A method and system for identifying potential hazards of a monoclinic gently inclined karst landslide

Abstract

The present invention discloses a method and system for identifying potential hazards of monoclinic gently-dipping karst landslides, which relates to the technical field of landslide potential hazard identification. Geological background parameters of the slope to be measured in the monoclinic gently-dipping karst area are obtained, including the magnitude relationship between the slope gradient and the formation dip angle, the included angle between the slope aspect and the formation strike, and the forward slope and reverse slope of the slope. According to the geological background parameters of the slope to be measured, the slope type of the slope to be measured is judged. Radar data for the whole year and a half is obtained, and the surface deformation data of the slope to be measured is extracted from the radar data. The slope type of the slope to be measured and the surface deformation data are input into the landslide prediction model, and then the probability of landslide disasters occurring is obtained. The present invention aims at the special geographical environment and geological characteristics of the monoclinic gently-dipping karst area, and at the same time combines the surface deformation information obtained from radar data, and can more accurately predict the probability of landslide disasters occurring in the monoclinic gently-dipping karst area.

Description

Technical Field

[0001] The invention relates to the technical field of landslide hidden danger identification, and more particularly to a method and system for identifying hidden dangers of monocline gently dipping karst landslides. Background Art

[0002] Monocline slow-dip karst landslide is an important and special type of landslide disaster. It is important because monocline slow-dip karst landslide has a wide range and great harm. The karst area has complex stratum lithology, and compound folds (which can be decomposed into multiple typical monocline slow-dip karst geological structures) are widely developed. The geological background for the occurrence of monocline slow-dip karst landslides is widely distributed. Once a landslide occurs in a monocline slow-dip limestone area, it is often large-scale and extremely harmful. It is special because the landslide mechanism of monocline slow-dip limestone landslide is special, highly concealed, and difficult to identify. Compared with soil landslides or other types of bedrock landslides, landslides in karst areas are more sudden, the surface shape before the landslide is small, the change characteristics of topography and landforms are not obvious, and the landslide mechanism is special. Under the action of tectonic-dissolution, karst areas are prone to develop steep slopes such as limestone escarpments. Most monocline slow-dip karst landslides occur at high positions, which are prone to form high-speed long-distance landslides, are highly concealed, and the technical difficulty of identifying landslide hazards is great.

[0003] At present, the method for identifying landslide hazard risks is a universal method. No targeted method has been proposed for the special mechanism of landslides in karst areas. The technical methods cannot keep up with the requirements of geological disaster prevention and control. Therefore, targeted identification of landslide hazard risks in monocline and gently sloping karst areas is an issue that technical personnel in this field urgently need to solve. Summary of the invention

[0004] In view of this, the present invention provides a method and system for identifying hidden dangers of monocline gently dipping karst landslides.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A method for identifying hidden dangers of monocline gently dipping karst landslides comprises the following steps:

[0007] Obtain the geological background parameters of the slope to be tested in the monocline gently dipping karst area;

[0008] Determine the slope type of the slope to be tested based on the geological background parameters of the slope to be tested;

[0009] Acquiring radar data within a preset time period, and extracting surface deformation data of the slope to be measured from the radar data;

[0010] The slope type and surface deformation data of the slope to be tested are input into the landslide prediction model to obtain the probability of landslide disaster occurrence.

[0011] Optionally, the geological background parameters include the magnitude relationship between the slope gradient and the formation dip angle, the included angle between the slope aspect and the formation strike, and the forward slope and reverse slope of the slope.

[0012] Optionally, the method for determining the slope type of the slope to be measured is specifically as follows:

[0013] When the slope gradient is greater than the formation dip angle, if the included angle between the slope aspect and the formation strike is less than or equal to 36°, the slope to be measured is a floating forward slope; if the included angle between the slope aspect and the formation strike is between 36° and 72°, the slope to be measured is a floating forward slope; if the included angle between the slope aspect and the formation strike is between 72° and 108°, the slope to be measured is a floating transverse slope; if the included angle between the slope aspect and the formation strike is between 108° and 144°, the slope to be measured is a floating reverse slope; if the included angle between the slope aspect and the formation strike is between 144° and 180°, the slope to be measured is a floating reverse slope;

[0014] When the slope gradient is less than the formation dip angle, if the included angle between the slope aspect and the formation strike is less than or equal to 36°, the slope to be measured is a lying forward slope; if the included angle between the slope aspect and the formation strike is between 36° and 72°, the slope to be measured is a lying forward slope; if the included angle between the slope aspect and the formation strike is between 72° and 108°, the slope to be measured is a lying transverse slope; if the included angle between the slope aspect and the formation strike is between 108° and 144°, the slope to be measured is a lying reverse slope; if the included angle between the slope aspect and the formation strike is between 144° and 180°, the slope to be measured is a lying reverse slope.

[0015] Optionally, the preset time period is selected as one and a half years.

[0016] Optionally, the surface deformation data includes the surface deformation rate and the surface deformation amount.

[0017] Optionally, the method for extracting the surface deformation data of the slope to be measured from the radar data is as follows:

[0018] Obtain the radar data within the preset time period;

[0019] Using the SAR image data in the radar data, based on the SABS-InSAR technology, obtain the deformation rate and deformation amount of the deformation point along the radar line of sight direction;

[0020] Using the DEM data, convert the deformation rate and deformation amount of the deformation point along the radar line of sight direction into the surface deformation rate and surface deformation amount along the vertical direction.

[0021] Optionally, the training process of the landslide prediction model is as follows:

[0022] Obtain the landslide samples in the monoclinic gently dipping karst area and the non-landslide samples with the same proportion as the landslide samples;

[0023] Extract the slope type features and surface deformation data features in landslide samples and non-landslide samples, assign landslide labels to landslide samples, and assign non-landslide labels to non-landslide samples; use the landslide samples and non-landslide samples with slope type features, surface deformation data features, landslide labels, and non-landslide labels as the training and test data sets;

[0024] Import the training and test data set into a random forest model for training to generate a trained random forest model as the landslide prediction model.

[0025] Optionally, according to the label results output by each sub-decision tree in the landslide prediction model, use the ratio of the number of sub-decision trees outputting landslide labels to the total number of sub-decision trees as the probability of landslide disasters occurring.

[0026] A hidden danger identification system for monoclinic gently dipping karst landslides, comprising:

[0027] A geological background parameter acquisition module, configured to acquire the geological background parameters of the slope to be measured in the monoclinic gently dipping karst area;

[0028] A slope type determination module, configured to determine the slope type of the slope to be measured according to the geological background parameters of the slope to be measured;

[0029] A surface deformation data acquisition module, configured to acquire radar data within a preset time period and extract the surface deformation data of the slope to be measured from the radar data;

[0030] A probability prediction module, configured to input the slope type and surface deformation data of the slope to be measured into the landslide prediction model to obtain the probability of landslide disasters occurring.

[0031] Through the above technical solutions, the present invention provides a hidden danger identification method and system for monoclinic gently dipping karst landslides in view of the special geographical environment and geological characteristics of monoclinic gently dipping karst. Compared with the prior art, the present invention has the following beneficial effects:

[0032] (1) In view of the special geographical environment and geological characteristics of the monoclinic gently dipping karst area, the present invention selects geological background parameters to break through the limitations of traditional single-factor analysis such as slope, aspect, stratigraphic lithology, and geological structure, and directly divides the characteristic geological background suitable for the karst area into slope structures according to the size relationship between the slope gradient and the stratigraphic dip angle and the correlation relationship between the slope aspect and the stratigraphic dip direction. At the same time, combined with the surface deformation information obtained from radar data, the probability of landslide disasters occurring in the monoclinic gently dipping karst area can be predicted more accurately.

[0033] (2) Determine the time period selection rule for radar data suitable for the karst area, which must be radar data for one and a half years, to eliminate the influence of periodic deformation in the karst area. Description of the Drawings

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on the provided drawings.

[0035] Figure 1 It is a schematic diagram of the method steps of the present invention;

[0036] Figure 2 It is a schematic diagram of the system structure of the present invention. Detailed implementation manners

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention 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 of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0038] The rock stratum structure in the monoclinic gently dipping karst area usually shows an obvious continuous non-uniform thickness interbedded (intercalated) structure of hard and soft rocks. Especially, thin mudstone, expansive soil, gypsum, etc. are widely developed between medium-thick hard limestone, which is a rock stratum structure prone to landslides.

[0039] Due to the solubility of carbonate rocks, the contact zone (main sliding zone) between limestone and soft rock strata such as mudstone is usually a strong runoff zone second only to karst pipelines. The water content of soft rock strata such as mudstone is quickly saturated, and the lubrication effect of groundwater is obvious; in addition, the water content of the karst aquifer changes dynamically on the time scale by dozens of times that of the soft rock strata. The soft rock in the contact zone (main sliding zone) is in an extreme state of continuous conversion between dry and wet, and the groundwater characteristics significantly exacerbate the sliding property of the monoclinic gently dipping karst area.

[0040] According to the relationship between the dip angle of the stratum and the slope, there are two types of slopes in the monoclinic gently dipping karst area. When the dip angle is greater than the slope of the slope, the forward slope, reverse slope, and oblique slope of the slope are not likely to have large-scale overall sliding. When the dip angle is less than the slope of the slope, the oblique slope of the slope is not likely to have large-scale overall sliding; for the forward slope, due to the cutting at the front edge (high position), a free face is formed, providing the basic topographic conditions for overall sliding; for the reverse slope, under the condition that the slope is nearly vertical (limestone steep cliffs are widely developed in the karst area), it is easy to trigger chain landslides by high-position collapses.

[0041] Karstification has two aspects of influence on landslide disasters. One is that the joints and fissures in the gently inclined monoclinic karst area are extremely easy to penetrate (especially the nearly vertical joints and fissures), which accelerates the cracking of carbonate rocks in the gently inclined monoclinic karst area and promotes the occurrence of landslides in the gently inclined monoclinic karst area. The other is that karstification requires the participation of water. The seasonal cycle of water makes the deformation in the karst area have hydrological periodicity, resulting in the cumulative deformation over the years being easily offset by positive and negative values.

[0042] Based on the above special geographical environment and geological characteristics of the gently inclined monoclinic karst, the embodiments of the present invention disclose a method for identifying potential hazards of landslides in the gently inclined monoclinic karst. See Figure 1 , including the following steps:

[0043] Step 1: Obtain the geological background parameters of the slope to be measured in the gently inclined monoclinic karst area, including the magnitude relationship between the slope gradient and the formation dip angle, the included angle between the slope aspect and the formation strike, and the forward slope and reverse slope of the slope.

[0044] In specific situations, the slope gradient can be obtained by a slope gradient measuring instrument, and the formation dip angle can be obtained by a formation dip angle measuring scale.

[0045] Step 2: Based on the geological background parameters of the slope to be measured, determine the slope type of the slope to be measured. See Table 1:

[0046] Table 1

[0047]

[0048] Step 3: Obtain radar data within a preset time period, and extract the surface deformation data of the slope to be measured from the radar data; the preset time period is selected as one and a half years. In specific embodiments, it can be selected as 1.5 years, 2.5 years, etc. to eliminate the influence of periodic deformation in the karst area;

[0049] In specific embodiments, the surface deformation data includes the surface deformation rate and the surface deformation amount.

[0050] Correspondingly, the method for extracting the surface deformation data of the slope to be measured from the radar data is as follows:

[0051] Step 3.1: Obtain radar data within a preset time period;

[0052] Step 3.2: Utilize the SAR image data in the radar data, and based on the SABS-InSAR technology, obtain the deformation rate and deformation amount of the deformation point along the radar line of sight direction;

[0053] Step 3.3: Utilize the DEM data to convert the deformation rate and deformation amount of the deformation point along the radar line of sight direction into the surface deformation rate and surface deformation amount along the vertical direction.

[0054] Step 4: Input the slope type of the slope to be measured and the surface deformation data into the landslide prediction model. According to the label results output by each sub-decision tree in the landslide prediction model, take the ratio of the number of sub-decision trees that output the landslide label to the total number of sub-decision trees as the probability of the occurrence of the landslide disaster.

[0055] In a specific embodiment, the training process of the landslide prediction model is as follows:

[0056] Step 4.1: Obtain landslide samples in a monoclinic gently dipping karst area and non-landslide samples with the same proportion as the landslide samples;

[0057] Step 4.2: Extract the slope type features and surface deformation data features from the landslide samples and non-landslide samples, and assign landslide labels to the landslide samples and non-landslide labels to the non-landslide samples; Take the landslide samples and non-landslide samples with slope type features, surface deformation data features, landslide labels, and non-landslide labels as the training and test data sets.

[0058] Step 4.3: Import the training and test data set into the random forest model for training to generate a trained random forest model as the landslide prediction model.

[0059] In other embodiments, the method further includes a warning step for giving a warning when the probability of the occurrence of the landslide disaster exceeds a preset threshold.

[0060] In another embodiment, a hidden danger identification system for monoclinic gently dipping karst landslides is also disclosed. See Figure 2 , including:

[0061] A geological background parameter acquisition module for acquiring the geological background parameters of the slope to be measured in a monoclinic gently dipping karst area;

[0062] A slope type determination module for determining the slope type of the slope to be measured according to the geological background parameters of the slope to be measured;

[0063] A surface deformation data acquisition module for acquiring radar data within a preset time period and extracting the surface deformation data of the slope to be measured from the radar data;

[0064] A probability prediction module for inputting the slope type of the slope to be measured and the surface deformation data into the landslide prediction model to obtain the probability of the occurrence of the landslide disaster.

[0065] In other embodiments, the system further includes a warning module for giving a warning of the landslide disaster.

[0066] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other. For the systems and devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For the relevant parts, reference can be made to the description in the method section.

[0067] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying hidden dangers of monoclinic gently inclined karst landslides, characterized in that, It includes the following steps: Obtain the geological background parameters of the slope to be measured in the monoclinic gently inclined karst area; Judge the slope type of the slope to be measured according to the geological background parameters of the slope to be measured; Obtain radar data within a preset time period, and extract the surface deformation data of the slope to be measured from the radar data; Input the slope type and surface deformation data of the slope to be measured into the landslide prediction model to obtain the probability of landslide disaster occurrence; The method for extracting the surface deformation data of the slope to be measured from the radar data is as follows: Obtain radar data within a preset time period; Utilize the SAR image data in the radar data, and based on the SABS-InSAR technology, obtain the deformation rate and deformation amount of the deformation points along the radar line of sight direction; Utilize the DEM data to convert the deformation rate and deformation amount of the deformation points along the radar line of sight direction into the surface deformation rate and surface deformation amount along the vertical direction.

2. The method for identifying potential hazards of a monoclinic gently inclined karst landslide according to claim 1, wherein The geological background parameters include the magnitude relationship between the slope gradient and the formation dip angle, the included angle between the slope aspect and the formation strike, and the forward slope and reverse slope of the slope.

3. The method for identifying hidden dangers of a monoclinic gently inclined karst landslide according to claim 2, characterized in that, The judgment of the slope type of the slope to be measured is specifically as follows: When the slope gradient is greater than the formation dip angle, if the included angle between the slope aspect and the formation strike is less than or equal to 36°, the slope to be measured is a floating forward slope; if the included angle between the slope aspect and the formation strike is between 36° and 72°, the slope to be measured is a floating forward slope; if the included angle between the slope aspect and the formation strike is between 72° and 108°, the slope to be measured is a floating transverse slope; if the included angle between the slope aspect and the formation strike is between 108° and 144°, the slope to be measured is a floating reverse slope; if the included angle between the slope aspect and the formation strike is between 144° and 180°, the slope to be measured is a floating reverse slope; When the slope gradient is less than the formation dip angle, if the included angle between the slope aspect and the formation strike is less than or equal to 36°, the slope to be measured is a lying forward slope; if the included angle between the slope aspect and the formation strike is between 36° and 72°, the slope to be measured is a lying forward slope; if the included angle between the slope aspect and the formation strike is between 72° and 108°, the slope to be measured is a lying transverse slope; if the included angle between the slope aspect and the formation strike is between 108° and 144°, the slope to be measured is a lying reverse slope; if the included angle between the slope aspect and the formation strike is between 144° and 180°, the slope to be measured is a lying reverse slope.

4. A method for identifying potential hazards of a monoclinic gently dipping karst landslide according to claim 1, characterized in that, The preset time period is selected as one and a half years.

5. A method for identifying potential hazards of a monoclinic gently-dipping karst landslide according to claim 1, characterized in that The surface deformation data includes the surface deformation rate and the surface deformation amount.

6. The hidden danger identification method for a monoclinic gently inclined karst landslide according to claim 1, wherein, The training process of the landslide prediction model is as follows: Obtain the landslide samples in the monoclinic gently inclined karst area and non-landslide samples with the same proportion as the landslide samples; Extract the slope type features and surface deformation data features in the landslide samples and non-landslide samples, and assign landslide labels to the landslide samples and non-landslide labels to the non-landslide samples; use the landslide samples and non-landslide samples with slope type features, surface deformation data features, landslide labels, and non-landslide labels as the training and test data sets; Import the training and test data sets into the random forest model for training to generate a trained random forest model as the landslide prediction model.

7. A method for identifying potential hazards of a monoclinic gently inclined karst landslide according to claim 6, characterized in that, According to the label results output by each sub - decision tree in the landslide prediction model, the ratio of the number of sub - decision trees that output the landslide label to the total number of sub - decision trees is used as the probability of landslide disaster occurrence.

8. A hidden danger identification system for a monoclinic gently inclined karst landslide, characterized in that, It includes: A geological background parameter acquisition module, which is used to acquire the geological background parameters of the slope to be measured in the monoclinic gently - dipping karst area; A slope type determination module, which is used to determine the slope type of the slope to be measured according to the geological background parameters of the slope to be measured; A surface deformation data acquisition module, which is used to acquire radar data within a preset time period and extract the surface deformation data of the slope to be measured from the radar data; A probability prediction module, which is used to input the slope type and surface deformation data of the slope to be measured into the landslide prediction model to obtain the probability of landslide disaster occurrence; The method for extracting the surface deformation data of the slope to be measured from the radar data is as follows: Acquire radar data within a preset time period; Using the SAR image data in the radar data, based on the SABS - InSAR technology, obtain the deformation rate and deformation amount of the deformation point along the radar line - of - sight direction; Using DEM data, convert the deformation rate and deformation amount of the deformation point along the radar line - of - sight direction into the surface deformation rate and surface deformation amount along the vertical direction.

Images