acoustic emission signal exceeding a threshold value in a test, the amplitude refers to a maximum amplitude value of the waveform, the duration refers to a time interval from when the acoustic emission signal first crosses a threshold voltage to when the acoustic emission signal finally falls to the threshold voltage, and the rise time refers to a time interval from when the acoustic emission signal first crosses the threshold voltage to when the acoustic emission signal maximum amplitude.

According to an embodiment of the present invention, wherein RA is X axis and AF is Y axis, the tension crack accumulation zone is determined by brazilian cleavage experiment.

According to an embodiment of the invention, the value range of the tension crack accumulation area is an area with a slope of 73.3008-90 degrees after the RA-AF coordinate normalization.

According to an embodiment of the present invention, wherein RA is X-axis and AF is Y-axis, the shear crack accumulation zone is determined by direct shear experiments.

According to an embodiment of the invention, the shear crack accumulation area is an area with a slope of 0 ° to 20.5560 ° after RA-AF coordinate normalization.

According to one embodiment of the invention, the acoustic emission signals of the shear crack accumulation area are set to be 1, the acoustic emission signals of the tension crack accumulation area are set to be 2, and the acoustic emission signals of types 1 and 2 are taken as learning values and are brought into a support vector machine for machine learning to obtain an initial crack signal recognition library.

According to an embodiment of the invention, the residual signal is set as the signal 3 to be identified, and 3 is classified as 1 or 2 by approximation, so as to form a final crack signal identification library.

One embodiment of the present invention has the following advantages or benefits:

the method for dividing the tension cracks and the shear cracks in the rockburst process is a classification method, and the division method for obtaining the tension cracks and the shear cracks is obtained by calculating RA and AF values of acoustic emission signals acquired in the experimental process. The support vector machine is used as a machine learning algorithm with supervised learning, and has good advantages in dividing nonlinear data. In the RA-AF chart, the low RA high AF area is a tension crack accumulation area, and the high RA low AF area is a shear crack accumulation area. Based on the method, the two are respectively taken as a training value and a target function of machine learning to carry out machine learning, and the tension crack and the shear crack can be accurately divided by utilizing an approximation method. The tension-shear signal dividing method based on the support vector machine can effectively divide tension cracks and shear cracks.

Drawings

The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

FIG. 1 is a flow chart illustrating a method of compartmentalizing tension and shear cracks during a rock burst, according to an exemplary embodiment.

FIG. 2 is a flowchart illustrating an algorithm for supporting vector machine based crack segmentation, according to an exemplary embodiment.

FIG. 3 is a schematic diagram illustrating a crack segmentation method according to an exemplary embodiment.

FIG. 4 is a schematic diagram illustrating an acoustic emission experimental stress path according to an exemplary embodiment.

FIG. 5 is a diagram illustrating RA-AF partitions obtained after computation by a machine learning algorithm, according to an exemplary embodiment.

FIG. 6 is a graph illustrating crack fraction per minute during a rock burst according to an exemplary embodiment.

FIG. 7 is a graph illustrating the overall process fraction per minute of a rock burst process according to an exemplary embodiment.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and thus their detailed description will be omitted.

The terms "a," "an," "the," "said" are used to indicate the presence of one or more elements/components/etc.; the terms "comprising" and "having" are intended to be inclusive and mean that there may be additional elements/components/etc. other than the listed elements/components/etc.

As shown in fig. 1 to 3, fig. 1 shows a flow chart of a method for dividing tension and shear cracks in a rock burst process provided by the invention. FIG. 2 shows a flow chart of a support vector machine algorithm provided by the present invention. Fig. 3 shows a schematic diagram of a RA-AF rendering method provided by the present invention.

The method for dividing the tension and shear cracks in the rock burst process comprises the following steps: s1, calculating RA and AF values through acoustic emission signals obtained by the acoustic emission experiment; s2, dividing a low RA high AF area, and taking the low RA high AF area as a tension crack aggregation area; s3, dividing a high RA low AF area, and taking the area as a shear crack aggregation area; s4, inputting acoustic emission signals of the tension crack gathering area and the shear crack gathering area as learning values into a vector machine for learning to obtain an initial crack signal identification library; s5, classifying the remaining signals to be identified into acoustic emission signals of a tension crack accumulation area and a shear crack accumulation area by adopting an approximation method, and obtaining a final crack signal identification library.

Among them, the phenomenon that a local source in a material rapidly releases energy to generate transient elastic waves is called Acoustic Emission (AE), and sometimes called stress wave Emission. The internal stress of the material is suddenly redistributed due to the change of the internal structure of the material; converting the mechanical energy into acoustic energy; generating elastic waves is a common nondestructive testing means. The acoustic emission experiment is a conventional laboratory experiment for simulating rock burst. The SVM refers to a support vector machine as a machine learning algorithm with supervised learning. Approximation is a mathematically common method.

In a preferred embodiment of the present invention, the method of compartmentalizing tension and shear fractures during a rock burst further comprises: and drawing an RA-AF (random-echo) graph according to the acoustic emission signals of the tension crack accumulation area and the shear crack accumulation area.

After the rock burst experiment, a tension crack gathering area and a shear crack gathering area are divided, RA serves as an X axis, AF serves as a Y axis, and an RA-AF diagram is drawn.

In a preferred embodiment of the present invention, RA is a ratio of rise time to amplitude, and AF is a ratio of a ringing count to a duration, wherein the ringing count is the number of ringing pulses of an acoustic emission signal exceeding a threshold value in a test, the amplitude is a maximum amplitude value of a waveform, the duration is a time interval taken for the acoustic emission signal to first cross a threshold voltage and finally fall to the threshold voltage, and the rise time is a time interval taken for the acoustic emission signal to first cross the threshold voltage and finally fall to the maximum amplitude. And RA is an X axis, AF is a Y axis, and a tension crack accumulation area is determined through Brazilian splitting experiments. The value range of the tension crack gathering area is an area with the slope of 73.3008-90 degrees after the RA-AF coordinate normalization. And RA is an X axis, AF is a Y axis, and a shear crack accumulation area is determined through a direct shear experiment. The shear crack aggregation zone is a region with the slope of 0-20.5560 degrees after the RA-AF coordinates are normalized.

As shown in fig. 1 to 3, the RA and AF values are calculated through indoor brazilian splitting experiments, and the traditional rock mechanics method considers that most of signals from the brazilian splitting experiments are tension signals, and the signals of the tension signals are very dense in the low RA and AF areas, so that the areas can be considered as tension crack accumulation areas, and similarly, shear crack accumulation areas can be found through indoor direct shear experiments (the signals from the direct shear experiments are mostly shear crack signals, and are densely concentrated in the high RA and low AF areas). A straight line is drawn from a coordinate origin by a mathematical method, and the straight line has a very small included angle with an X axis, namely an RA axis, and is a high RA low AF area, namely a shear crack gathering area, and has a very small included angle with a Y axis, namely an AF axis, and is a high AF low RA area, namely a tension crack gathering area. The specific value of the included angle between the X axis and the Y axis is obtained through Brazilian splitting experiments and direct shear experiments and processing. After the RA-AF coordinates are normalized, the range of a shear crack gathering area is 0-20.5560 degrees, and the range of a tension crack gathering area is 73.3008-90 degrees.

In a preferred embodiment of the invention, the acoustic emission signals of the shear crack gathering area are set to be 1, the acoustic emission signals of the tension crack gathering area are set to be 2, and the acoustic emission signals of types 1 and 2 are taken as learning values and are brought into a support vector machine for machine learning to obtain an initial crack signal recognition library. Setting the residual signal as a signal 3 to be identified, and classifying the signal 3 into 1 or 2 by an approximation method to form a final crack signal identification library.

As shown in fig. 1 to 3, by the support vector machine algorithm, firstly defining the acoustic emission signal of the shear crack accumulation area as 1, the acoustic emission signal of the tension crack accumulation area as 2, and the rest is the signal to be identified as 3. And secondly, respectively taking the 1 and 2 acoustic emission signals as learning values, and taking the corresponding {1} and {2} as learning results to be brought into a support vector machine for machine learning to obtain an initial crack signal identification library. The signal to be identified is then classified into classes 1 and 2 using "approximation".

FIG. 4 shows a schematic diagram of an acoustic emission experimental stress path provided by the present invention. Fig. 5 shows a schematic diagram of an RA-AF partition obtained after calculation by a machine learning algorithm according to the present invention. Fig. 6 shows a schematic diagram of crack ratio per minute during a rock burst process provided by the present invention. Fig. 7 shows a schematic diagram of the total process ratio per minute in the process of rock burst according to the present invention.

As shown in fig. 4 to 7, in a preferred embodiment of the present invention, the experimental rock is red sandstone, and the rock burst is simulated by a loading mode of "loading-single-sided sudden unloading-holding-axial loading", where the initial ground stress level of the loading is: σ v/σ H =26.36/33.63/21.83 MPa. The stress path of the experiment is shown in fig. 4. And performing RA and AF calculation according to the obtained acoustic emission signals, and performing stretch-shear crack division on the whole rockburst process by using the machine learning algorithm. Fig. 5 shows the RA-AF partition obtained after calculation by a machine learning algorithm. Fig. 6 and 7 show the crack fraction per minute and the total process fraction during a rock burst, respectively. As can be seen from FIGS. 6 and 7, the proportion of the tension cracks is the majority in the rock burst experiment process, and the tension cracks are dominant in the whole rock burst process. According to the existing theoretical research, the rock body must have a process of splitting into plates before rock burst occurs, and the splitting into plates is caused by the fact that the tension cracks play a leading role in the whole process. In the process of loading to the rock burst, the phenomenon that the proportion of tension cracks is reduced and the proportion of shear cracks is increased occurs, which corresponds to the process of 'shearing into blocks' in the process of rock burst inoculation, and the proportion of shear cracks is continuously increased along with the continuous reduction of the proportion of tension cracks, and finally the rock burst occurs.

The result calculated by the method is consistent with the actual situation, which shows that the dividing result of the method accords with the actual situation and has rationality. Compared with the traditional method, the method greatly reduces manual intervention by utilizing a machine learning algorithm, and the calculated crack proportion curve well reflects the processes of splitting into plates → shearing into blocks → ejecting the rock blocks in the rock burst inoculation process.

In embodiments of the present invention, the term "plurality" means two or more unless explicitly defined otherwise. The terms "mounted," "connected," "secured," and the like are to be construed broadly, and for example, "connected" may be a fixed connection, a removable connection, or an integral connection. Specific meanings of the above terms in the embodiments of the present invention can be understood by those of ordinary skill in the art according to specific situations.

In the description of the embodiments of the present invention, it should be understood that the terms "upper", "lower", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience in describing the embodiments of the present invention and simplifying the description, but do not indicate or imply that the referred devices or units must have a specific direction, be configured in a specific orientation, and operate, and thus, should not be construed as limiting the embodiments of the present invention.

In the description herein, the appearances of the phrase "one embodiment," "a preferred embodiment," or the like, are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the embodiments of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the present embodiment by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention.