Safety grading early warning method for risk area of tunnel surrounding rock

Abstract

The invention provides a safety grading early warning method for a risk area of tunnel surrounding rock, and relates to the technical field of geotechnical engineering. Firstly, a three-dimensional laser scanning technology is adopted to perform regular surface scanning on a selected tunnel section, and high-precision tunnel point cloud data is obtained and preprocessed to ensure data quality (S1). And based on the preprocessed tunnel point cloud data, analyzing and determining spatial deformation distribution of the surrounding rock and an evolution law thereof with time (S2). And according to the spatial deformation characteristics, developing a set of identification algorithm for the potential risk area (S3), and establishing a set of quantitative index system to evaluate the risk level of surrounding rock deformation and damage (S4). And finally, on the basis of the research results, constructing a hierarchical safety early warning system for monitoring and early warning the change condition of the surrounding rock risk area in real time, thereby ensuring the safety during tunnel construction and operation (S5).

Description

Technical Field

[0001] The present invention belongs to the technical field of geotechnical engineering, and particularly relates to a method for safety classification and early warning of risk areas of tunnel surrounding rock. Background Technique

[0002] The stress redistribution caused by tunnel excavation unloading may induce the deformation of the surrounding rock, and even cause damage. In order to timely discover and deal with the engineering problems induced by the deformation of the surrounding rock, implementing the deformation monitoring of the surrounding rock can timely understand the safety status of the surrounding rock, reduce the adverse impact on the tunnel structure, extend the service life of the tunnel, eliminate potential safety hazards, effectively avoid serious engineering accidents, and also provide valuable experience and reference for similar projects, improving the efficiency and quality of tunnel construction. The traditional deformation monitoring method is displacement measurement, with a single measurement form, which can only obtain the deformation characteristics of a single measurement point and cannot accurately reflect the overall trend of the surrounding rock deformation, thus unable to identify potential risk areas; moreover, the deformation of the measurement points is mostly obtained manually, with a large operation intensity, a long operation time, and low efficiency; in addition, personnel need to be exposed in the tunnel for a long time, and there is a certain risk to personnel safety.

[0003] Therefore, there is an urgent need for a simple and effective method to identify the development process of potential risk areas during the deformation of the surrounding rock as a whole, and quantitatively describe the risk degree of potential risk areas, so as to realize the safety early warning classification of tunnel surrounding rock deformation. Summary of the Invention

[0004] The purpose of the embodiment of the present invention is to provide a method for safety classification and early warning of risk areas of tunnel surrounding rock, so as to solve the problems that the traditional deformation monitoring method can only obtain the deformation characteristics of a single measurement point or a certain cross-section, cannot accurately identify potential risk areas, and is mostly manually obtained, with a large operation intensity, low efficiency, and low safety.

[0005] To solve the above technical problems, the technical solution adopted by the present invention is a method for safety classification and early warning of risk areas of tunnel surrounding rock, which is specifically carried out according to the following steps:

[0006] S1. Select a tunnel section for risk area identification and monitoring early warning, and regularly scan the selected tunnel surface through a three-dimensional laser scanning system to obtain tunnel point clouds and preprocess them;

[0007] S2. Based on the tunnel point clouds obtained in S1, determine the spatial deformation distribution and evolution process of the tunnel surrounding rock;

[0008] S3. Construct a method for identifying potential risk areas based on the spatial deformation distribution characteristics of the surrounding rock;

[0009] S4. Establish and determine the risk index of surrounding rock deformation and failure;

[0010] S5. Establish a safety classification and early warning method for surrounding rock risk areas.

[0011] Further, the specific steps of S2 are as follows:

[0012] S201. Take the tunnel point cloud of the first period as the reference point cloud and the tunnel point cloud of the second period as the comparison point cloud. Calculate the coordinate transformation matrix between the comparison point cloud and the reference point cloud according to the spatial geometric characteristics of the tunnel point cloud at different times, obtain the coordinate transformation relationship of the comparison point cloud at different times relative to the reference point cloud, and convert the comparison point cloud at different times to the coordinate system where the reference point cloud is located;

[0013] S202. Construct a covariance matrix based on the point set composed of each point in the reference point cloud and its k nearest neighbors, determine the eigenvalues and eigenvectors of the covariance matrix, and determine that the eigenvector corresponding to the minimum eigenvalue is the local normal vector of this point. Calculate the normal average distance from this point along the direction of its local normal vector to the corresponding point in the comparison point cloud in the same coordinate system, so as to obtain the deformation of each point in the comparison point cloud relative to the reference point cloud;

[0014] S203. Take the tunnel point cloud of the first period as the reference point cloud and the tunnel point cloud of the third period as the comparison point cloud. Obtain the deformation of the tunnel point cloud of the third period relative to the tunnel point cloud of the first period according to the steps of S201 - S202, and then calculate the deformation of each point in the point clouds of the second, third,... Nth periods relative to the reference point cloud to obtain the spatio-temporal deformation development process of the surrounding rock in the same tunnel section.

[0015] Further, the specific steps of S3 are as follows:

[0016] S301. Determine the non-coordinated deformation coefficient according to the deformation magnitude of each point of the tunnel surrounding rock obtained in S2:

[0017]

[0018] In the formula, α is the non-coordinated deformation coefficient, U i is the deformation at a certain position of the tunnel surrounding rock, U avg is the average deformation of the tunnel surrounding rock, and D is the equivalent tunnel diameter;

[0019] S302. Determine the non-coordinated deformation coefficient thresholds of the side wall and the crown respectively;

[0020] S303. Determine the non-coordinated deformation coefficient of each point of the tunnel side wall and crown according to formula (1), and compare it with the non-coordinated deformation coefficient threshold of the side wall or crown. If the non-coordinated deformation coefficient is greater than the non-coordinated deformation coefficient threshold, then the area where this point is located is the potential risk area of the surrounding rock.

[0021] Further, the specific steps of S4 are as follows:

[0022] S401. Determine the comprehensive index I reflecting the overall and local deformation characteristics of the potential risk area of the tunnel surrounding rock tunnel as follows:

[0023] I tunnel = F(D, P, A, V) (2)

[0024] In the formula, F is the weight function of the comprehensive index, D is the equivalent tunnel diameter, P is the deformation of the characteristic point, V is the incremental deformation volume of the potential risk area, and A is the area of the potential risk area;

[0025] S402. Determine the area of the potential risk area of the tunnel surrounding rock;

[0026] S403. Determine the incremental deformation volume of the potential risk area of the tunnel surrounding rock;

[0027] S404. Determine the deformation of the characteristic point of the potential risk area of the tunnel surrounding rock.

[0028] Furthermore, the S402 is specifically carried out according to the following steps:

[0029] S4021. If the potential risk area is located on the side wall of the tunnel, the area determination method is as follows:

[0030] S40211. First, project the point cloud of the potential risk area onto a plane parallel to the side wall to obtain a two-dimensional point set; and calculate the boundary vertices based on the outer contour of its projected point cloud; when the projected point cloud of the potential risk area is a convex contour, use the two-dimensional convex hull algorithm to calculate the boundary vertices of the projected point cloud of the potential risk area; if the projected point cloud is a non-convex contour, use the corresponding boundary recognition method to calculate its boundary vertices;

[0031] S40212. Use the angle sorting method to sort the calculated boundary vertices counterclockwise. At this time, the area enclosed by the closed polygon formed by the connection of the boundary vertices is the area A of the potential risk area:

[0032]

[0033] In the formula, x[i] and y[i] are the abscissa and ordinate of the i-th polygon vertex after counterclockwise sorting, x[i + 1] and y[i + 1] are the abscissa and ordinate of the (i + 1)-th polygon vertex after counterclockwise sorting, and n is the number of polygon vertices;

[0034] S4022. If the potential risk area is located on the top arch of the tunnel, the area determination method is as follows:

[0035] S40221. For the point cloud of the tunnel crown area, the ellipsoidal cylinder point cloud plane mapping method is used to map the surface point cloud into a plane point cloud. For each point P(x i , y i , z i ) in the potential risk area, its coordinates (s i , y i , ρ i ) in the mapping coordinate system are determined according to Equation (4):

[0036]

[0037] In the formula, θ i is the included angle between the positive direction of the horizontal axis, i.e., the x-axis, where the center (x c , z c ) of the ellipse obtained by fitting is located, and the ray formed by the center (x c , z c ) of the self-fitting ellipse and the point P(x i , y i , z i ). s i is the arc length starting from the center (x c , z c ) of the ellipse along the positive direction of the horizontal axis, along the edge of the ellipse, in the clockwise or counterclockwise direction to the corresponding point of point P on the ellipse. ρ i is the distance from the point P(x i , y i , z i ) to the central axis;

[0038] S40222. According to the two-dimensional plane point cloud after mapping the point cloud of the crown risk area, calculate the boundary vertices, obtain the closed polygon formed by connecting the boundary vertices, and calculate the area of the two-dimensional plane closed polygon according to the method in S4021 to obtain the area of the crown risk area.

[0039] Furthermore, S403 is specifically carried out according to the following steps:

[0040] S4031. Use the Delaunay triangulation algorithm to triangulate the plane point set obtained in S4211 or S4221 to generate triangulation triangles;

[0041] S4032. Use the boundary recognition algorithm to identify the boundary of the non-convex region of the planar point set, then extract the boundary vertices of the non-convex region to form a closed polygon; calculate the incenter of each triangulation triangle in the planar point set, and obtain the set of triangles whose incenter is located inside the boundary of the non-convex region according to whether the incenter of each triangle is located inside the boundary of the non-convex region; finally, use the Delaunay algorithm to triangulate the planar point set corresponding to the set of triangles whose incenter is located inside the non-convex region.

[0042] S4033. For the case where the potential risk region is located on the tunnel sidewall, first obtain the risk region point cloud based on the potential risk region recognition method determined in S3 and the surrounding rock spatial deformation of the tunnel obtained in S2, project the risk region point cloud onto a plane parallel to the tunnel sidewall, that is, the reference plane, to obtain a two-dimensional reference plane point set; use the non-convex region triangulation method described in S4031 - S4032 to triangulate the two-dimensional reference plane point set to obtain a triangular mesh representing the shape of the risk region projection point cloud, then calculate the volume of the triangular prism corresponding to this triangle from the triangulated triangular mesh and the average deformation corresponding to the triangle vertices, and finally obtain the incremental deformation volume ΔV of the potential risk region:

[0043]

[0044] In the formula, A i is the area of the i-th triangle in the generated triangular mesh, N is the number of triangles in the generated triangular mesh, is the average deformation of the surrounding rock corresponding to the vertices of the i-th triangle:

[0045]

[0046] In the formula, D j 、D j+1 、D j+2 respectively refer to the deformations corresponding to the three vertices of the triangulation triangle;

[0047] For the calculation of the incremental deformation volume of the crown risk region, first use the elliptical cylinder point cloud plane mapping method to convert the surface point cloud into a two-dimensional plane point cloud, then use the non-convex region triangulation method to triangulate this point set to obtain a triangular mesh representing the shape of the risk region projection point cloud, and then calculate the volume of the triangular prism corresponding to this triangle from the triangulated triangular mesh and the average deformation corresponding to the triangle vertices, that is, the incremental deformation volume ΔV of the potential risk region.

[0048] Furthermore, the specific process of S404 is as follows: First, screen out the points whose deformation at each moment is greater than zero in the potential risk region, and among these points, take the arithmetic mean of the deformations of the corresponding points whose deformation amount exceeds the median of the deformation amounts in the risk region as the deformation of the potential risk region characteristic points.

[0049] Further, the S5 is specifically carried out according to the following steps:

[0050] Describe the deformation and failure characteristics during the progressive development of the surrounding rock failure through the actual failure photos of the tunnel surrounding rock at different times, and the failure degree of the local surrounding rock in different time periods, and thereby determine the corresponding deformation risk level; according to the potential risk areas identified by the non-coordinated deformation of the tunnel surrounding rock and their development process, calculate the comprehensive deformation index of the potential risk areas and its change over time by using the comprehensive index of the surrounding rock deformation and failure, establish the relationship between the comprehensive index and its change characteristics of the potential risk areas and the failure characteristics and corresponding risk levels of the tunnel surrounding rock in different time periods, and determine the index thresholds corresponding to "safe", "risk", and "dangerous" for the comprehensive index of the potential risk areas, so as to establish an early warning method composed of three levels of "safe", "risk", and "dangerous".

[0051] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0052] 1. Based on the tunnel point clouds at different times, the present invention reconstructs the spatio-temporal development process of the tunnel surrounding rock deformation, and establishes a prediction method for potential risk areas based on the non-coordinated deformation characteristics of the surrounding rock, realizing the prediction of potential risk areas of the surrounding rock as a whole, facilitating the discovery of abnormal situations that are not easily found during on-site inspections. The steps are simple, the efficiency is high, the manpower and material resources are saved, and it is more in line with the actual deformation situation of the tunnel.

[0053] 2. Based on the prediction of potential risk areas of the surrounding rock, the present invention establishes a comprehensive index for quantifying the risk degree of potential failure areas, calculates the corresponding comprehensive index values for different time periods according to the deformation development process of the potential risk areas, and comprehensively determines the comprehensive index threshold of the potential risk areas of the tunnel surrounding rock in combination with the actual failure situation of the tunnel surrounding rock, thereby realizing the classification of deformation safety early warning of the tunnel surrounding rock, being able to timely discover and respond to engineering problems induced by the surrounding rock deformation; with a relatively high accuracy rate, it can provide technical support for timely understanding the safety status of the tunnel surrounding rock. Description of the Drawings

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0055] Figure 1 It is the spatial deformation distribution diagram of the tunnel surrounding rock in Embodiment 1; (a) is the spatial deformation distribution of the tunnel surrounding rock on the 145th day after the first scan, and (b) is the spatial deformation distribution of the tunnel surrounding rock on the 424th day after the first scan;

[0056] Figure 2 It is the potential risk area map corresponding to the non - coordinated deformation coefficient threshold of 0.4% in Embodiment 1;

[0057] Figure 3 It is the actual failure area map of the right side wall of the tunnel surrounding rock in Embodiment 1;

[0058] Figure 4 It is the line graph of the characteristic indexes of this embodiment changing with time;

[0059] Figure 5 It is the trend map of the potential risk degree of the tunnel surrounding rock in the process of deformation development of this embodiment. Specific implementation mode

[0060] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0061] S1. Obtain tunnel point cloud. The three - dimensional laser scanning technology can efficiently obtain the surface spatial geometric morphology of the tunnel surrounding rock in a non - contact manner, that is, the tunnel point cloud. By obtaining the tunnel point cloud of the same tunnel section at different times, it prepares for the subsequent analysis of the spatio - temporal deformation evolution process of the tunnel surrounding rock. The specific steps are introduced as follows:

[0062] S11. Select the tunnel section for which risk area identification and monitoring and early warning are required;

[0063] S12. Regularly scan the selected tunnel surface using a three - dimensional laser scanning system to obtain the tunnel point cloud for analysis; ensure that the conditions for each scan are as consistent as possible to reduce the influence of external factors (such as temperature, humidity, etc.) on the results.

[0064] S13. Pre - process the tunnel point cloud data obtained in S12, clean and correct the original point cloud data obtained by scanning, and remove noise points, overlapping points, etc. to improve the quality of the point cloud data.

[0065] S2. Calculate the spatial deformation distribution of the tunnel surrounding rock and obtain its evolution process. The specific steps are introduced as follows:

[0066] S21. Refer to Figure 1 (a), Figure 1 (b), the point cloud of the first period is the point cloud data obtained by the earliest three - dimensional laser scanning of the tunnel section; the point clouds of the second, third,..., Nth periods are the point cloud data obtained by three - dimensional laser scanning of the same tunnel section at different time points after the first period.

[0067] S22. Take the tunnel point cloud of the first period as the reference point cloud and the tunnel point cloud of the second period as the comparison point cloud. Calculate the coordinate transformation matrix between the comparison point cloud and the reference point cloud according to the spatial geometric characteristics of the tunnel point cloud at different times, obtain the coordinate transformation relationship of the comparison point cloud at different times relative to the reference point cloud, and transform the comparison point cloud at different times into the coordinate system where the reference point cloud is located.

[0068] S23. For each point in the reference point cloud, find its k nearest neighbor points. Construct a covariance matrix for the point set composed of this point and its k nearest neighbors, solve the eigenvalues and eigenvectors, and the eigenvector corresponding to the minimum eigenvalue is the local normal vector of this point. Then calculate the normal average distance from this point to the corresponding point in the comparison point cloud in the same coordinate system along its local normal vector direction, obtain the deformation of the corresponding point in the comparison point cloud relative to the reference point cloud in the normal direction, and further calculate the deformation of each point in the comparison point cloud relative to the reference point cloud in the normal direction.

[0069] In this embodiment, the central axis of the cylinder is defined by a point in the reference point cloud and its corresponding local normal vector. Then the cylinder of this point can be completely defined by the radius and length of the cylinder. Thus, the point sets belonging to the cylinder in the reference point cloud and the comparison point cloud can be obtained. Then project the subsets corresponding to this point in the reference point cloud and the comparison point cloud along the local normal vector direction respectively, and calculate the average center of the projected point sets respectively. The distance between the projection centers is called the normal average distance.

[0070] S24. Take the tunnel point cloud of the first period as the reference point cloud and the tunnel point cloud of the third period as the comparison point cloud. Obtain the normal relative deformation of the tunnel point cloud of the third period relative to the tunnel point cloud of the first period through these two calculation steps of S22 and S23, and further calculate the deformation of each point in the point clouds of the second, third,..., N periods relative to the reference point cloud in the normal direction.

[0071] Thus, the development process of the surrounding rock spatio-temporal deformation at different times within the same tunnel section can be obtained.

[0072] S3. Construct a method for identifying potential risk areas based on the spatial deformation distribution characteristics of the surrounding rock.

[0073] Through the aforementioned calculations, the spatio-temporal development process of the deformation of the tunnel surrounding rock can be obtained from the tunnel point clouds at different times. Due to factors such as the differences in in-situ stress, geological structure, and the mechanical properties of the surrounding rock mass, there are certain differences in the deformation of different parts of the tunnel surrounding rock. When the deformation of the surrounding rock develops to a certain extent, by comparing the spatial distribution characteristics of the cumulative deformation of the tunnel surrounding rock at different time periods, the preliminary location of the potential risk area can be qualitatively described. However, this method requires subjective judgment by humans and lacks a quantitative judgment standard. Therefore, based on the deformation characteristics of the surrounding rock in the potential risk area, a discriminant method for identifying the potential risk area is proposed. The specific steps are introduced as follows:

[0074] S31. When the deformation of the surrounding rock is small, there is no potential risk area or the potential risk is small in the tunnel surrounding rock. At this time, the deformation of the surrounding rock has a coordinated characteristic. While when the deformation of the surrounding rock is large, due to the anisotropy of the deformation of the tunnel surrounding rock, the deformation differences in different parts are more significant, corresponding to a relatively large deformation in the local area of the surrounding rock. At this time, the deformation of the surrounding rock shows a certain degree of non-coordinated characteristic. Thus, the deformation characteristic of the surrounding rock risk area is that the relative deformation of the local risk area is large, while the relative deformation of the non-risk area adjacent to it is small, that is, the deformation of the local risk area and its adjacent non-risk area is non-coordinated. According to the local-global non-coordinated characteristic of the spatial deformation of the local risk area, the relative magnitude of the deformation of the tunnel surrounding rock is calculated, and a method for identifying the local potential risk area of the tunnel surrounding rock is proposed.

[0075] According to the deformation magnitude of each point of the tunnel surrounding rock obtained in S2, the average deformation of the tunnel surrounding rock is calculated, that is, the sum of the deformations of each point of the tunnel surrounding rock divided by the total number of point clouds, and the discriminant method for the local potential risk area is described based on the characteristic hole diameter of the tunnel. The calculation formula of the non-coordinated deformation coefficient is as shown in Equation (1):

[0076]

[0077] In the formula, α is the non-coordinated deformation coefficient, U i is the deformation at a certain position of the tunnel surrounding rock, U avg is the average deformation of the tunnel surrounding rock, and D is the equivalent hole diameter of the tunnel. Equation (1) reflects the relative magnitude of the non-coordinated deformation of the local risk area of the surrounding rock.

[0078] S32. When the non-coordinated deformation of the surrounding rock develops to a certain extent, the non-coordinated deformation of the surrounding rock increases significantly. At this time, obvious fractures have occurred in the local surrounding rock, corresponding to: when the non-coordinated deformation coefficient α is greater than a certain threshold, non-coordinated deformation appears in the surrounding rock. Due to the action of the in-situ stress field and the differences in the mechanical properties of the surrounding rock mass and geological conditions, the deformation development of the tunnel surrounding rock is also different. At the same time, considering that the deformation and failure of the tunnel crown have a sudden characteristic, while the deformation of the tunnel sidewall usually shows a progressive development process. Therefore, here it is necessary to obtain the threshold of the non-coordinated deformation coefficient of the sidewall and the threshold of the non-coordinated deformation coefficient of the crown respectively.

[0079] In actual application, the magnitude of the non-coordinated deformation coefficient threshold needs to be determined by the measured data of the actual spatial deformation of the tunnel surrounding rock. When the measured data is insufficient, empirical values can be selected to identify the potential risk areas of deformation. The calculated potential risk areas should be given key attention and verified on-site, such as checking whether obvious cracks, excessive local deformation, or large deformation occur on the surface of the surrounding rock at this part. Ensure that the identified potential risk areas match the actual damaged areas of the tunnel. In this embodiment, the measured data of the actual spatial deformation of the tunnel surrounding rock mainly includes the failure characteristics of the tunnel surrounding rock, which mainly originate from the local failure photos of the tunnel surrounding rock.

[0080] S33. Calculate the non-coordinated deformation coefficient of each point on the side wall and the crown arch in the tunnel surrounding rock according to formula (1), and compare it with the non-coordinated deformation coefficient threshold of the side wall or the crown arch. If the non-coordinated deformation coefficient is greater than the non-coordinated deformation coefficient threshold, then the area where this point is located is the potential risk area of the surrounding rock.

[0081] Therefore, the method for identifying local potential risk areas based on the spatio-temporal deformation process of the tunnel surrounding rock is to check the operation status of the tunnel surrounding rock as a whole, which is convenient for finding abnormal situations that are not easily detected during on-site inspections, and can also pre-identify potential risk areas.

[0082] S4. Establish and calculate the risk index of the surrounding rock deformation and failure.

[0083] Traditional deformation monitoring technologies, such as total stations and displacement meters, can only obtain the deformation amount, deformation rate of a single measuring point and the curve of its change with time, and the manual operation intensity of obtaining the surrounding rock deformation using traditional deformation monitoring technologies is relatively large and the collection time is long. Operating in the tunnel environment for a long time increases the safety risks of the operators.

[0084] S41. To make up for the deficiencies of traditional deformation monitoring technologies, based on the potential risk areas of the surrounding rock obtained in S3 and their change process with time, calculate a comprehensive index reflecting the overall and local deformation characteristics of the potential risk areas of the tunnel surrounding rock. Among them, the overall deformation characteristics include the area of the potential risk area and the incremental deformation volume of the potential risk area, and the local deformation characteristics include the deformation of characteristic points. If there are multiple discontinuous potential risk areas, they are divided according to the point cloud segmentation method, such as the k-nearest neighbor algorithm, etc., and their comprehensive potential risk area indexes are calculated respectively. The comprehensive index I of the potential risk area of the tunnel surrounding rock tunnel is:

[0085] I tunnel = F(D, P, A, V) (2)

[0086] In the formula, F is the comprehensive index weight function, D is the equivalent tunnel diameter, P is the deformation of the characteristic points, V is the incremental deformation volume of the potential risk area, and A is the area of the potential risk area. According to the actual failure conditions of the tunnel surrounding rock in different periods, the weights of each quantity in the weight function are determined and then weighted and summed up.

[0087] Since the comprehensive index calculation formula for the potential risk area of the tunnel surrounding rock needs to comprehensively consider the incremental deformation volume of the potential risk area, the area of the potential risk area, and the deformation of the characteristic points, and these indexes are all calculated based on the identified potential risk area and its development process.

[0088] S42. Calculation of the area of the potential risk area.

[0089] There are significant differences in the deformation characteristics between the tunnel side wall and the crown arch. Therefore, the calculation of the area of the potential risk area is divided into the tunnel side wall area and the crown arch for separate calculations. The specific steps are as follows:

[0090] S421. If the potential risk area is located in the tunnel side wall, the area calculation method is as follows:

[0091] S4211. First, project the point cloud of the potential risk area onto a plane parallel to the side wall, that is, project it onto a plane parallel to the YOZ plane to obtain a two-dimensional point set; and calculate the boundary vertices based on the outer contour of its projected point cloud. When the projected point cloud of the potential risk area is a convex contour, the two-dimensional convex hull algorithm can be used to calculate the boundary vertices of the projected point cloud of the potential risk area; if the projected point cloud is a non-convex contour, corresponding boundary recognition methods need to be used to calculate its boundary vertices.

[0092] S4212. Then, sort the calculated boundary vertices counterclockwise using the angle sorting method. At this time, the area enclosed by the closed polygon formed by the connection of the boundary vertices is the area A of the potential risk area:

[0093]

[0094] In the formula, x[i] and y[i] are the abscissa and ordinate of the i-th polygon vertex after counterclockwise sorting respectively, x[i + 1] and y[i + 1] are the abscissa and ordinate of the (i + 1)-th polygon vertex after counterclockwise sorting respectively, and n is the number of polygon vertices.

[0095] S422. If the potential risk area is located in the tunnel crown arch, the area calculation method is as follows:

[0096] S4221. For the point cloud at the tunnel crown arch part, the ellipsoidal cylinder point cloud plane mapping method is used to map the surface point cloud into a plane point cloud. The ellipsoidal cylinder point cloud plane mapping method is to use the ellipsoidal cylinder fitting method to determine the fitting ellipse parameters of the location of the risk area: the horizontal semi-axis length a, the vertical semi-axis length b, and the coordinates of the center of the fitted ellipse (xc , z c ); For each point P(x i , y i , z i ) in the potential risk area, its coordinates (s i , y i , ρ i ) in the mapped coordinate system are determined according to Equation (4):

[0097]

[0098] In the formula, θ i is the positive direction of the x-axis, which is the horizontal axis where the center (x c , z c ) of the ellipse obtained by fitting is located, and the included angle between the ray formed from the center (x c , z c ) of the fitted ellipse to the point P(x i , y i , z i ). s i is the arc length starting from the center (x c , z c ) of the ellipse along the positive direction of the horizontal axis and along the edge of the ellipse, either clockwise or counterclockwise, to the corresponding point of point P on the ellipse. ρ i is the distance from the point P(x i , y i , z i ) to the central axis. Therefore, (s i , y i ) are the two-dimensional plane coordinates after the mapping of the elliptical cylinder point cloud plane.

[0099] S4222. According to the two-dimensional plane point cloud after the mapping of the top arch risk area points, calculate the boundary vertices, obtain the closed polygon formed by connecting the boundary vertices, and calculate the area of the two-dimensional plane closed polygon to obtain the area of the top arch risk area.

[0100] S43. Calculation of the incremental deformation volume of the potential risk area.

[0101] Considering the differences in the deformation development characteristics and sudden characteristics between the tunnel sidewall and the top arch, the incremental deformation volume of the potential risk area is calculated separately for the tunnel sidewall and the top arch in two parts.

[0102] Generally, the Delaunay triangulation algorithm triangulates all points in the convex region surrounding the plane point set. At this time, in addition to triangulating the non-convex region of the plane point set, it also triangulates the concave region. This will lead to non-conformity with the actual geometric characteristics of the point cloud and affect the accuracy of the calculation results.

[0103] Therefore, the conventional Delaunay triangulation algorithm cannot achieve the triangulation of non-convex regions. This embodiment provides a non-convex region triangulation method based on the Delaunay triangulation algorithm.

[0104] S431. Use the Delaunay triangulation algorithm to triangulate the plane point set obtained in S4211 or S4221 to generate triangulation triangles; at this time, the concave regions and non-convex regions within the convex region surrounded by the plane point set are triangulated.

[0105] S432. Determine the calculation boundary of the non-convex region from the plane point set, that is, use the boundary recognition algorithm to identify the boundary of the non-convex region of the plane point set, then extract the boundary vertices of the non-convex region to form a closed polygon; and calculate the incenter of each triangulation triangle within the plane point set. According to whether the incenter of each triangle is located inside the boundary of the non-convex region, obtain the set of triangles whose incenter is located within the non-convex region. Finally, use the Delaunay algorithm to triangulate the plane point set corresponding to the set of triangles whose incenter is located within the non-convex region. According to the above calculation process, the triangulation of the non-convex region within the plane can be achieved.

[0106] S433. Calculate the incremental deformation volume of the potential risk region.

[0107] S4331. For the case where the potential risk region is located on the tunnel side wall, first obtain the risk region point cloud based on the potential risk region identification method obtained in S33 and the surrounding rock spatial deformation of the tunnel obtained in S23. Project the risk region point cloud onto a plane parallel to the tunnel side wall, that is, the reference plane, to obtain a two-dimensional reference plane point set. Then, use the non-convex region triangulation method described in S431 - S432 to triangulate this point set to obtain a triangular mesh representing the shape of the projected point cloud of the risk region. Then, calculate the volume of the triangular prism corresponding to this triangle, that is, the incremental deformation volume corresponding to this triangle, from the triangulated triangular mesh and the average deformation corresponding to the triangle vertices. Finally, obtain the incremental deformation volume ΔV of the potential risk region. The calculation formula is as shown in Equation (5):

[0108]

[0109] In the formula, A i is the area of the i-th triangle in the generated triangular mesh, N is the number of triangles in the generated triangular mesh, is the average deformation of the surrounding rock corresponding to the vertices of the i-th triangle:

[0110]

[0111] In the formula, D j 、Dj+1 and D j+2 respectively refer to the deformations corresponding to the three vertices of the dissected triangle, which are obtained from S23;

[0112] S4332. Calculation of the incremental volume of deformation in the top arch risk area. First, the surface point cloud is converted into a two-dimensional plane point cloud by the ellipsoidal cylinder point cloud plane mapping method, and then the non-convex area triangulation method is used to triangulate the point set to obtain a triangular mesh representing the shape of the risk area projection point cloud. Then, the volume of the triangular prism corresponding to the triangle is calculated from the triangulated triangular mesh and the average deformation corresponding to the triangle vertices, that is, the incremental deformation volume ΔV of the potential risk area.

[0113] S44. Calculation of the deformation of the characteristic points in the potential risk area.

[0114] Using the point with the largest deformation in the potential risk area as the characteristic point can only represent the deformation of a certain point, but cannot characterize the local characteristic deformation of the risk area. In addition, when the overall deformation of the potential risk area is large, the point with the maximum deformation is likely to be missing due to local spalling. To avoid the occurrence of the above situation, first, the points with deformation greater than zero at each moment are screened out within the potential risk area. Among these points, the arithmetic mean of the deformations of the corresponding points whose deformation exceeds the median of the deformations within the risk area is used as the deformation of the characteristic points in the potential risk area.

[0115] According to the deformation of the potential risk area in different time periods, the development processes of these indicators can be obtained by the above calculation methods of the area, incremental deformation volume, and characteristic point deformation of the potential risk area. Thus, the comprehensive index of the potential risk area and its change process over time are calculated. During the calculation process, according to the actual failure conditions of the tunnel surrounding rock in different time periods, the weight values corresponding to the characteristic point deformation, characteristic area, and incremental deformation volume are comprehensively determined.

[0116] The index weights describing the deformation development characteristics of the potential risk area should be related to the actual failure development process of the surrounding rock risk area, and the index weights that can more comprehensively reflect the characteristics of each different time period are larger. The relative magnitude relationship of the weight values corresponding to the characteristic point deformation, characteristic area, and incremental deformation volume is that the weight of the incremental deformation volume index is the largest, while the weight of the characteristic point deformation index is the smallest. Their value ranges are recommended to be [0.15 - 0.25], [0.2 - 0.4], and [0.4 - 0.6] respectively. The sum of the weight values of the characteristic point deformation, characteristic area, and incremental deformation volume indicators is 1.

[0117] S5. Establish a classification safety warning for the surrounding rock risk area.

[0118] The deformation and damage characteristics of the surrounding rock during the progressive development of the surrounding rock damage and the degree of damage to the local surrounding rock in different periods are described through the actual damage photos of the tunnel surrounding rock at different times, and the corresponding deformation risk level is determined accordingly (Table 1). According to the potential risk area identified by the uncoordinated deformation of the tunnel surrounding rock and its development process, the deformation comprehensive index of the potential risk area and its change over time are calculated using the surrounding rock deformation and damage comprehensive index, and the relationship between the comprehensive index of the potential risk area and its change characteristics and the damage characteristics of the tunnel surrounding rock at different periods and the corresponding risk level is established. The indicator thresholds corresponding to the "safety", "risk (warning)" and "danger" corresponding to the comprehensive index of the potential risk area are determined, so as to establish an early warning method consisting of three levels of "safety", "risk (warning)" and "danger".

[0119] Table 1 Tunnel surrounding rock failure characteristics and corresponding risk levels at different times

[0120] Time period Degree of deformation and damage Risk level Comprehensive index value Period 1 Safe Safe <![CDATA[I1]]> Period 2 Continuous deformation, slight damage Safe <![CDATA[I2]]> Period 3 Significant increase in discontinuous deformation Early warning <![CDATA[I3]]> Period 4 Potentially unstable block Early warning <![CDATA[I4]]> Period 5 Small rockfall Dangerous <![CDATA[I5]]> … … …

[0121] Example 1

[0122] S1. Obtain the point cloud of tunnel surface contour.

[0123] S11. A 30m long section in a deep buried tunnel was selected as the research object;

[0124] S12. The tunnel section is scanned regularly using a 3D laser scanning system to obtain the tunnel surface contour point cloud of the tunnel section at 6 different times. The first scan date is 2018 / 8 / 30, and this point is used as the starting point cloud, as the first phase tunnel point cloud; the time intervals of the subsequent five scans from the first scan are 75 days, 145 days, 215 days, 303 days, and 424 days, respectively, as the second, third, fourth, and fifth phase tunnel point clouds. This is used as the analysis object, and the above steps are used to analyze the potential risk areas of the tunnel surrounding rock and carry out deformation safety risk warning.

[0125] S13, pre-processing the tunnel point cloud data obtained in S12, cleaning and correcting the original point cloud data obtained by scanning, removing noise points, overlapping points, etc., to improve the quality of the point cloud data.

[0126] S2. Calculate the spatial deformation distribution of the surrounding rock of the tunnel and obtain its evolution process.

[0127] S21, the tunnel point cloud scanned on 2018 / 8 / 30 is used as the reference point cloud, and the tunnel point cloud scanned 75 days later is used as the comparison point cloud;

[0128] S22, calculating a coordinate transformation matrix between the comparison point cloud and the reference point cloud according to the spatial geometric features of the two phase tunnel point clouds; and transforming the coordinates of the comparison point cloud into the coordinate system of the reference point cloud using the coordinate transformation matrix;

[0129] S23. Calculate the local normal vector of each point of the reference point cloud, and calculate the normal average distance from each point to the corresponding comparison point cloud along the direction of the local normal vector, so as to obtain the deformation amount of the point relative to the reference point cloud. By traversing all points of the reference point cloud, calculate the deformation of each point in the comparison point cloud relative to the reference point cloud, so as to obtain the spatial deformation distribution of the tunnel surrounding rock 75 days after the first scan.

[0130] S24. Adopt the calculation steps of S22 above to obtain the spatial deformation distributions of the tunnel surrounding rock on the 145th day, 215th day, 303rd day, and 424th day after the first scan, that is, the spatio-temporal deformation development process of the tunnel surrounding rock. The spatial deformation distributions of the tunnel surrounding rock on the 145th day and 424th day after the first scan are as Figure 1 a. Figure 1 as shown in b.

[0131] S3. Construct a potential risk area identification method based on the spatial deformation distribution characteristics of the surrounding rock.

[0132] Refer to Figure 1 a. Figure 1 b. According to the spatial deformation of the tunnel surrounding rock and its development process, significant discontinuous deformation and even spalling occur in a local area of the right side wall of the tunnel, as shown in the blue area in Figure 2 . If spalling occurs, the calculated deformation amount is negative. In addition, the tunnel failure investigation results also show that outward bulging has occurred in some areas of the side wall of this tunnel section, and local spalling has even occurred. Looking inward from the spalling area, it can be seen that there is a large gap between the lining and the surrounding rock, that is, significant discontinuous deformation has occurred between the lining and the surrounding rock.

[0133] S31. Take the spatial deformation distribution of the tunnel point cloud on the 424th day after the first scan of the tunnel surrounding rock as the analysis object, and calculate the non-coordination deformation coefficient of each point on the right side wall of the tunnel surrounding rock; the equivalent tunnel diameter is calculated to be 5 m from the designed cross-sectional dimensions of the tunnel; the purpose of selecting the tunnel point cloud on the 424th day is that significant discontinuous deformation has actually occurred in the tunnel surrounding rock at this time. This is selected according to the actual failure situation of the tunnel surrounding rock.

[0134] S32. In order to quantitatively identify the occurrence of significant deformation of the side wall of the tunnel surrounding rock, according to the potential risk area identification method of the tunnel surrounding rock, it is necessary to first determine the threshold of the non-coordination deformation coefficient of the tunnel surrounding rock based on the discontinuous deformation characteristics of the tunnel surrounding rock. For this purpose, discuss the potential risk areas corresponding to different non-coordination deformation thresholds, and compare them with the actual tunnel failure investigation results in the same area at the same time, so as to determine the threshold of the non-coordination deformation coefficient of the tunnel surrounding rock.

[0135] Calculate the non-coordination deformation coefficient threshold α according to Equation (1) thresholdThe potential risk area point clouds corresponding to 0.1%, 0.2%, 0.3%, 0.4%, 0.5%, and 0.6% respectively are compared with the actual damaged area. As Figure 2 , Figure 3 shown, when the non-coordinated deformation threshold α threshold is 0.4%, the identified potential risk area coincides with the actual damaged area of the right sidewall of the tunnel. Therefore, in this embodiment, the non-coordinated deformation threshold is taken as 0.4%.

[0136] Based on the determined non-coordinated deformation coefficient threshold α threshold of the surrounding rock deformation, the potential risk areas corresponding to all times of the right sidewall of the tunnel are calculated by Equation (1), and thus the point cloud of the potential risk area and its development process of spatial deformation over time are obtained.

[0137] S4. Establish and calculate the risk index of surrounding rock deformation and failure.

[0138] S41. In order to calculate the comprehensive index reflecting the overall and local deformation characteristics of the potential risk area of the tunnel surrounding rock sidewall, that is, the area of the potential risk area of the sidewall, the incremental deformation volume of the potential risk area, and the deformation of the characteristic points. According to the overall and local deformation characteristics of the potential risk area of the tunnel surrounding rock, the calculation formula for the comprehensive index of the potential risk area of the tunnel surrounding rock is as follows:

[0139] I tunnel = F(D, P, A, V)

[0140] In the formula, F is the comprehensive index weight function, D is the equivalent tunnel diameter, P is the deformation of the characteristic point, V is the incremental deformation volume of the potential risk area, and A is the area of the potential risk area.

[0141] Since the characteristic tunnel diameter D involved in this embodiment is 5.0 m, and the potential risk areas at different time periods have been calculated in the previous step.

[0142] To illustrate the specific calculation methods of the area, incremental deformation volume, and characteristic point deformation of the potential risk area of the sidewall, taking the potential risk area of the tunnel sidewall after 424 days of long-term deformation of the surrounding rock with August 30, 2018 as the start time of calculation as an example, the calculation processes are discussed respectively.

[0143] S42. Calculation of the area of the potential risk area.

[0144] S421. Since the identified is the potential risk area of the tunnel sidewall, the area calculation method is as follows:

[0145] S4211. The point cloud of the potential risk area on the tunnel sidewall is actually coordinates in three-dimensional space. To facilitate the calculation of the area of this area, it is first projected onto the YOZ plane to obtain a two-dimensional point set; and the boundary vertices are calculated based on the outer contour of its projected point cloud. The projected point cloud of the potential risk area is a non-convex contour. At this time, if the two-dimensional convex hull algorithm is used to calculate the boundary of this area, a large error will occur. To accurately identify the boundary of this area, the built-in function boundary in MATLAB is used to identify its boundary vertices;

[0146] S4212. Then, the vertices of the identified potential risk area are sorted counterclockwise using the angle sorting method. The connection lines of the vertices of the potential risk area after counterclockwise sorting form a closed polygon surrounding this area, and the following formula is used to calculate the area of this closed polygon.

[0147]

[0148] In the formula, x[i] and y[i] are the abscissa and ordinate of the i-th polygon vertex after counterclockwise sorting respectively, x[i + 1] and y[i + 1] are the abscissa and ordinate of the (i + 1)-th polygon vertex after counterclockwise sorting respectively, and n is the number of polygon vertices.

[0149] S43. Calculation of the incremental deformation volume of the potential risk area.

[0150] S431. To facilitate the calculation of the incremental deformation volume of the potential risk area on the sidewall, the three-dimensional coordinates corresponding to the potential risk area on the sidewall are projected onto a plane parallel to the tunnel sidewall, that is, the YOZ plane. Since the potential risk area on the sidewall projected onto the YOZ plane is a non-convex area, according to the non-convex area triangulation method in Embodiment 1, first, the built-in triangulation function delaunayTriangulation in MATLAB is used to triangulate the plane projection point set of the potential risk area of the surrounding rock.

[0151] S432. The built-in boundary function boundary in MATLAB is used to calculate the non-convex boundary of the plane projection point set of the potential risk area of the surrounding rock, and the incenter of each triangulated triangle is calculated. Based on this, it is determined whether each triangulated triangle is located within the non-convex boundary, and thus a set of triangles whose incenter is located within the non-convex area is obtained. Finally, from the set of triangles whose incenter is located within the non-convex area and the plane point set, and using the triangulation function triangulation to triangulate the plane point set, the triangulation of the non-convex area formed by the plane projection point set of the potential risk area of the surrounding rock is realized.

[0152] S433. According to the above triangulation steps of the non-convex region, a triangular mesh representing the shape of the projected point cloud of the risk region is obtained, and the volume of the triangular prism corresponding to each triangle in the triangulated region and the average deformation corresponding to its vertices is calculated, that is, the incremental deformation volume corresponding to this triangle. Finally, the incremental deformation volume ΔV of the potential risk region is obtained, as shown in the following formula:

[0153]

[0154] In the formula, A i is the area of the i-th triangle in the generated triangular mesh, N is the number of triangles in the generated triangular mesh, is the average deformation of the surrounding rock corresponding to the vertices of the i-th triangle,

[0155]

[0156] In the formula, D j 、D j+1 、D j+2 respectively refer to the deformations corresponding to the three vertices of the triangulated triangle.

[0157] S44. Calculation of the deformation of the characteristic points of the potential risk region of the side wall.

[0158] When calculating the deformation of the characteristic points of the potential risk region, it is required that the deformations corresponding to all points in the calculation region are greater than zero, that is, the falling block region is not included. For this reason, based on the identified potential risk region of the side wall, first, the points with deformations greater than zero are screened out, and then the arithmetic mean of the deformations of the corresponding points whose deformation amounts exceed the median of the deformation amounts of the risk region is used as the deformation of the characteristic points of the potential risk region.

[0159] According to the above calculation steps, the characteristic indexes of the potential risk region of the tunnel side wall in a certain time period can be obtained, and then the size of the comprehensive index of the potential risk region corresponding to this time period is calculated according to the comprehensive index calculation formula of the potential risk region of the tunnel surrounding rock (Formula (2)). Thus, the change process of the comprehensive index of the potential risk region over time in all time periods can be obtained, as shown in Table 2.

[0160] According to the actual failure conditions of the tunnel surrounding rock in different time periods, the weight values corresponding to the deformation of the characteristic points, the characteristic area, and the incremental deformation volume are comprehensively determined to be 0.15, 0.35, and 0.50 respectively. Then, the comprehensive index of the potential risk region in the corresponding time period is calculated from the deformation of the characteristic points, the characteristic area, and the incremental deformation volume corresponding to different time periods, as shown in Table 2.

[0161] Table 2 Characteristic deformation indexes of the potential risk region in different time periods

[0162] Time span / day Deformation of characteristic point / m <![CDATA[Feature area / m 2 > <![CDATA[Incremental deformation volume / m 3 > Comprehensive index value 75 0.0701 0.0 0.0 0.0105 145 0.0396 0.0837 0.0031 0.0368 215 0.0422 0.3669 0.0142 0.1418 303 0.0433 1.1375 0.0462 0.4277 424 0.0516 1.4392 0.0697 0.5463

[0163] S5. Establish a hierarchical safety warning for the surrounding rock risk areas.

[0164] Based on the actual damage conditions of the right side wall at different times obtained from the tunnel damage investigation, determine the deformation and failure characteristics during the progressive development of the surrounding rock failure, the damage degree of the tunnel side wall in different time periods, and thereby determine their corresponding deformation risk levels, as shown in Table 3.

[0165] Table 3 Failure characteristics, risk levels and their corresponding comprehensive index values of tunnel surrounding rock in different time periods

[0166]

[0167] As shown in Table 3, according to the relationship between the comprehensive index values calculated from the potentially risky areas identified in the corresponding time periods and the actual deformation and failure characteristics of the right side wall of the tunnel surrounding rock, and considering the reserved safety margin, the index thresholds corresponding to "safe" and "risk (warning)", and "risk (warning)" and "danger" for the comprehensive index of the potentially risky area on the right side wall of the tunnel are determined to be 0.12 and 0.50 respectively. In this embodiment, according to the calculation results, when the comprehensive index is 0.1418, a potentially risky area appears. At this time, the corresponding index value is set as the risk threshold, but considering a certain safety factor, its value is determined to be 0.12. The comprehensive index value of 0.12 corresponds to the deformation changing from "safe" to "risk".

[0168] The index thresholds corresponding to "risk" and "danger" are 0.50 respectively.

[0169] Through the above calculation method and the corresponding index thresholds, the risk degree of the potentially risky area can be comprehensively determined, thereby realizing the safety grading warning of the tunnel surrounding rock.

[0170] In the prior art, deformation measurement focuses on recording the displacement of a single measurement point over time. However, in tunnel engineering, if local damage occurs to the surrounding rock or it is affected by excavation blasting, it may cause damage to the measurement points or loss of data. In addition, the data of a single measurement point is not sufficient to comprehensively reflect the deformation trend of the entire tunnel surrounding rock, because such single and discrete data points can only capture local deformation characteristics, similar to the "characteristic deformation" defined in this embodiment (such as Figure 4 shown by the blue solid line in). To more accurately evaluate the overall stability of the tunnel surrounding rock, this embodiment introduces two more comprehensive indicators: characteristic area and incremental deformation volume (i.e., characteristic volume). Among them, the "characteristic area" ( Figure 4 shown by the blue dashed line in) is used to describe the trend of the gradually expanding potentially risky area; while the "incremental deformation volume" or "characteristic volume" ( Figure 4The pink solid line (in []) is used to quantify the degree of discontinuous deformation occurring during the development of the potential risk area. By combining the use of these comprehensive indicators in this embodiment, the potential risk level of the tunnel surrounding rock during the entire deformation process can be monitored and evaluated more comprehensively (see Figure 5 ), thus providing more reliable support for engineering decisions. This method not only makes up for the limitations of traditional single-point deformation measurement but also enhances the understanding and prediction ability of the tunnel safety condition.

[0171] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and reference can be made to the corresponding part of the method embodiment for the relevant content.

[0172] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A risk area safety classification early warning method for tunnel surrounding rock, characterized in that: Follow these steps: S1. Select a tunnel section for risk area identification and monitoring and early warning. Use a 3D laser scanning system to regularly scan the selected tunnel surface to obtain tunnel point cloud and pre-process it. S2, based on the tunnel point cloud data obtained in S1, determine the spatial deformation distribution and evolution process of the tunnel surrounding rock; S3. Construct a potential risk area identification method based on the spatial deformation distribution characteristics of surrounding rock; S4. Establish and determine the risk index of surrounding rock deformation and failure; S5. Establish a safety classification and early warning method for surrounding rock risk areas.

2. A risk area safety classification early warning method for tunnel surrounding rock according to claim 1, characterized in that: The S2 is specifically performed according to the following steps: S201, taking the first phase tunnel point cloud as the reference point cloud and the second phase tunnel point cloud as the comparison point cloud, calculating the coordinate transformation matrix between the comparison point cloud and the reference point cloud according to the spatial geometric features of the tunnel point cloud at different times, obtaining the coordinate transformation relationship of the comparison point cloud at different times relative to the reference point cloud, and converting the comparison point clouds at different times to the coordinate system of the reference point cloud; S202, constructing a covariance matrix based on each point in the reference point cloud and a point set consisting of its k nearest neighbors, and determining the eigenvalues ​​and eigenvectors of the covariance matrix, determining that the eigenvector corresponding to the minimum eigenvalue is the local normal vector of the point, and calculating the average normal distance from the local normal vector of the point to the corresponding point in the comparison point cloud, thereby obtaining the deformation of each point in the comparison point cloud relative to the reference point cloud; S203. Take the first phase tunnel point cloud as the reference point cloud and the third phase tunnel point cloud as the comparison point cloud, and follow the steps of S201 to S202 to obtain the normal relative deformation of the third phase tunnel point cloud relative to the first phase tunnel point cloud, and then calculate the deformation of each point in the second, third, ...N phase point clouds relative to the reference point cloud, and obtain the spatiotemporal deformation development process of the surrounding rock in the same tunnel section.

3. The risk area safety classification early warning method of tunnel surrounding rock according to claim 1 is characterized in that: The S3 is specifically performed according to the following steps: S301. Determine the incompatible deformation coefficient based on the deformation size of each point of the tunnel surrounding rock obtained in S2: Where α is the non-coordinated deformation coefficient, U i is the deformation of a certain position of the tunnel surrounding rock, U avg is the average deformation of the tunnel surrounding rock, and D is the equivalent diameter of the tunnel; S302, respectively determining the incompatible deformation coefficient threshold of the side wall and the incompatible deformation coefficient threshold of the top arch; S303. Determine the incompatible deformation coefficient of each point on the tunnel side wall and top arch according to formula (1), and compare it with the incompatible deformation coefficient threshold of the side wall or top arch. If the incompatible deformation coefficient is greater than the incompatible deformation coefficient threshold, then the area where this point is located is a potential risk area of ​​the surrounding rock.

4. The risk area safety classification early warning method of tunnel surrounding rock according to claim 1 is characterized in that: The S4 is specifically performed according to the following steps: S401. Determine the comprehensive index I reflecting the overall and local deformation characteristics of the potential risk area of ​​the tunnel surrounding rock tunnel for: I tunnel =F(D,P,A,V) (2) Where F is the comprehensive index weight function, D is the equivalent tunnel diameter, P is the deformation of the feature point, V is the incremental deformation volume of the potential risk area, and A is the area of ​​the potential risk area. S402, determine the potential risk area of ​​the tunnel surrounding rock; S403, determining the incremental deformation volume of the potential risk area of ​​the tunnel surrounding rock; S404. Determine the deformation of characteristic points in the potential risk area of ​​the tunnel surrounding rock.

5. A risk area safety classification early warning method for tunnel surrounding rock according to claim 4, characterized in that: The S402 is specifically performed according to the following steps: S4021. If the potential risk area is located on the tunnel side wall, the area is determined as follows: S40211. First, project the point cloud of the potential risk area onto a plane parallel to the side wall to obtain a two-dimensional point set; and calculate the boundary vertices according to the outer contour of the projected point cloud; when the projected point cloud of the potential risk area is an outer convex contour, use a two-dimensional convex hull algorithm to calculate the boundary vertices of the projected point cloud of the potential risk area; if the projected point cloud is a non-outer convex contour, use a corresponding boundary recognition method to calculate its boundary vertices; S40212. Sort the calculated boundary vertices counterclockwise using an angle sorting method. At this time, the area enclosed by the closed polygon formed by the boundary vertices is the area A of the potential risk area: Where x[i] and y[i] are the horizontal and vertical coordinates of the ith polygon vertex after counterclockwise sorting, x[i+1] and y[i+1] are the horizontal and vertical coordinates of the i+1th polygon vertex after counterclockwise sorting, and n is the number of polygon vertices. S4022. If the potential risk area is located in the tunnel crown, the area is determined as follows: S40221. For the point cloud of the tunnel arch, the curved point cloud is mapped into a plane point cloud using the elliptical cylinder point cloud plane mapping method. For each point P(x i ,y i ,z i ), and determine its coordinates (s) in the mapping coordinate system according to formula (4) i ,y i ,ρ i ): In the formula, θ i is the center of the ellipse obtained by fitting (x c ,z c ) is located on the horizontal axis, that is, the positive direction of the x-axis and the center of the self-fitting ellipse (x c ,z c ) to point P(x i ,y i ,z i ) form the angle between the rays, s i is the distance from the center of the ellipse (x c ,z c ) Starting from the positive direction of the horizontal axis, along the edge of the ellipse, the arc length of the corresponding point of point P on the ellipse in a clockwise or counterclockwise direction is ρ i For point P(x i ,y i ,z i ) to the central axis; S40222. Based on the two-dimensional plane point cloud mapped from the top arch risk area point cloud, calculate the boundary vertices according to the method in S4021, obtain the closed polygon formed by the boundary vertex lines, calculate the area of ​​the two-dimensional plane closed polygon, and obtain the area of ​​the top arch risk area.

6. A risk area safety classification early warning method for tunnel surrounding rock according to claim 4, characterized in that: The S403 is specifically performed according to the following steps: S4031, triangulate the plane point set obtained in S4211 or S4221 using the Delaunay triangulation algorithm to generate triangulated triangles; S4032, using a boundary recognition algorithm to identify the boundary of the non-convex region of the plane point set, and then extracting the boundary vertices of the non-convex region to form a closed polygon; calculating the incenter of each subdivided triangle in the plane point set, and obtaining a set of triangles whose incenters are located in the non-convex region according to whether the incenter of each triangle is located inside the boundary of the non-convex region; finally, using the Delaunay algorithm to triangulate the plane point set corresponding to the set of triangles whose incenters are located in the non-convex region; S4033. For a situation where the potential risk area is located at the tunnel side wall, firstly, based on the potential risk area identification method determined in S3, a point cloud of the risk area is obtained and the spatial deformation of the surrounding rock of the tunnel obtained in S2, and the point cloud of the risk area is projected onto a plane parallel to the tunnel side wall, i.e., a reference plane, to obtain a two-dimensional reference plane point set; The non-convex area triangulation method described in S4031-S4032 is used to triangulate the two-dimensional reference plane point set to obtain a triangular mesh representing the shape of the projected point cloud of the risk area. The volume of the triangular prism corresponding to the triangle is calculated based on the triangular mesh after triangulation and the average deformation corresponding to the triangle vertices. Finally, the incremental deformation volume ΔV of the potential risk area is obtained: In the formula, A i is the area of ​​the i-th triangle in the generated triangular mesh, N is the number of triangles in the generated triangular mesh, is the average deformation of the surrounding rock corresponding to the i-th triangle vertex: Where D j , D j+1 , D j+2 They refer to the deformations corresponding to the three vertices of the subdivided triangle respectively; For the calculation of the incremental deformation volume of the risk area of ​​the top arch, the surface point cloud is first converted into a two-dimensional plane point cloud using the elliptical cylinder point cloud plane mapping method, and then the point set is triangulated using the non-convex area triangulation method to obtain a triangular mesh representing the shape of the projected point cloud of the risk area. The triangular prism volume corresponding to the triangle is calculated based on the triangular mesh after triangulation and the average deformation corresponding to the triangle vertices, that is, the incremental deformation volume ΔV of the potential risk area.

7. The risk area safety classification early warning method of tunnel surrounding rock according to claim 4 is characterized in that: The S404 is specifically performed according to the following process: first, points whose deformation at each moment is greater than zero are screened out in the potential risk area, and among these points, the arithmetic mean of the deformation of the corresponding points whose deformation exceeds the median of the deformation in the risk area is taken as the deformation of the characteristic points of the potential risk area.

8. The method for early warning of risk area safety classification of tunnel surrounding rock according to claim 1 is characterized in that: The S5 is specifically performed according to the following steps: The actual destruction photos of the tunnel surrounding rock at different times are used to describe the deformation and destruction characteristics of the surrounding rock in the progressive development process, the degree of local surrounding rock destruction in different periods, and thus determine the corresponding deformation risk level; according to the potential risk area identified by the uncoordinated deformation of the tunnel surrounding rock and its development process, the deformation comprehensive index of the potential risk area and its change over time are calculated using the surrounding rock deformation and destruction comprehensive index, and the relationship between the comprehensive index of the potential risk area and its change characteristics and the tunnel surrounding rock destruction characteristics and corresponding risk levels at different periods is established, and the indicator thresholds corresponding to the "safety", "risk" and "danger" corresponding to the comprehensive index of the potential risk area are determined, thereby establishing an early warning method consisting of three levels of "safety", "risk" and "danger".

Images