CN115511850A - Method for identifying stable state of side landslide - Google Patents

Abstract

The invention discloses a method for identifying a stable state of a side landslide, which comprises the following steps: carrying out fixed-time interval aerial photography on the landslide through unmanned aerial vehicle photogrammetry, establishing a DOM (document object model), a DSM (digital document model) and a three-dimensional point cloud model, carrying out boundary identification of the landslide, and identifying crack development and position distribution; filtering point cloud data by an iCSF method for correcting the gravity direction to obtain a visual displacement field of the side landslide; measuring and recording the crack development condition by adopting a crack monitoring device; carrying out dynamic continuous monitoring on critical position monitoring points and crack development in the edge landslide area, the edge landslide front edge, the edge landslide rear edge and the edge landslide middle part, recording and identifying the crack development and deformation conditions of the edge landslide, and carrying out primary identification on the edge landslide development stage; the obtained side landslide crack development condition and the obtained deformation condition are integrated to further identify the state of the side landslide; the method can timely and accurately capture the side landslide crack and deformation development trend, and has strong adaptability to the terrain complex area.

Description

A method for identifying the stable state of side landslides

technical field

The invention belongs to the technical field of geotechnical engineering, and in particular relates to a method for identifying a stable state of a side landslide.

Background technique

The identification of the stable state of a side landslide consists of two aspects: fracture development and surface deformation. Generally speaking, the accurate judgment of the landslide development stage requires the support of many data including deep deformation, surface deformation, and surface crack development. However, the acquisition of deep deformation requires drilling and burying inclinometer equipment and the data are easily affected by the environment. The reflection of the landslide evolution stage on the surface is more intuitive;

Rainfall, strong earthquakes, and engineering disturbances often cause a certain degree of damage to the mountain, resulting in a large number of cracks of different scales on the slope. With the gradual accumulation of damage to the rock mass in the damaged part, the cracks in different parts of the slope penetrate and merge, and the failure boundary will be formed in the slope, thereby reducing the stability of the slope and accelerating the instability of the slope; The role of the dominant channel for surface water infiltration also accelerated the initiation of side landslides. Therefore, the identification and monitoring of cracks in key positions of side landslides is an important part of geological disaster investigation and monitoring.

For the identification of the stable state of the above-mentioned side landslide, the traditional measurement technology has the defects of mobility, time-consuming and labor-intensive, feedback lag, and low precision in the traditional shooting method.

Contents of the invention

The purpose of the present invention is to solve the problems of traditional measurement technology in terms of mobility, time-consuming and labor-intensive, feedback lag, and low precision in the traditional measurement technology for the identification of the stable state of the side landslide.

In order to solve the above problems, the present invention specifically relates to a method for identifying a stable state of a side landslide, comprising the following steps:

Step 1. Carry out aerial photography of the landslide at fixed time intervals by UAV photogrammetry, establish DOM model, DSM model and 3D point cloud model, extract surface slope information based on DSM model and roughness information based on 3D point cloud model extraction; On this basis, identify the boundary of side landslides, and identify the development and location distribution of cracks;

Step 2. Based on the 3D point cloud model obtained in step 1, filter the point cloud data by correcting the gravity direction iCSF method, separate and extract ground points, check the relative position of the image control points to obtain the relative error of monitoring, M3C2 (multi-scale The model-to-model cloud comparison) method is used to monitor the deformation of the multi-temporal point cloud model to obtain the visualized displacement field of the side landslide;

Step 3. Based on the crack position determined in step 1, use a crack monitoring device to measure the crack development status, and perform regular readings and records on the crack monitoring device through manual/drone inspection;

Step 4, when the crack develops to a certain stage and forms a clear landslide boundary, arrange monitoring points at positions reflecting the deformation characteristics of the side landslide, such as the front edge and the rear edge of the side landslide;

Step 5: Set up the target prism at the monitoring point, place the GPS receiver antenna, and set up a reference point in the area other than the side landslide. The monitoring of the reference point can invert the deformation of the monitoring point;

Step 6. Carry out dynamic and continuous monitoring of the side landslide area, the front edge of the side landslide, the rear edge of the side landslide, and the key position monitoring points (GPS RTK) and crack development (crack meter) in the middle of the side landslide, and record and identify the development and development of side landslide cracks in real time. Deformation status, preliminary identification of the development stage of side landslides;

Step 7, further identifying the state of the side landslide by combining the crack development status of the side landslide obtained in step 6 and the deformation status obtained in step 2.

Further, in the step 1, the identification of the boundary of the side landslide is carried out, and the specific method for identifying the development and location distribution of the cracks is: compared with its local neighbors, there is a sudden change in the slope of the steep slope or crack of the side landslide, and the DSM slope map identification For larger scarps, the DOM model identified fractures developed near the main fractures at the trailing edge of the landslide.

Further, the method for filtering the point cloud data by correcting the iCSF method of the gravity direction in the second step is:

The CSF algorithm first inverts the 3D point cloud data, and then covers the inverted point cloud by simulating the "cloth", and generates an approximate surface according to the position of the "cloth" surface covering the point cloud under the action of gravity; by comparing the original point The distance between points in the cloud data and the surface of the generated "cloth", ground points are identified from the raw point cloud data, and distance points are separated for further feature extraction;

Simulated cloth According to Newton's second law, the relationship between cloth position and force is determined by the following formula:

In the formula, X represents the position of the particle at time t; F _ext (X,t) represents the external force, which is composed of gravity and the collision force generated by obstacles:

F _ext (X,t)＝mg+f _interact (X,t) (2)

When the particle encounters some objects in its moving direction; F _int (X,t) represents the internal force of the particle at position X and time t, and the internal force is generated by interacting with the point cloud (boundary); because both internal and external forces change with time t change, so the above equations are solved by numerical integration (such as Euler's method) in the realization of cloth simulation;

The above method is suitable for gentle surfaces, and is not friendly to the terrain characteristics of side landslides. Therefore, the gravity direction of F _ext (X,t) is changed from the vertical direction of the traditional CSF method to the direction perpendicular to the slope; this direction is passed The median value of the slope of the DSM model is determined; the steps are as follows:

(1) Obtain the median α of the regional slope based on the DSM model generated by the UAV photography technology, and determine the normal vector [N _x , N _y , N _z ] according to the median slope;

(2) The first gravitational item on the right side of the modified formula (2) g′=[N _x , N _y , N _z ]·9.81;

(3) Bring the corrected gravity term back to formula (2) to start filtering.

Further, the specific method of the third step is: fix a steel nail at both ends of the crack to ensure the synchronization between the steel nail and the movement of the soil on both sides, and fix the initial scale end and the free end of the electronic vernier caliper on the two cracks respectively. On the steel nail, the development of cracks can drive the steel nail to move, thereby changing the reading of the electronic vernier caliper, and regularly reading and recording the crack meter vernier caliper through manual/drone inspection;

Further, the reference point in step five is set in a stable area 30m away from the side landslide body.

Further, the specific method of said step six is:

动态监测阶段夹角为

解算公式如下：Assume that the reference point P1 is arranged on the relatively stable side of the fracture, and the monitoring point P2 is arranged on the unstable side of the fracture. First, use RTK to locate the reference point P1 and monitoring point P2 to obtain the world coordinates (X ₁₀ , Y ₁₀ , Z ₁₀ ) of P1 and P2 points; then perform dynamic continuous monitoring on the reference point P1 and monitoring point P2 to obtain two points The world coordinates (X _1t , Y _1t , Z _1t ), (X _2t , Y _2t , Z _2t ), through the calculation, the data of the dislocation length l, fracture width d and dislocation height h during the fracture development process can be obtained, assuming The angle between the line connecting P1 and P2 and the x-axis is

The included angle in the dynamic monitoring stage is

The solution formula is as follows:

h＝Z _2t -Z _1t

Based on the data of dislocation length l, fissure width d and dislocation height h in the process of fracture development, the development stages of side landslides are initially identified.

Further, the method for preliminary identification of the development stage of the side landslide based on the data of the dislocation length l, the fissure width d and the dislocation height h in the crack development process is as follows:

(1) For a side landslide without cracks, it is judged to be in a stable state;

(2) For those with a small amount of cracks but no obvious side landslide boundaries, it is judged to be in a relatively stable state;

(3) When the cracks continue to develop and form a clear edge landslide boundary, it is judged to be in a relatively unstable state;

(4) For the side landslides with bulging at the leading edge and dislocation of the trailing edge, it is judged to be in an unstable state. When the change rate of the three parameters of displacement length l, crack width d and dislocation height h is significantly increased, it is necessary to carry out early warning.

Further, the method for further identifying the state of the side landslide by the crack development status of the side landslide obtained in the comprehensive step 6 of the step 7 and the deformation status obtained in the step 2 is:

When there are no cracks in the side landslide area and the deformation is small, it is judged that the side landslide area is in a stable state; when some cracks are produced in the area, but the boundary has not yet formed and the deformation rate does not exceed 10mm/month, it is judged that the side landslide area is in a relatively stable state; Cracks continue to develop on the surface to form a clear boundary of the landslide and the deformation rate exceeds 10mm/month, which is judged to be in a relatively unstable state; in the local landslide area, the cracks at the trailing edge of the landslide are dislocated, and the leading edge bulges to appear feathery cracks, and the deformation rate exceeds 50mm/month month, the judgment area is in an unstable state.

Generally speaking, compared with the prior art, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

(1) The side landslide stable state recognition method based on unmanned aerial vehicle technology of the present invention can realize replacing artificial by using unmanned aerial vehicle, all-weather 360 degree without dead angle scanning, monitoring of side landslide form.

(2) The side landslide stable state recognition method based on unmanned aerial vehicle technology of the present invention can timely and accurately capture side landslide cracks and deformation development trends, and the results provided comprehensively and objectively reflect the actual situation of side landslide states.

(3) The side landslide stable state recognition method based on unmanned aerial vehicle technology of the present invention can monitor and evaluate complex terrain areas such as other landslides, railways, road side landslides, etc., and has strong adaptability.

Description of drawings

Fig. 1 is a schematic diagram of a logic flow of a preferred embodiment of the present invention;

Fig. 2 is the preferred implementation of the present invention based on the DSM image generation side landslide topographic gradient figure;

Fig. 3 is the DOM image figure based on landslide scarp and crack recognition of preferred implementation of the present invention;

Fig. 4 is a comparison result diagram of the point cloud of a certain landslide on December 18, 2021 and the point cloud of December 23, 2021;

Fig. 5 is the schematic diagram of crack monitoring by unmanned aerial vehicle in a preferred embodiment of the present invention;

Fig. 6 is the deformation monitoring schematic diagram of the key position of side landslide of preferred embodiment of the present invention;

Fig. 7 is a photogram of a stable state for aerial photography in a preferred embodiment of the present invention;

Fig. 8 is a photo diagram of a relatively stable state for aerial photography in a preferred embodiment of the present invention;

Fig. 9 is a photo diagram of a relatively unstable state for aerial photography in a preferred embodiment of the present invention;

Fig. 10 is a photograph diagram of an unstable state for aerial photography in a preferred embodiment of the present invention;

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not constitute a conflict with each other.

Please refer to Fig. 1, a kind of edge landslide stable state identification method, comprises the following steps:

Step 1. Carry out aerial photography of the landslide at fixed time intervals by UAV photogrammetry, establish DOM model, DSM model and 3D point cloud model, extract surface slope information based on DSM model and roughness information based on 3D point cloud model extraction; On this basis, identify the boundary of side landslides, and identify the development and location distribution of cracks;

In the step 1, carry out side landslide boundary identification, and the specific method for identifying crack development and location distribution is: compared with its local neighbors, there is a sudden change in slope at the steep slope or crack of the side landslide, and the DSM slope map identifies a larger and steeper slope. Kan, the DOM model identifies fractures developed near the main fractures at the trailing edge of the landslide.

Figure 2 shows the topographic slope map of a landslide generated based on DSM images for the identification of a landslide boundary (slope in the figure represents the slope). The slope distribution obtained in the study area is between 0° and 90°. The slope calculation results show that compared with the local neighborhood, the cracks and steep slopes on the perimeter of active landslides generally have a slope higher than 50°, which is related to the sudden change of surface cracks and scarps. Compared with its local neighbors, steep slopes or fissures have abrupt changes in slope (mutantly greater than 50°). By combining DOM and DSM data to identify trailing edge fissures and determine landslide boundaries, DSM slope maps can identify larger steep slopes , the DOM model can identify the small cracks that develop near the main cracks at the trailing edge of the landslide and are arranged in an approximately concentric arc shape on the plane. The final identification results of the landslide boundary are shown in Figure 3.

Step 2. Based on the 3D point cloud model obtained in step 1, filter the point cloud data by correcting the gravity direction iCSF method, separate and extract ground points, and check the relative position of the image control points to obtain the relative error of monitoring. The phase point cloud model is used for deformation monitoring to obtain the visualized displacement field of the side landslide;

Among them, the method of filtering point cloud data by correcting the iCSF method of gravity direction is as follows:

The CSF algorithm first inverts the 3D point cloud data, and then covers the inverted point cloud by simulating the "cloth", and generates an approximate surface according to the position of the "cloth" surface covering the point cloud under the action of gravity; by comparing the original point The distance between points in the cloud data and the surface of the generated "cloth", ground points are identified from the raw point cloud data, and distance points are separated for further feature extraction;

Simulated cloth According to Newton's second law, the relationship between cloth position and force is determined by the following formula:

In the formula, X represents the position of the particle at time t; F _ext (X,t) represents the external force, which is composed of gravity and the collision force generated by obstacles:

F _ext (X,t)＝mg+f _interact (X,t) (2)

When the particle encounters some objects in its moving direction; F _int (X,t) represents the internal force of the particle at position X and time t, and the internal force is generated by interacting with the point cloud (boundary); because both internal and external forces change with time t changes, so the above equations are usually solved by numerical integration (such as Euler's method) in the realization of cloth simulation;

The above method is suitable for gentle surfaces and is not friendly to the topographical characteristics of side landslides. Therefore, this project proposes a modified CSF method, changing the gravity direction of F _ext (X,t) from the vertical direction of the traditional CSF method to the direction of the slope. The direction perpendicular to the surface; this direction is determined by the median value of the slope of the DSM model; the steps are as follows:

(1) Obtain the median α of the regional slope based on the DSM model generated by the UAV photography technology, and determine the normal vector [N _x , N _y , N _z ] according to the median slope;

(2) The first gravitational item on the right side of the modified formula (2) g′=[N _x , N _y , N _z ]·9.81;

(3) Bring the corrected gravity term back to formula (2) to start filtering.

For deformation monitoring, Figure 4 shows the comparison results of the point cloud of a landslide on December 18, 2021 and the point cloud on December 23, 2021. The point cloud-based landslide deformation monitoring provides a visualized material loss in the surface change area ( collapse) or gain (build-up) situations. Under continuous blasting operations, the exposed position of the fault collapsed, causing a landslide in the strongly weathered soil overlying the fault, and finally three fan-shaped accumulations (#1, #2, #3) were accumulated on the bottom platform.

Step 3, please refer to Figure 5-Figure 10, based on the position of the crack determined in Step 1, use a crack monitoring device (combination of electronic vernier caliper and steel nails) to measure the development of the crack, and inspect the crack by manual/drone inspection The crack monitoring device takes regular readings and records; the specific method is: fix a steel nail at both ends of the crack to ensure the synchronization of the steel nail and the movement of the soil on both sides. On a steel nail, the crack development can drive the steel nail to move, thereby changing the reading of the electronic vernier caliper, and regularly reading and recording the crack meter vernier caliper through manual/drone inspection;

Step 4, when the crack develops to a certain stage and forms a clear landslide boundary, arrange monitoring points at positions reflecting the deformation characteristics of the side landslide, such as the front edge and the rear edge of the side landslide;

Step 5: Set up the target prism at the monitoring point, place the GPS receiver antenna, and set up the reference point in the area outside the side landslide (in step 5, the reference point is set in a stable area 30m away from the side landslide body). GPS receivers and GPS RTK reference station receivers are set up on the points, and the deformation of the monitoring points can be reversed through regular monitoring of the reference points in the stable area;

Step 6. Carry out dynamic and continuous monitoring of the side landslide area, the front edge of the side landslide, the rear edge of the side landslide, and the key position monitoring points (GPS RTK) and crack development (crack meter) in the middle of the side landslide, and record and identify the development and development of side landslide cracks in real time. Deformation status, preliminary identification of the development stage of side landslides

The concrete method of described step six is:

动态监测阶段夹角为

解算公式如下：Assume that the reference point P1 is arranged on the relatively stable side of the fracture, and the monitoring point P2 is arranged on the unstable side of the fracture. First, use RTK to locate the reference point P1 and monitoring point P2 to obtain the world coordinates (X ₁₀ , Y ₁₀ , Z ₁₀ ) of P1 and P2 points; then perform dynamic continuous monitoring on the reference point P1 and monitoring point P2 to obtain two points The world coordinates (X _1t , Y _1t , Z _1t ), (X _2t , Y _2t , Z _2t ), through the calculation, the data of the dislocation length l, fracture width d and dislocation height h during the fracture development process can be obtained, assuming The angle between the line connecting P1 and P2 and the x-axis is

The included angle in the dynamic monitoring stage is

The solution formula is as follows:

h＝Z _2t -Z _1t

Based on the data of dislocation length l, fissure width d and dislocation height h in the process of fracture development, the development stages of side landslides are initially identified. According to the data of dislocation length l, fissure width d and dislocation height h in the process of fracture development, the method for preliminary identification of the development stage of side landslide is as follows:

(1) For a side landslide without cracks, it is judged to be in a stable state;

(2) For those with a small amount of cracks but no obvious side landslide boundaries, it is judged to be in a relatively stable state;

(3) When the cracks continue to develop and form a clear edge landslide boundary, it is judged to be in a relatively unstable state;

(4) For the side landslides with bulging at the leading edge and dislocation of the trailing edge, it is judged to be in an unstable state. When the change rate of the three parameters of displacement length l, crack width d and dislocation height h is significantly increased, it is necessary to carry out early warning.

Step 7, further identify the state of the side landslide based on the development status of the side landslide cracks obtained in the comprehensive step 6 and the deformation status obtained in the step 2; the specific method is:

When there are no cracks in the side landslide area and the deformation is small, it is judged that the side landslide area is in a stable state; when some cracks are produced in the area, but the boundary has not yet formed and the deformation rate does not exceed 10mm/month, it is judged that the side landslide area is in a relatively stable state; Cracks continue to develop on the surface to form a clear boundary of the landslide and the deformation rate exceeds 10mm/month, which is judged to be in a relatively unstable state; in the local landslide area, the cracks at the trailing edge of the landslide are dislocated, and the leading edge bulges to appear feathery cracks, and the deformation rate exceeds 50mm/month month, the judgment area is in an unstable state.

To sum up, the identification methods of different stages of landslide stability state are summarized as follows:

1. The stage of no cracks in the slope: On the one hand, regular manual and UAV joint inspections are carried out to check whether the side landslides have cracks; Measure to obtain the control pile coordinates. The boundary of the side landslide is determined by the position of the crack, and the monitoring points of the front, middle, and rear edges of the side landslide are selected, and the deformation of the front, middle, and rear edge monitoring points is obtained by dynamic measurement through RTK.

2. Crack generation stage: adopt a simple crack monitoring device, that is, fix a steel nail at both ends of the crack, and then fix the initial scale end and free end of the electronic vernier caliper on the two steel nails, and the crack development will drive the steel nail to move. Therefore, the reading of the electronic vernier caliper is changed; on the other hand, the fracture is located through the DSM model gradient mutation and the DOM image model.

3. Crack development stage: regularly use the drone to carry the zoom camera to take pictures of the vernier caliper installed at the crack, and read the electronic reading of the vernier caliper to obtain the crack width. When the cracks develop to a certain extent, and the trailing edge of the side landslide has a dislocation, the height of the dislocation and the width of the crack can be obtained directly by solving the coordinates of the RTK monitoring points; on the other hand, the development of the crack can be quantified through the gradient mutation of the DSM model and the DOM image model Extract and judge the approximate range of the landslide.

4. The stage of bulging at the front edge and dislocation at the rear edge: when the crack develops to a certain extent, the front edge of the slope will bulge, and the rear edge of the slope will be dislocated. ; In terms of deformation, the deformation is extracted through GPS RTK monitoring and multi-temporal point cloud model change detection.

5. Recognition of the development stage of side landslides: Combining crack meter readings obtained by drones, 3D point cloud model deformation and deformation trends of key positions of side landslides obtained by RTK, to form a visualized side landslide deformation, damage development trend, and opposite side landslide development stage for preliminary judgment.

It is easy for those skilled in the art to understand that the above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention, etc., All should be included within the protection scope of the present invention.

Claims (8)

1. A side landslide stable state identification method, is characterized in that, comprises the steps: Step 1. Carry out aerial photography of side landslides at fixed time intervals through UAV photogrammetry, establish DOM models, DSM models and 3D point cloud models, and extract surface slope information based on DSM models; on this basis, carry out side landslide boundary recognition, identify Crack development and location distribution; Step 2. Based on the 3D point cloud model obtained in step 1, filter the point cloud data by correcting the gravity direction iCSF method, separate and extract ground points, and check the relative position of the image control points to obtain the relative error of monitoring. The phase point cloud model is used for deformation monitoring to obtain the visualized displacement field of the side landslide; Step 3. Based on the crack position determined in step 1, use a crack monitoring device to measure the crack development status, and perform regular readings and records on the crack monitoring device through manual/drone inspection; Step 4, when the crack develops to a certain stage and forms a clear landslide boundary, arrange monitoring points at positions reflecting the deformation characteristics of the side landslide, such as the front edge and the rear edge of the side landslide; Step 5: Set up the target prism at the monitoring point, place the GPS receiver antenna, and set up a reference point in the area other than the side landslide. The monitoring of the reference point can reverse the deformation of the fracture; Step 6. Carry out dynamic and continuous monitoring of the front edge of the side landslide, the rear edge of the side landslide, and the key position monitoring points and crack development in the middle of the side landslide in the area of the side landslide, record and identify the development and deformation of the side landslide cracks in real time, and the development stage of the side landslide carry out preliminary identification; Step 7, further identifying the state of the side landslide by combining the crack development status of the side landslide obtained in step 6 and the deformation status obtained in step 2. 2. the side landslide stable state identification method according to claim 1, is characterized in that, in described step 1, carry out side landslide boundary identification, the concrete method of identifying crack development and position distribution is: the scarp or crack place of side landslide Compared with its local neighbors, there are abrupt changes in the slope. The DSM slope map identifies larger scarps, and the DOM model identifies fractures developed near the main fractures at the trailing edge of the landslide. 3. side landslide stable state identification method according to claim 1, is characterized in that, described step 2 is by the method for filtering point cloud data by the iCSF method of correcting gravity direction: The CSF algorithm first inverts the 3D point cloud data, and then covers the inverted point cloud by simulating the "cloth", and generates an approximate surface according to the position of the "cloth" surface covering the point cloud under the action of gravity; by comparing the original point The distance between points in the cloud data and the surface of the generated "cloth", ground points are identified from the raw point cloud data, and distance points are separated for further feature extraction; Simulated cloth According to Newton's second law, the relationship between cloth position and force is determined by the following formula:

In the formula, X represents the position of the particle at time t; F _ext (X,t) represents the external force, which is composed of gravity and the collision force generated by obstacles: F _ext (X,t)＝mg+f _interact (X,t)(2) When the particle encounters some objects in its moving direction; F _int (X,t) represents the internal force of the particle at position X and time t, and the internal force is generated by interacting with the point cloud (boundary); because both internal and external forces change with time t change, so the above equations are solved by numerical integration (such as Euler's method) in the realization of cloth simulation; The above method is suitable for gentle surfaces, but not suitable for landslide terrain characteristics. Therefore, the gravity direction of F _ext (X, t) is changed from the vertical direction of the traditional CSF method to the direction perpendicular to the slope; this direction is passed The median value of the slope of the DSM model is determined; the steps are as follows: (1) Obtain the median α of the regional slope based on the DSM model generated by the UAV photography technology, and determine the normal vector [N _x , N _y , N _z ] according to the median slope; (2) The first gravitational item on the right side of the modified formula (2) g′=[N _x , N _y , N _z ]·9.81; (3) Bring the corrected gravity term back to formula (2) to start filtering. 4. side landslide stable state identification method according to claim 1, it is characterized in that, the concrete method of described step 3 is: fix a steel nail respectively at crack two ends, guarantee the distance between steel nail and soil mass movement on both sides Synchronization, the initial scale end and free end of the electronic vernier caliper are respectively fixed on two steel nails, the crack development can drive the steel nail to move, thereby changing the reading of the electronic vernier caliper, and the crack meter vernier caliper is regularly read through manual/drone inspection and record. 5. The method for identifying a stable state of a side landslide according to claim 1, wherein the reference point in the step 5 is arranged in a stable area 30m away from the side landslide body. 6. side landslide stable state identification method according to claim 1, is characterized in that, the concrete method of described step 6 is: Assume that the reference point P1 is arranged on the relatively stable side of the fracture, and the monitoring point P2 is arranged on the unstable side of the fracture. First, use RTK to locate the reference point P1 and monitoring point P2 to obtain the world coordinates (X ₁₀ , Y ₁₀ , Z ₁₀ ) of P1 and P2 points; then perform dynamic continuous monitoring on the reference point P1 and monitoring point P2 to obtain two points The world coordinates (X _1t , Y _1t , Z _1t ), (X _2t , Y _2t , Z _2t ), through the calculation, the data of the dislocation length l, fracture width d and dislocation height h during the fracture development process can be obtained, assuming The angle between the line connecting P1 and P2 and the x-axis is

The included angle in the dynamic monitoring stage is

The solution formula is as follows:

h＝Z _2t -Z _1t Based on the data of dislocation length l, fissure width d and dislocation height h in the process of fracture development, the development stages of side landslides are initially identified. 7. the side landslide stable state identification method according to claim 4, is characterized in that, described according to the dislocation length l, the crack width d and the dislocation height h data in the crack development process carry out preliminary identification to the side landslide development stage The method is: (1) For a side landslide without cracks, it is judged to be in a stable state; (2) For those with a small amount of cracks but no obvious side landslide boundaries, it is judged to be in a relatively stable state; (3) When the cracks continue to develop and form a clear edge landslide boundary, it is judged to be in a relatively unstable state; (4) For side landslides with bulging at the leading edge and dislocation of the trailing edge, it is judged to be in an unstable state; when the rate of change of the three parameters of dislocation length l, crack width d, and dislocation height h is significantly increased, it is necessary to carry out early warning. 8. according to claim 1-7 described side landslide stable state identification method described in any claim, it is characterized in that, in described step 7, the side landslide crack development situation that comprehensive step 6 obtains and the deformation state that step 2 obtains is opposite to the side The method for further identification of the state of the landslide is as follows: When there are no cracks in the side landslide area and the deformation is small, it is judged that the side landslide area is in a stable state; when there are some cracks in the area, but the boundary has not yet formed and the deformation rate does not exceed 10mm/month, it is judged that the side landslide area is in a relatively stable state; Cracks continue to develop on the surface to form a clear landslide boundary and the deformation rate exceeds 10mm/month, which is judged to be in a relatively unstable state; in the local landslide area, the cracks on the trailing edge of the landslide are dislocated, and the leading edge bulges to appear feathery cracks, and the deformation rate exceeds 50mm/month month, the judgment area is in an unstable state.

Images