CN115791803B - Deep-buried tunnel surrounding rock blasting damage test system and test method - Google Patents
Deep-buried tunnel surrounding rock blasting damage test system and test method Download PDFInfo
- Publication number
- CN115791803B CN115791803B CN202211615303.5A CN202211615303A CN115791803B CN 115791803 B CN115791803 B CN 115791803B CN 202211615303 A CN202211615303 A CN 202211615303A CN 115791803 B CN115791803 B CN 115791803B
- Authority
- CN
- China
- Prior art keywords
- surrounding rock
- damage
- coefficient
- blasting
- rock
- 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 215
- 230000006378 damage Effects 0.000 title claims abstract description 177
- 238000005422 blasting Methods 0.000 title claims abstract description 77
- 238000012360 testing method Methods 0.000 title claims abstract description 39
- 238000010998 test method Methods 0.000 title claims abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000011156 evaluation Methods 0.000 claims description 30
- 238000001514 detection method Methods 0.000 claims description 28
- 238000012937 correction Methods 0.000 claims description 27
- 230000008859 change Effects 0.000 claims description 17
- 238000005553 drilling Methods 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000001965 increasing effect Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 8
- 238000010586 diagram Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 6
- 238000003672 processing method Methods 0.000 claims description 3
- 239000000779 smoke Substances 0.000 claims description 3
- 230000007847 structural defect Effects 0.000 claims description 3
- 230000001681 protective effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 9
- 238000010276 construction Methods 0.000 description 7
- 238000011176 pooling Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 6
- 230000002829 reductive effect Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 239000000523 sample Substances 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000002427 irreversible effect Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000000149 penetrating effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000036039 immunity Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000005641 tunneling Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Abstract
The invention relates to a surrounding rock blasting damage test system and a test method for a deep buried tunnel. The method improves the accuracy of the surrounding rock blasting damage test, and can rapidly judge the damage state of the surrounding rock, so that a series of protective measures are taken for the surrounding rock, and the stability of the tunnel is ensured.
Description
Technical Field
The invention belongs to the field of tunnel engineering, and particularly relates to a deep-buried tunnel surrounding rock blasting damage test system and a test method.
Background
In recent years, with the increasing demands of society on transportation, tunnel engineering projects are increased increasingly, and the tunnel engineering projects tend to be deeply buried and grow in the growing direction. However, tunnel engineering is complex, the technical aspect is imperfect, and the process has stronger uncertainty, wherein the construction safety is inevitably threatened by structural deformation in the deep buried tunnel. In the process of tunnel blasting, uncontrollable factors are more, life or property loss is easy to occur, and the method for reducing influencing factors is lacking, so that damage to surrounding rock in the blasting process is reduced to an ideal standard.
The damage forms encountered by the deep-buried tunnel in the construction process are different, but the tunnel periphery deformation caused by the stress redistribution of surrounding rock is substantial, the deep-buried tunnel can overcome the terrain obstacle, shorten the distance between two places, enable the line to be smoother, and improve the safety quality of transportation and traffic, so that the state of the surrounding rock in the deep-buried tunnel needs to be accurately monitored.
At present, a plurality of approaches can observe the state of surrounding rock, but accuracy and convincing are generally lacking, guidance is provided for construction by means of scientific technology, and the method is a great trend of future tunnel engineering and is developed towards the intelligent direction. The surrounding rock is continuously deteriorated in mechanical property under the continuous blasting effect, and the damage is continuously accumulated to cause the loss of a stable state, so that the phenomenon of tunnel collapse occurs. In general, the damage state of surrounding rock after blasting is observed, the internal structure of the surrounding rock can be observed through drilling, namely, the surrounding rock is perforated at a certain position, and then the data are detected and recorded by utilizing a camera shooting technology; the damage state in the surrounding rock can be detected by an ultrasonic method, namely the elastic wave velocity and the attenuation condition of the rock mass can be effectively measured after the sound wave penetrates through the surrounding rock, so that the damage and destruction condition of the surrounding rock can be judged; or taking the quality of the surrounding rock as a reference, such as the joint crack density of the surrounding rock, the quality grade of the surrounding rock and the like, evaluating the blasting damage of the surrounding rock or correcting the test result. The existing damage method for testing surrounding rock blasting is less, the deviation from the actual situation is larger, the problems of tunnel collapse or overlarge explosive consumption and the like caused by misjudgment are easy to occur, the life safety of constructors or property loss are extremely easy to threaten, meanwhile, the difficulty of construction is also greatly increased, and the construction progress is influenced.
Disclosure of Invention
The invention provides a surrounding rock blasting damage test system and a surrounding rock blasting damage test method for a deep-buried tunnel, which aim to solve the defects of the background technology, improve the accuracy of the surrounding rock blasting damage test, and rapidly judge the damage state of the surrounding rock, so that a series of protective measures are adopted for the surrounding rock, and the stability of the tunnel is ensured.
In order to achieve the purpose, the invention provides a method for testing a blasting damage system of surrounding rocks of a deep-buried tunnel, which comprises the following specific steps and technical scheme:
The system specifically comprises a surrounding rock identification module, an ultrasonic detection module, a shooting and crack identification module and a surrounding rock blasting damage evaluation module, wherein the surrounding rock identification module, the ultrasonic detection module, the shooting and crack identification module are all connected with the surrounding rock blasting damage evaluation module;
the surrounding rock identification module is used for generating a surrounding rock point cloud image after a surrounding rock is scanned by using a convolutional neural network to identify laser radar, identifying microscopic structural defects existing on the surface of the surrounding rock and outputting a correction coefficient alpha;
the ultrasonic detection module is specifically used for acquiring surrounding rock internal information by using an acoustic wave instrument, and judging a surrounding rock macroscopic damage coefficient D 1 by using the change characteristics of acoustic waves before and after blasting;
The shooting and crack identification module is specifically a method for utilizing a digital panoramic drilling shooting technology and image processing, and judges the microscopic damage coefficient D 2 of the surrounding rock by observing the increment of the crack length and width, the increment of the crack depth and the change of the crack angle before and after blasting;
The surrounding rock damage evaluation module determines a comprehensive index damage coefficient [ D ] through a correction coefficient alpha input by the surrounding rock identification module, a macroscopic damage coefficient D 1 input by the ultrasonic detection module and a microscopic damage coefficient D 2 input by the camera shooting and crack identification module, and corrects the surrounding rock comprehensive index damage coefficient [ D ] for a plurality of times.
The method comprises the steps of (1) establishing a network model through a surrounding rock identification module according to collected tunnel face images and data, generating point cloud images through laser radar scanning tunnel face images, calculating areas of target areas in the point cloud images, and according to the ratio of areas of cracks or holes to tunnel face, setting a correction coefficient alpha and inputting the correction coefficient alpha into a surrounding rock blasting damage evaluation module;
The ultrasonic detection module acquires surrounding rock information by utilizing a sound wave test, obtains a macroscopic damage coefficient D 1 of the surrounding rock and inputs the macroscopic damage coefficient D 1 into the surrounding rock blasting damage evaluation module;
The third step, shooting and crack identification module obtains clear visible plane expansion diagram outline images by using a digital panoramic drilling shooting technology and an image processing method, processes the images, identifies cracks in the images, obtains a microscopic damage coefficient D 2 by increasing the length and width of the cracks before and after blasting, increasing the depth of the cracks and changing the angles of the cracks, and inputs the microscopic damage coefficient D 2 into the surrounding rock blasting damage evaluation module;
And step four, inputting the correction coefficient alpha, the macroscopic damage coefficient D 1 and the microscopic damage coefficient D 2 into a surrounding rock blasting damage evaluation module, determining the comprehensive index damage coefficient [ D ], and correcting the surrounding rock comprehensive index damage coefficient [ D ] for a plurality of times.
Further, the calculation formula of the macroscopic damage coefficient D 1 in the step (two) is as follows:
Wherein: d 1 is the macroscopic damage coefficient of surrounding rock obtained by the ultrasonic detection module, E 0 is the elastic modulus of the rock mass before blasting, E is the equivalent elastic modulus of the rock mass after blasting, V pm、Vpr is the elastic longitudinal wave velocity of the affected rock mass and the complete rock mass after blasting is finished, η is the sound wave reduction rate, and when η is greater than 10%, it is determined that the rock mass is damaged by blasting damage, and the corresponding smoke table damage threshold is D cr =0.19.
Further, in the step (three), the calculation formula of the mesoscopic damage coefficient D 2 is as follows:
Wherein: d 2 is a mesoscopic damage coefficient obtained by a camera shooting and fracture identification module, theta 1 is a fracture angle change quantity, l 1 is a fracture length increase quantity, b 1 is a fracture width increase quantity, h 1 is a fracture depth increase quantity, l is a fracture length, b is a fracture width, and h is a fracture depth.
Further, the fracture angle θ is calculated as follows:
Wherein: θ is the angle of the crack, g 2(y)、g1 (x) is the tangential direction of each point, Is the direction of the average tangent line.
Further, the comprehensive index damage coefficient [ D ] correction method comprises the following steps:
the initial formula: [D] =α (aD 1+bD2)
One-time allocation formula:
Scaling factor a:
scaling factor b:
The secondary allocation formula: [D] =α (a 1D1+b1D2)
N-time allocation formula: [D] =α (a n-1D1+bn-1D2)
Wherein: alpha is a correction coefficient, a is a proportionality coefficient of a macroscopic damage coefficient D 1 of surrounding rock, b is a proportionality coefficient of a microscopic damage coefficient D 2 of the surrounding rock, D 1 is a surrounding rock damage coefficient acquired by an ultrasonic detection module, D 2 is a surrounding rock damage coefficient acquired by a shooting and crack identification module, and n is correction times.
The invention has the beneficial effects that:
Through the combined work of the modules, the rock mass damage judgment is more accurate and accords with the actual situation, and the data processing is more intelligent, rapid and efficient;
by the deep learning method, rock mass details can be accurately captured, and compared with a manual mode, the efficiency is greatly improved and the safety is ensured to a certain extent;
The laser radar is used for quantitatively acquiring rock mass information instead of the traditional photographing mode, so that the disadvantage of unclear imaging caused by environmental factors is avoided;
The accurate identification of rock damage can greatly weaken the influence of larger errors of subsequent field blasting or technical schemes, and protect the loss of lives and properties;
the influence of low accuracy of manual data acquisition can be effectively overcome, the work of manual data acquisition is replaced to a great extent, and the construction efficiency is improved.
Drawings
FIG. 1 is a flow chart of a deep buried tunnel surrounding rock blasting damage test system;
FIG. 2 is a schematic diagram of a surrounding rock identification module;
FIG. 3 is a schematic flow chart of an ultrasonic detection module;
FIG. 4 is a schematic diagram of a camera and crack identification module;
FIG. 5 is a graph of image processing effects in the camera and crack recognition module;
FIG. 6 is a front view panoramic borehole television display;
fig. 7 is a plan expanded view of the structural face.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present patent is further described in detail below with reference to the accompanying drawings and detailed description. It should be understood that the detailed description is presented by way of example only and is not intended to limit the invention.
A surrounding rock blasting damage test system for a deep buried tunnel specifically comprises four parts, namely a surrounding rock identification module, an ultrasonic detection module, a shooting and crack identification module and a surrounding rock blasting damage evaluation module. The surrounding rock identification module, the ultrasonic detection module, the camera shooting and crack identification module are all connected with the surrounding rock blasting damage evaluation module through signals or electricity.
The surrounding rock identification module is used for generating a surrounding rock point cloud image after a surrounding rock is scanned by using a convolutional neural network to identify laser radar, identifying microscopic structural defects such as microcracks or micropore holes on the surface of the surrounding rock, and outputting a correction coefficient alpha. The purpose of using the laser radar is to solve the problem of dark light in the tunnel, and the trouble of unclear conventional image acquisition can be effectively avoided by using a method for generating point cloud to generate image data. Under various load actions, tiny cracks or holes are easy to expand into macroscopic cracks and holes until irreversible damage is developed, so that the rock body loses part of mechanical properties, the microscopic defects on the surrounding rock are detected through a convolutional neural network, feedback is timely given to the actual condition of the surrounding rock after blasting, reference data of the damage degree of the surrounding rock are given, a subsequent blasting operation adjustment scheme can be adopted, irreversible damage to the surrounding rock is prevented, the reference data are input into a surrounding rock blasting damage evaluation module, and the surrounding rock damage condition is further tested.
The ultrasonic detection module is specifically used for acquiring surrounding rock internal information by utilizing an RSM-SY5 intelligent sonic wave instrument, and judging a surrounding rock macroscopic damage coefficient D 1 by utilizing the change characteristics of sound waves before and after blasting. By adopting a plurality of test holes, all drilling holes in the acoustic wave test need to incline downwards by a certain angle, the inside of the test holes is prevented from being far away from the exploded rock mass, the measured damage coefficient is smaller, and all drilling holes are kept parallel. According to the sound wave test, the sound wave speed of a medium is calculated according to different propagation characteristics of sound waves in different surrounding rocks, the sound wave tester emits high-frequency elastic pulse waves, and according to the Huygens principle, when the sound waves reach a structural surface, reflection, scattering, diffraction and the like are generated, so that propagation paths are prolonged by micro cracks and macroscopic cracks, the sound wave speed is reduced, the sound wave reduction degree has close relation with the number and width of cracks, and the cracks are continuously increased and expanded along with the increase of blasting times, so that the sound wave speed is continuously reduced; the sound waves can be divided into longitudinal waves, transverse waves, surface waves and the like, the air medium and the water medium can only transmit the longitudinal waves, and in the transmission process of the sound waves, the sound waves can be influenced by various factors, such as stress states of rock mass, integrity of the rock mass and the like, and the sound waves are transmitted quickly in the hard and compact rock mass; slower propagation in a rock mass containing more fissures; when the sound wave speed is attenuated, if the sound wave attenuation is serious, the surrounding rock damage is serious, otherwise, the surrounding rock damage is slight. Therefore, accurate results can be obtained by judging the damage degree of the surrounding rock by utilizing the change characteristics of the sound waves.
The shooting and fracture identification module is specifically a method for utilizing a digital panoramic drilling shooting technology and image processing, and judges the surrounding rock microscopic damage coefficient D 2 by observing the increment of the fracture length width, the increment of the fracture depth and the change of the fracture angle before and after blasting. The camera shooting and crack identifying module comprises a hardware part and a software part, wherein the hardware part consists of a panoramic camera, a depth pulse generator, a computer, a cable and the like, the panoramic camera is main equipment, is provided with a light source, is used for illuminating and shooting the hole wall in real time, and internally comprises a reflecting mirror capable of obtaining a panoramic image, and is converted into the panoramic image after being refracted by the reflecting mirror; the depth pulse generator consists of a measuring wheel, a photoelectric rotation angle encoder, a depth signal acquisition board and an interface board, and has the main functions of determining the position of a probe and automatically detecting a system; the main functions of the software part are that image data are monitored and output in real time, images can be processed to form result images, the occurrence gap width of a structural surface can be calculated, and the like, and the detection results can be statistically analyzed and a database can be established. The digital panoramic drilling and shooting technology has a series of advantages that construction work can be completed efficiently, for example, the digital panoramic drilling and shooting technology has panoramic observation capability, real-time monitoring function, capability of storing output images in a computer and continuously playing, timely processing and analyzing hole wall images, and real-time displaying, wherein a front-view panoramic drilling television display diagram and a structural surface display diagram are shown in fig. 6 and 7.
The surrounding rock damage evaluation module is used for analyzing the rock burst damage degree on a macroscopic measure and a microscopic measure. Due to the complexity of rock damage properties, the actual condition of the rock can be determined more accurately by considering rock damage for two different measures, and the surrounding rock blasting damage evaluation module mainly analyzes data on macroscopic measures and microscopic measures through algorithms or functional relations, and proposes a correction method on the whole layer, and finally determines a module of a result. And transmitting sound waves in the test hole by the ultrasonic detection module to the macroscopic extent, judging the rock mass damage degree according to the change of the sound wave speed, and if the sound wave speed is attenuated, indicating that the surrounding rock is seriously damaged if the sound wave attenuation is serious, otherwise, slightly damaging the surrounding rock. Outputting a macroscopic damage coefficient D 1 of the surrounding rock through a functional relation; the macroscopic measurement is judged by factors such as the number and the area of cracks or holes, the image of the holes is shot and converted into a panoramic image by a shooting and crack recognition module, then the panoramic image is restored into a plane expansion chart, the plane expansion chart is recognized by a deep learning technology, the number and the area of the cracks or holes are marked, and a microscopic damage coefficient D 2 is obtained by a functional relation; and comprehensively analyzing output results of the ultrasonic detection module and the camera shooting and fracture recognition module, determining a comprehensive index damage coefficient [ D ] through a functional relation, recognizing and outputting a correction coefficient alpha according to the area of the tunnel face by the surrounding rock recognition module, correcting the comprehensive index damage coefficient [ D ], and finally determining a specific numerical value of a rock damage variable.
As shown in fig. 1, a testing method of a deep-buried tunnel surrounding rock blasting damage testing system comprises the following working procedures:
Establishing a network model by using a data set which is manufactured by labelimg software through a surrounding rock identification module, wherein the data set is a plurality of tunnel face pictures and data files containing marked cracks or holes, and the pictures are acquired through shooting and the like; the method comprises the steps of performing label description on an image through frame selection of a characteristic image, storing a label file into an animation folder of a VOC2007 folder, and storing a picture file into JPEGImages under the VOC2007 folder before training; scanning tunnel face images through a laser radar to generate point cloud images; calculating the area of a target area in the point cloud image, and according to the ratio of the area of a crack or a hole to the tunnel face, drawing a correction coefficient alpha and inputting the correction coefficient alpha into a surrounding rock blasting damage evaluation module;
As shown in fig. 2, the specific steps are as follows:
Firstly, a YOLOv network model is established, the convolutional neural network mainly comprises an input layer, a convolutional layer, a pooling layer, a full-connection layer and an output layer, the convolutional layer is a core of the convolutional neural network, each convolutional layer carries out convolutional operation by a corresponding convolutional kernel, wherein the more the number of layers of the convolutional layer is, the deeper the extracted features are, the convolutional kernels have two characteristics of local connection and weight sharing, and the weight sharing means that each convolutional kernel in the convolutional layer repeatedly acts in a receptive field to operate an image. The pooling layer reduces the complexity of the feature graphs through pooling operation mainly under the condition of large data volume, namely the number of the output feature graphs is unchanged, the size of the feature graphs is reduced, the calculation complexity can be effectively reduced, and the common pooling method comprises the maximum pooling, the random pooling and the average pooling. The fully connected layer is generally the last layer of the convolutional neural network, and mainly compresses and flattens multi-dimensional image data, thereby being beneficial to enhancing the mapping capability of the network and limiting the size of the network scale.
YOLOv5 the network structure is divided into 3 parts, backbone (Backbone part), neck and Head. The backspace function is mainly characterized by extracting features; neck acts primarily on a wave mixing and combining of the features and passing these features to the prediction layer; the Head effect is mainly to make the final prediction output.
And training a convolutional neural network, developing a network model based on Microsoft Visual Vtudio 2015, adopting Python language, and selecting a YOLOv model based on a pytorch framework. The invention uses VOC format to train, utilizes labelimg to make data set, selects characteristic image through frame, carries out label description to image, stores label file into the animation file in VOC2007 file, stores picture file into JPEGImages under VOC2007 file before train.
Processing the manufactured data set through the network model, training the network model to a standard with higher accuracy, identifying the generated training set, testing the identification accuracy of the network model, if the accuracy is lower, increasing the number of the data sets, carrying out network training again, identifying the training set, and repeating the steps until the accuracy of the output result reaches the standard and basically accords with the real situation. The result is typically output using Relu functions, relu functions being essentially functions that take a maximum value, relu functions outputting 0 if the input is negative and directly outputting the value if the input is positive; the specific formula is as follows:
f(x)＝max(0,x)(1)
Further, the point cloud image is subjected to preliminary treatment, in order to accurately and rapidly acquire the area of the crack or hole, the edges of the tiny crack or hole are treated by adopting a median filtering method, the gray value of each pixel point is set to be the median value of the gray values of all pixel points in a certain neighborhood window of the point, so that noise can be effectively eliminated, and meanwhile, the edge information of the crack or hole is well protected;
further, the area calculation is performed on the target area in the point cloud image, and the specific formula is as follows, wherein the total pixel number is calculated and then multiplied by the area of a single pixel:
S Total area of ＝S area of single pixel total number of pixels (2)
Further, according to the ratio of the area of the crack or the hole to the tunnel face, a correction coefficient alpha is formulated as follows:
Wherein: alpha is a correction coefficient, n is the number of cracks or holes, S Total area of is the area of a single crack or hole, and S Tunnel face area is the total area of the tunnel face.
And finally, inputting the drawn correction coefficient alpha into a surrounding rock blasting damage evaluation module, waiting for the output data of the ultrasonic detection module and the shooting and crack identification module, and waiting for cooperative work.
The ultrasonic detection module acquires surrounding rock information by utilizing a sound wave test, obtains a macroscopic damage coefficient D 1 of the surrounding rock and inputs the macroscopic damage coefficient D 1 into the surrounding rock blasting damage evaluation module;
as shown in fig. 3, the specific steps are as follows:
Firstly, surrounding rock near a tunneling surface is drilled, the test holes are parallel to the cross section of a tunnel, a plurality of test holes are arranged according to specific conditions, the distances among the test holes can meet actual requirements, and a certain angle is formed between the test holes and the horizontal direction;
And (3) preparing the RSM-SY5 intelligent sonic wave instrument, filling water into the test hole after the RSM-SY5 intelligent sonic wave instrument is prepared, inserting a sonic wave transmitting probe and a sonic wave receiver into the test hole, and starting sonic wave testing from the bottom of the hole and collecting the speed. The specific sonic velocity C p is calculated as follows:
Wherein: l is the distance m of the sound wave penetrating through the surrounding rock, and t is the time s of the sound wave penetrating through the rock mass.
After the underground tunnel surrounding rock is subjected to the action of blasting dynamic load, the rock mass structure can be damaged, micro cracks or holes can be generated, the damage can have great influence on subsequent engineering, stress concentration can be generated, stress redistribution is caused, the surrounding rock mass is unstable due to rock mass damage, the life safety of constructors is seriously threatened, the corresponding relation of surrounding rock damage coefficients can be obtained preliminarily through the rock mass integrity coefficient K v according to the difference of rock mass integrity coefficients K v and sound wave reduction rate eta, and the specific formulas are as follows:
Wherein: k v is the rock mass integrity coefficient, and V pm、Vpr is the elastic longitudinal wave velocity of the affected rock mass and the complete rock mass after blasting is completed.
Wherein: d 1 is the macroscopic damage coefficient of surrounding rock obtained by the ultrasonic detection module, E 0 is the elastic modulus of the rock mass before blasting, E is the equivalent elastic modulus of the rock mass after blasting, V pm、Vpr is the elastic longitudinal wave velocity of the affected rock mass and the complete rock mass after blasting is finished, η is the sound wave reduction rate, and when η is greater than 10%, it is determined that the rock mass is damaged by blasting damage, and the corresponding smoke table damage threshold is D cr =0.19.
The method comprises the steps of judging the integrity degree of surrounding rock according to the damage coefficient of the surrounding rock, crushing the surrounding rock with larger damage coefficient and complete surrounding rock with smaller damage coefficient, guiding the classification work of the surrounding rock by using the damage coefficient of the surrounding rock, providing a certain reference, inputting the macroscopic damage coefficient D 1 of the surrounding rock into a surrounding rock blasting damage evaluation module, waiting for the shooting and crack recognition module to output data, and waiting for cooperative work.
The third step, shooting and crack recognition module obtains clear visible plane expansion map outline images by using a digital panoramic drilling shooting technology and an image processing method and processes the images; identifying cracks in the image, obtaining a mesoscopic damage coefficient D 2 through the increment of the length and width of the cracks, the increment of the depth of the cracks and the change of the angles of the cracks before and after blasting, and inputting the mesoscopic damage coefficient D 2 into a surrounding rock blasting damage evaluation module;
as shown in fig. 4, the specific steps are as follows:
The panoramic camera is placed into the obtained borehole through the cable to obtain the optical image of the rock wall in the borehole, and the obtained image is shown in fig. 6. The panoramic camera is provided with a light source, the hole wall is illuminated and shot, the image is transformed by the reflecting mirror to form a panoramic image, the shot image is transmitted to the video distributor by the special cable, one path of the shot image enters the video recorder, the whole detection process is recorded, the other path of the shot image enters the computer to be digitized, the measuring wheel on the winch measures the position of the probe in real time, and the depth value is placed in a special port in the computer through the interface board; the capturing mode of the capturing card is controlled by the depth value, the panoramic image is quickly restored to be a plane unfolding diagram, and a specific image is shown as a figure 7 and is displayed in real time for on-site monitoring;
The image is subjected to graying, the image is processed by utilizing a color psychology formula, the commonly obtained image is a color image, redundant information interference is easy to generate due to interference of colors, sunlight and the like, in many image processing algorithms, the image is subjected to graying, and then the algorithm is designed and improved by using the graying image, wherein the specific formula is as follows:
Gray＝0.30×R+0.59×G+0.11×B(7)
The edges of the cracks are detected through an edge detection operator (Canny operator), the outlines of the cracks are highlighted, the noise immunity and the detection accuracy are good, as many actual edges as possible are identified, the edges are as close to the actual edges in the image as possible, and the edges are only identified once; firstly converting a gray level diagram, then carrying out filtering noise reduction treatment, using a first-order finite difference to calculate gradients to obtain two matrixes of partial derivatives of an image in the x and y directions, using a Sobel operator as a gradient operator in a Canny operator, then inhibiting data of non-maximum values, and also understanding the possibility of eliminating the non-maximum values as edges of the non-maximum values, using a double threshold to screen the image, and selecting a proper large threshold and a small threshold to obtain an edge image closest to the real edge of the image;
The specific example image is shown in fig. 5: fig. 5 is a tunnel face fracture image, wherein the image is processed by detecting the edges of the fracture by an image graying and edge detection operator (Canny operator), so as to obtain specific data of image processing: the threshold information is 0.54; area information is 15454.25; the length information is 461.00; the maximum width information is 77.00; the minimum width information is 25.00; the shape information is a longitudinal slit.
The fracture identification part mainly refers to angle change of the fracture before and after blasting, and then utilizes a functional relation to express the influence brought by the fracture change, and the fracture identification part is embodied through a specific damage coefficient expression. The angle change of the fracture is mostly counted by adopting a method of counting the directions of each point on the main skeleton, establishing a histogram of 8 directions, then solving the inclination angle through a gravity center method, and then calculating the overall average tangential direction of the skeleton by solving the tangential directions of each point of the skeleton, wherein the specific formula is as follows:
Wherein: g 2(y)、g1 (x) is the tangential direction of each point, In the average tangential direction, λ is a characteristic value, and m is the number of points.
After solving the direction of the average tangent line, the angle of the crack can be obtained through an inverse trigonometric function, and the specific formula is as follows:
Wherein: θ is the angle of the crack, g 2(y)、g1 (x) is the tangential direction of each point, Is the direction of the average tangent line.
The width and length of the crack and the depth of the crack are identified according to a digital panoramic drilling shooting technology in a shooting and crack identification module, data are output, and the change of the crack before and after the explosion is accurately judged by combining three indexes through observing the increment of the length and width of the crack before and after the explosion, the increment of the crack depth and the change of the crack angle, wherein the specific formula is as follows:
Wherein: d 2 is a damage coefficient obtained by the camera shooting and fracture identification module, theta 1 is a fracture angle change amount, l 1 is a fracture length increase amount, b 1 is a fracture width increase amount, h 1 is a fracture depth increase amount, l is a fracture length, b is a fracture width, and h is a fracture depth.
Inputting the microscopic damage coefficient D 2 of the surrounding rock into a surrounding rock blasting damage evaluation module, and waiting for cooperative work.
And step four, inputting the correction coefficient alpha, the macroscopic damage coefficient D 1 and the microscopic damage coefficient D 2 into a surrounding rock blasting damage evaluation module, determining the comprehensive index damage coefficient [ D ], and correcting the surrounding rock comprehensive index damage coefficient [ D ] for a plurality of times.
The surrounding rock blasting damage evaluation module integrates data output by the surrounding rock identification module, the ultrasonic detection module and the camera shooting and crack identification module, evaluates the surrounding rock damage of the tunnel, and synthesizes various indexes to draw a functional relation according to the correlation of the surrounding rock damage evaluation of each module so as to finish the final surrounding rock damage test method; and according to the correlation of each module on the surrounding rock damage evaluation, integrating various indexes such as a correction coefficient alpha, a macroscopic damage coefficient D 1 and a microscopic loss coefficient D 2 of the surrounding rock, and drawing a functional relation to finish the final surrounding rock damage test method. The proportion distribution of the surrounding rock damage values of each module is dynamically adjusted through the on-site real-time blasting operation, and the concrete method is as follows:
The surrounding rock blasting damage evaluation module judges the surrounding rock damage by utilizing a functional relation for N times (N= {1,2 … N-1 }), compares the independent initial surrounding rock damage coefficients of all the modules when not blasted, performs calculation and analysis, and generates proportion distribution data of the surrounding rock damage for the first time; the nth time data is used as data of surrounding rock damage by utilizing a functional relation formula for the (n+1) th time, then the data is compared with the independent surrounding rock damage coefficients of all modules during the nth time blasting, calculation and analysis are performed again, the (n+1) th time data are distributed in proportion to the surrounding rock damage, the work is repeated for N times, and the surrounding rock damage coefficients gradually and accurately tend to the real situation.
Specifically, after the surrounding rock damage is judged by using a functional relation for the first time, the initial surrounding rock damage coefficient of each module is compared when not blasted, calculation and analysis are performed, the proportion distribution data of the surrounding rock damage is generated for the first time, the data is used as the data for judging the surrounding rock damage by using the functional relation for the second time, the comparison is performed with the surrounding rock damage coefficient of each module when blasted for the first time, calculation and analysis are performed again, the proportion distribution data of the surrounding rock damage is generated for the second time, the work is repeated, the surrounding rock damage coefficient gradually and accurately tends to the real situation, and the specific formula is as follows:
The initial formula: [D] =α (aD 1+bD2) (12)
One-time allocation formula:
Scaling factor a:
scaling factor b:
The secondary allocation formula: [D] =α (a 1D1+b1D2) (16)
N-time allocation formula: [D] =α (a n-1D1+bn-1D2) (17)
Wherein: alpha is a correction coefficient, a is a proportion coefficient of a macroscopic damage coefficient D 1 of surrounding rock, b is a proportion coefficient of a microscopic damage coefficient D 2 of the surrounding rock, D 1 is a surrounding rock damage coefficient acquired by an ultrasonic detection module, D 2 is a surrounding rock damage coefficient acquired by a shooting and crack recognition module, and n is the number of times when the difference between front and rear damage coefficient distribution is smaller.
By correcting the surrounding rock damage coefficient for many times, the accuracy rate is more and more close to the real condition, and the method can dynamically, efficiently and accurately judge the damage degree of the surrounding rock of the tunnel, and has important significance.
Claims (5)
1. A test method of a deep buried tunnel surrounding rock blasting damage test system is characterized by comprising the following steps of: the testing system specifically comprises a surrounding rock identification module, an ultrasonic detection module, a shooting and crack identification module and a surrounding rock blasting damage evaluation module, wherein the surrounding rock identification module, the ultrasonic detection module and the shooting and crack identification module are all connected with the surrounding rock blasting damage evaluation module;
the surrounding rock identification module is used for generating a surrounding rock point cloud image after a surrounding rock is scanned by using a convolutional neural network to identify laser radar, identifying microscopic structural defects existing on the surface of the surrounding rock and outputting a correction coefficient alpha;
the ultrasonic detection module is specifically used for acquiring surrounding rock internal information by using an acoustic wave instrument, and judging a surrounding rock macroscopic damage coefficient D 1 by using the change characteristics of acoustic waves before and after blasting;
The shooting and crack identification module is specifically a method for utilizing a digital panoramic drilling shooting technology and image processing, and judges the microscopic damage coefficient D 2 of the surrounding rock by observing the increment of the crack length and width, the increment of the crack depth and the change of the crack angle before and after blasting;
The surrounding rock damage evaluation module determines a comprehensive index damage coefficient [ D ] through a correction coefficient alpha input by the surrounding rock identification module, a macroscopic damage coefficient D 1 input by the ultrasonic detection module and a microscopic damage coefficient D 2 input by the camera shooting and crack identification module, and corrects the surrounding rock comprehensive index damage coefficient [ D ] for a plurality of times;
the test method comprises the following steps:
The method comprises the steps of firstly, establishing a network model through a surrounding rock identification module by using collected tunnel face images and data, generating point cloud images through scanning tunnel face images by using a laser radar, calculating areas of target areas in the point cloud images, and according to the ratio of the areas of cracks or holes to the tunnel face, setting up a correction coefficient alpha and inputting the correction coefficient alpha into a surrounding rock blasting damage evaluation module;
The ultrasonic detection module acquires surrounding rock information by utilizing a sound wave test, obtains a macroscopic damage coefficient D 1 of the surrounding rock and inputs the macroscopic damage coefficient D 1 into the surrounding rock blasting damage evaluation module;
The third step, shooting and crack identification module obtains clear visible plane expansion diagram outline images by using a digital panoramic drilling shooting technology and an image processing method, processes the images, identifies cracks in the images, obtains a microscopic damage coefficient D 2 by increasing the length and width of the cracks before and after blasting, increasing the depth of the cracks and changing the angles of the cracks, and inputs the microscopic damage coefficient D 2 into the surrounding rock blasting damage evaluation module;
And step four, inputting the correction coefficient alpha, the macroscopic damage coefficient D 1 and the microscopic damage coefficient D 2 into a surrounding rock blasting damage evaluation module, determining the comprehensive index damage coefficient [ D ], and correcting the surrounding rock comprehensive index damage coefficient [ D ] for a plurality of times.
2. The method for testing the deep-buried tunnel surrounding rock blasting damage testing system according to claim 1, wherein: the calculation formula of the macroscopic damage coefficient D 1 in the second step is as follows:
Wherein: d 1 is the macroscopic damage coefficient of surrounding rock obtained by the ultrasonic detection module, E 0 is the elastic modulus of the rock mass before blasting, E is the equivalent elastic modulus of the rock mass after blasting, V pm、Vpr is the elastic longitudinal wave velocity of the affected rock mass and the complete rock mass after blasting is finished, η is the sound wave reduction rate, and when η is greater than 10%, it is determined that the rock mass is damaged by blasting damage, and the corresponding smoke table damage threshold is D cr =0.19.
3. The method for testing the deep-buried tunnel surrounding rock blasting damage testing system according to claim 1, wherein: in the step (III), the calculation formula of the mesoscopic damage coefficient D 2 is as follows:
Wherein: d 2 is a mesoscopic damage coefficient obtained by a camera shooting and fracture identification module, theta 1 is a fracture angle change quantity, l 1 is a fracture length increase quantity, b 1 is a fracture width increase quantity, h 1 is a fracture depth increase quantity, l is a fracture length, b is a fracture width, and h is a fracture depth.
4. A method of testing a deep tunnel surrounding rock burst damage testing system according to claim 3, wherein: the fracture angle θ is calculated as follows:
Wherein θ is the angle of the crack, g 2(y)、g1 (x) is the tangential direction of each point, Is the direction of the average tangent line.
5. The method for testing the deep-buried tunnel surrounding rock blasting damage testing system according to claim 1, wherein: the comprehensive index damage coefficient [ D ] correction method comprises the following steps:
the initial formula: [D] =α (aD 1+bD2)
One-time allocation formula:
Scaling factor a:
scaling factor b:
The secondary allocation formula: [D] =α (a 1D1+b1D2)
N-time allocation formula: [D] =α (a n-1D1+bn-1D2)
Wherein: alpha is a correction coefficient, a is a proportionality coefficient of a macroscopic damage coefficient D 1 of surrounding rock, b is a proportionality coefficient of a microscopic damage coefficient D 2 of the surrounding rock, D 1 is a surrounding rock damage coefficient acquired by an ultrasonic detection module, D 2 is a surrounding rock damage coefficient acquired by a shooting and crack identification module, and n is correction times.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211615303.5A CN115791803B (en) | 2022-12-15 | Deep-buried tunnel surrounding rock blasting damage test system and test method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211615303.5A CN115791803B (en) | 2022-12-15 | Deep-buried tunnel surrounding rock blasting damage test system and test method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115791803A CN115791803A (en) | 2023-03-14 |
| CN115791803B true CN115791803B (en) | 2024-07-09 |
Family
ID=
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104949868A (en) * | 2015-05-21 | 2015-09-30 | 中国矿业大学 | Blasting damaged rock sample preparation and micro-macro combined damage degree determination method |
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104949868A (en) * | 2015-05-21 | 2015-09-30 | 中国矿业大学 | Blasting damaged rock sample preparation and micro-macro combined damage degree determination method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3382386B1 (en) | Defect detection using ultrasound scan data | |
| US20080195330A1 (en) | Concrete Structure Crack Inspection Device and Crack Inspection Method | |
| CN102928435A (en) | Aircraft skin damage identification method and device based on image and ultrasound information fusion | |
| CN111754463B (en) | Method for detecting CA mortar layer defects of ballastless track based on convolutional neural network | |
| CN113553763A (en) | Tunnel surrounding rock rapid grading method and system | |
| CN114548278A (en) | In-service tunnel lining structure defect identification method and system based on deep learning | |
| den Bieman et al. | Deep learning video analysis as measurement technique in physical models | |
| US20230324577A1 (en) | Real-Time Microseismic Magnitude Calculation Method and Device Based on Deep Learning | |
| CN104181233A (en) | B ultrasound scanning detection method of strain clamp crimping defect based on feature enhancement | |
| CN112949380B (en) | Intelligent underwater target identification system based on laser radar point cloud data | |
| CN115797354A (en) | Method for detecting appearance defects of laser welding seam | |
| CN111783616B (en) | Nondestructive testing method based on data-driven self-learning | |
| CN105411624A (en) | Ultrasonic three-dimensional fluid imaging and speed measuring method | |
| CN117825516A (en) | Ultrasonic defect detection system and method with accurate positioning | |
| CN115616674A (en) | Ground penetrating radar training data set simulation amplification and road nondestructive testing method and system | |
| CN114942436A (en) | Tunnel lining disease identification method and system based on geological radar and deep learning | |
| CN115791803B (en) | Deep-buried tunnel surrounding rock blasting damage test system and test method | |
| CN106645399A (en) | Composite material damage detection and evaluation method and system | |
| CN115791803A (en) | Deep-buried tunnel surrounding rock blasting damage test system and test method | |
| Yamashita et al. | Inspection of visible and invisible features of objects with image and sound signal processing | |
| CN114354351A (en) | Rock destruction process real-time monitoring and early warning system and method based on multi-source heterogeneous data | |
| CN113917562A (en) | Macro-micro structure representation and three-dimensional space construction method and device for deep-buried soft interlayer | |
| CN105891339A (en) | Method, device and system for processing signal of geotechnical engineering medium | |
| CN108007866B (en) | Fruit rheological parameter detection method and system | |
| CN116196038B (en) | Ultrasonic multi-mode image fusion bone mineral density measurement system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |