CN-116858850-B - Dynamic identification and real-time prediction method for coal and rock fracture track

CN116858850BCN 116858850 BCN116858850 BCN 116858850BCN-116858850-B

Abstract

The invention provides a dynamic identification and real-time prediction method for a coal rock fracture track. The method establishes a coal rock internal and external crack development track recognition model, tracks the coal rock internal and external crack development track in real time, and realizes the dynamic recognition of the crack development track of the coal rock internal and external crack combination. The method has the advantages that a real-time self-correction prediction model of the internal and external cracks of the coal rock is built, the prediction result of the development track of the internal and external cracks of the coal rock is subjected to real-time self-correction, and the dynamic forward prediction of the development path of the internal and external cracks of the coal rock is realized. The coal rock crack identification prediction model constructed by the method can master the development condition of coal rock cracks in real time, and the self-correction prediction model has the characteristics of wide applicability, high reliability and high precision, and is suitable for the development of coal rock cracks in different prediction scenes. Based on the real-time identification and self-correction prediction model of the internal and external cracks of the coal rock, the method is favorable for timely defining and predicting the stability of the coal rock.

Inventors

ZOU QUANLE
CHEN CHUNMEI
CHEN ZIHAN
ZHAN JINFEI
Lv Feixiang
LIU HAN
Ran Qican

Assignees

重庆大学

Dates

Publication Date: 20260512
Application Date: 20230619

Claims (8)

1. The dynamic identification and real-time prediction method for the coal and rock fracture track is characterized by comprising the following steps of: 1) N acoustic emission probes are uniformly arranged on the side wall of the coal rock sample, and a high-speed camera is arranged on the outer side of the coal rock sample; 2) Dividing the acquisition period T into a plurality of acquisition steps according to the time step delta T, collecting the internal fracture expansion process of the coal rock and the external fracture expansion track condition of the coal rock in the acquisition period T by adopting an acoustic emission device and a high-speed camera in the mechanical experiment process, and obtaining internal fracture coordinates at the moments of T, t+1, t+2, t+3 and the term (t)= Obtaining the external fracture coordinates at the time points of t, t+1, t+2, t+3 and the time points of the third and the fourth, so as to establish an external fracture tip coordinate change data set W (t) =with time sequence characteristics The acquisition and calculation process of the single time step comprises the following sub-steps: a) The method comprises the steps of collecting acoustic emission information in a coal rock cracking process by adopting acoustic emission equipment, identifying the development track of the internal cracks of the coal rock at the moment t, recording the acoustic emission information in the whole process of the development of the internal cracks of the coal rock by adopting acoustic emission technology, further carrying out the identification of the internal cracks of the coal rock by adopting a sound source positioning method, reversing the development track of the cracks on the basis of acoustic emission positioning technology, and determining the spatial position of the development of the internal cracks of a coal rock sample; b) Recording a series of external crack development processes by adopting a high-speed camera, extracting a crack track image, identifying the external crack development track of the coal rock at the moment t, identifying the external crack of the coal rock by using an intelligent image identification algorithm, performing binarization treatment on the extracted coal rock crack image, separating cracks in the image from a background, and further extracting track coordinates in the crack development process to obtain data of the external crack position changing along with time; c) Based on the internal coordinate and the external coordinate of the t moment extracted in the step a) and the step b), matching the points of the internal and the external cracks by utilizing a point position corresponding method, so as to realize the real-time fusion of the internal and the external cracks; 3) Learning the internal and external fracture tip coordinate change data set extracted in the step 2) as a historical data set to obtain a fracture tip coordinate prediction model, predicting the internal and external fracture development tracks by adopting the fracture tip coordinate prediction model, predicting fracture tip coordinates at the future moment, realizing the prediction of the dynamic extension process of the internal and external fracture at the next moment, and outputting the coordinate prediction value of the internal and external fracture development dynamic process And ; 4) Continuously correcting the prediction step length according to the prediction result of the internal and external fracture prediction model, and updating the model in real time to realize the forward-delay self-correcting prediction of the development track of the internal and external fracture of the coal rock; 5) The method comprises the steps of carrying out identification prediction on internal and external fracture development tracks, carrying out matching on internal and external fracture points by utilizing a point position corresponding method based on the coal rock internal and external fracture coordinate prediction values obtained in the step 3) and the step 4), realizing internal and external fracture point position fusion based on a time sequence, and then obtaining a predicted forward fracture track.
2. The method for dynamically identifying and predicting the fracture track of the coal rock according to claim 1, wherein in the step 1), the mechanical experiment is a tensile failure mechanical experiment.
3. The method for dynamically identifying and predicting the fracture track of the coal rock according to claim 1, wherein n is more than or equal to 4 in the step 1).
4. The method for dynamically identifying and predicting the fracture trajectory of coal and rock according to claim 1, wherein the step a) comprises the following steps: a1 Establishing a space rectangular coordinate system by taking a central point of a coal rock sample as an origin to obtain space coordinates (x n ,y n ,z n ) of n sensors, marking a space seat corresponding to a crack development path as (x i ,y i ,z i ), marking the time of sound source signals received by the n sensors as t n , and marking the time of the signals transmitted from the sound emission sources as t; (1) Solving the solution (1) to obtain a plurality of groups of coordinate points, and solving the geometric centers of the plurality of groups of coordinate points, wherein the geometric centers are the space coordinates (x i ,y i ,z i ) corresponding to the required crack development path; a2 Extracting internal fracture coordinate changes in the fracture development process, and establishing a space coordinate data set ={(x 1 ,y 1 ,z 1 ) (x 2 ,y 2 ,z 2 ) (x 3 ,y 3 ,z 3 )……(x n ,y n ,z n )}; A3 By updating the space coordinates of the coal rock fracture tip at the current moment in real time, an internal fracture tip coordinate change data set with time sequence characteristics is established (t)= 。
5. The method for dynamically identifying and predicting the fracture trajectory of coal and rock according to claim 1, wherein the step b) comprises the following steps: b1 Recording a track image of external crack development, wherein a coal rock test piece center point is selected as a coordinate origin, and a plane coordinate system is drawn; b2 Performing binarization treatment on the image of the coal rock external fracture development process obtained in the step b 1); b3 Based on the slit image after binarization treatment in the step b 2), scanning line by line, and traversing the pixel value corresponding to any point (x, y) in the slit image in sequence And stored in set W i ; wherein i is 0,1, 2..n; (2) the tip of the slit has a change in pixel value, the first threshold change point being defined as the slit tip point, i.e. the The corresponding (x, y) is noted as the fracture tip coordinates; b4 Repeating the tip searching step in b 3) to obtain a series of processes of the development of the external fracture of the coal rock, further obtaining the coordinate change of the track in the process of the fracture development, and recording as Ω={(x 1 ,y 1 ) (x 2 ,y 2 ) (x 3 ,y 3 )……(x n ,y n )}; B5 By updating the coal rock external fracture tip coordinates in real time, an external fracture tip coordinate change data set with time sequence characteristics is established and is marked as W (t) = 。
6. The method for dynamically identifying and predicting the fracture track of the coal rock according to claim 1 is characterized by comprising the steps of c) combining internal coordinates reflecting the fracture morphology and size information with external coordinates reflecting the displacement and deformation information of the fracture, combining the external coordinates corresponding to a certain moment t with the corresponding internal fracture coordinates, further inverting the fracture development track of the coal rock test piece at the moment t, matching the internal fracture point position with the external fracture point position by utilizing a point position corresponding method based on the extracted internal coordinates and the external coordinates at the moment t, realizing the real-time fusion of the internal fracture and the external fracture, and forming a three-dimensional fracture model according to the fusion result, so that the positions and the morphologies of the internal fracture and the external fracture can be identified more accurately.
7. The method for dynamically identifying and predicting the fracture track of the coal rock according to claim 1, wherein in the step 3), training is performed by using the extracted data set of the tip coordinates of the inner and outer cracks of the coal rock so as to obtain a crack development track prediction model, and the tip coordinates of the inner and outer cracks of the coal rock at a future moment are predicted by using the model so as to predict the crack development track of the inner and outer cracks of the coal rock, wherein the prediction model comprises a long and short time memory network LSTM.
8. The method for dynamically identifying and predicting the fracture track of the coal rock according to claim 1, wherein in the step 4), the real-time self-correction method is used for collecting and monitoring the fracture development data in real time and establishing a corresponding feedback mechanism, and in the implementation process, a corresponding error range threshold value is set for a prediction model based on research precision, and when the model prediction value exceeds the error threshold value, self-correction adjustment is needed for the model, so that the prediction precision is improved.

Description

Dynamic identification and real-time prediction method for coal and rock fracture track Technical Field The invention relates to the technical field of rock mechanics, in particular to a dynamic identification and real-time prediction method for a coal rock fracture track. Background In underground engineering such as tunnel construction and coal mining, geotechnical engineering disasters caused by expansion and penetration of rock mass cracks occur, and the destabilization damage of geotechnical engineering is mostly related to the expansion and penetration of cracks, and is mainly represented by the phenomena of initiation, expansion and penetration of rock cracks, and further causes macroscopic cracking. In this regard, students studied the crack propagation and deformation process of the coal rock sample under load by using a digital image correlation technique observation system, analyzed the initiation, propagation and interconnected evolution process of the external cracks of the coal rock, and although many developments in the study of rock cracks have been advanced, the current study is mostly focused on the identification and prediction of the internal or external crack development process of a single coal rock, and the simultaneous observation of internal and external cracks is not realized. The occurrence of the internal and external cracks is often influenced mutually, so that the two are required to be combined for consideration in actual engineering, meanwhile, the development process of the prefabricated cracks in the rock test piece under the action of the observation load is found to have the difference in the development of the internal and external cracks, and the development process of the internal and external cracks can be effectively combined in few researches at present. In addition, in the face of large-scale real-time data, the implementation of real-time prediction still faces many challenges, and in addition, different working conditions or engineering conditions may have different accuracy requirements for model prediction, however, an existing prediction model can usually only predict for a specific working condition, and an automatic correction mechanism is lacking, which means that the accuracy and the accuracy of the prediction model in different scenes are limited, and the prediction result often has difficulty in meeting the actual engineering needs. In order to address these challenges, it is desirable to propose more accurate and stable recognition of predictive models, and further to modify and optimize the predictive models for different operating conditions. Therefore, a dynamic identification method combining the development tracks of the internal and external cracks of the coal rock is needed. Disclosure of Invention The invention aims to provide a dynamic identification and real-time prediction method for a coal and rock fracture track, which aims to solve the problems in the prior art. The technical scheme adopted for realizing the purpose of the invention is that the method for dynamically identifying and predicting the fracture track of the coal rock in real time comprises the following steps: 1) And (5) meshing the surface of the sample. N acoustic emission probes are uniformly arranged on the side wall of the coal rock sample, and a high-speed camera is arranged on the outer side of the coal rock sample. 2) The acquisition period T is divided into a plurality of acquisition steps according to the time step Δt. In the mechanical experiment process, an acoustic emission device and a high-speed camera are adopted to collect the coal rock internal crack expansion process and the coal rock external crack expansion track condition in the acquisition period T. Obtaining internal fracture coordinates at times t, t+1, t+2, t+3, and..once.n. t+n, thereby establishing an internal fracture tip coordinate change dataset with time sequence characteristics(t)=. Obtaining the external slit coordinates at times t, t+1, t+2, t+3, and..once.t+n, thereby establishing an external slit tip coordinate variation data set W (t) =with time sequence characteristics. The acquisition and calculation process of the single time step comprises the following sub-steps: a) And collecting acoustic emission information in the coal rock cracking process by adopting acoustic emission equipment. And identifying the development track of the internal fissure of the coal rock at the moment t. And recording acoustic emission information of the whole development process of the internal fissures of the coal rock by adopting an acoustic emission technology, further carrying out identification of the internal fissures of the coal rock by adopting a sound source positioning method, and determining the spatial position of the internal fissure development of the coal rock sample based on the acoustic emission positioning technology. B) And recording a series of processes of external crack development by adop