A landslide early warning method based on GB-InSAR monitoring and deformation region expansion

Abstract

The application discloses a landslide early warning method based on GB-InSAR monitoring and deformation area expansion. The application utilizes the technical advantage that GB-InSAR can realize surface scene deformation monitoring, and based on the feature that the deformation area will present regular expansion in the development process of the landslide, proposes an early warning method based on an area-time curve. By setting multiple deformation thresholds, the calculation results of the tangent angles of multiple deformation-time curves are fused, and the tangent angles are used for distinguishing the landslide early warning stage. Compared with the traditional early warning method based on single-point deformation information, the method using the area-time curve can identify the overall change trend of the landslide surface deformation, the early warning information is more comprehensive and accurate, and the accuracy of distinguishing the landslide early warning stage can be effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of alarm in response to disaster events, and particularly relates to a landslide early warning method based on GB-InSAR monitoring and deformation region expansion. BACKGROUND

[0002] At present, the landslide disaster early warning method is mainly based on the contact type sensor on the landslide body, such as: ground GNSS, inclinometer, soil pressure box, crack meter, etc. The contact type equipment is scattered on the target landslide body, and can only monitor the stress and strain information of a small number of point positions, and there are still a large number of information blind areas on the landslide body, which is easy to cause false negatives or false positives in actual application. The mainstream landslide early warning method at present is the deformation time curve based on ground GNSS monitoring, and the tangent angle method is used to distinguish the landslide early warning stage. This early warning method has two obvious defects: 1) the information source is a single point deformation time curve, which can only effectively represent the deformation state of the ground surface near the monitoring equipment, and does not necessarily represent the deformation characteristics of the entire target area; 2) the GNSS monitoring station is installed on the slope before the landslide occurs, which cannot ensure that it is installed in the strong deformation area when the landslide occurs, and is easy to cause false negatives. Therefore, the early warning method based on ground GNSS monitoring and deformation time curve has uncertainty, and it is difficult to ensure the accuracy of early warning.

[0003] The slope deformation evolution often starts from a local point, and the deformation time curve based on single point monitoring can better identify the deformation trend of the point in the slope deformation process, and the dangerous state of the point can be distinguished through deformation acceleration and deformation curve tangent angle. When the slope has a significant damage, the slope releases a large internal stress, and the deformation speed will slow down, but the deformation area may still continue to expand to form a secondary landslide risk, at this time, the single point deformation time curve can only identify the slow deformation and make a low early warning mistake. SUMMARY

[0004] Therefore, the present application provides a landslide early warning method based on GB-InSAR monitoring and deformation region expansion, which uses the area-time curve to calculate the tangent angle, and can realize accurate discrimination of the landslide early warning stage.

[0005] The landslide early warning method based on GB-InSAR monitoring and deformation region expansion of the present application comprises:

[0006] Step 1, continuously and real-time monitoring the scene by using GB-InSAR, and obtaining the cumulative deformation of the monitoring scene in the current measurement period; according to the maximum value in the cumulative deformation, designing a plurality of deformation thresholds;

[0007] Step 2, obtaining the area-time curve:

[0008] For each deformation threshold, the total area of pixel points whose accumulated deformation exceeds the deformation threshold is calculated in real time at the current time; each time corresponds to a total area of pixel points exceeding the deformation threshold, and with the passage of time, an area-time curve is formed;

[0009] Step 3, the tangent angle of the area-time curve corresponding to each deformation threshold at the current time is calculated respectively;

[0010] Step 4, the maximum of the tangent angles of the area-time curves corresponding to each deformation threshold is the early warning decision tangent angle, which is matched with the set early warning level to perform early warning.

[0011] Preferably, in step 1, the N deformation thresholds are designed as follows:

[0012] The maximum accumulated deformation value S in the current measurement period in the monitored scene max ≤100mm, The deformation threshold S k =10·k, k=1~N, Indicates rounding down;

[0013] The maximum accumulated deformation value S in the current measurement period in the monitored scene max >100mm, The deformation threshold

[0014] Preferably, in step 3, the tangent angle of the area-time curve A k (t) corresponding to the deformation threshold S k is calculated as follows:

[0015] First, the average speed B k from the initial time t1 to each time t i of the area-time curve A k,i (t) is calculated, and then the ordinate A k,i of the area-time curve is transformed into T k,i with the same dimension as the abscissa time t by dividing the area by the average speed; the slope of T k,i is the tangent angle a k of the area-time curve A i (t) at t k,i .

[0016] Preferably, the tangent angle a k,i is:

[0017]

[0018] Where, Δt is the time window in the tangent angle calculation.

[0019] Preferably, the Δt is set as a fixed value according to the deformation characteristics of the monitoring target, or is adaptively determined by the variation degree of the tangent angle curve under different time windows Δt.

[0020] Preferably, in the step 4, the warning levels are four: attention level θ1≤α<θ2, warning level θ2≤α<θ3, alert level θ3≤α<θ4 and alarm level α≥θ4; wherein θ1∈[35°, 45°], θ2∈[40°, 50°], θ3∈[60°, 80°], θ4∈[70°, 90°], and θ1<θ2<θ3<θ4 are satisfied; and α is the tangent angle of the warning decision.

[0021] Beneficial effects:

[0022] The present application has the technical advantage of realizing surface scene deformation monitoring by using GB-InSAR, and based on the regular expansion characteristics of the deformation area in the landslide development process, proposes a warning method based on the area-time curve. By setting multiple deformation thresholds, the tangent angle calculation results of multiple deformation time curves are fused, and the landslide warning stage is discriminated based on the tangent angle. Compared with the traditional single-point deformation information warning method, the area-time curve method can identify the overall change trend of the landslide surface deformation, and the warning information is more comprehensive and accurate, which can effectively improve the accuracy of the landslide warning stage discrimination.

[0023] GB-InSAR has the measurement advantages of all-weather, all-day, non-contact and near real-time, can realize regional surface scene deformation measurement, and to a great extent, overcomes the two defects of GNSS monitoring, and can realize accurate warning of landslide disasters. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The area-time curve corresponding to the multiple deformation thresholds.

[0025] Figure 2 The tangent angle-time curve corresponding to the multiple area-time curves.

[0026] Figure 3 The warning decision tangent angle-time curve.

[0027] Figure 4 The landslide warning level.

[0028] Figure 5 The method flowchart of the present application. DETAILED DESCRIPTION

[0029] The present application will be described in detail below with reference to the accompanying drawings and examples.

[0030] The application provides a landslide early warning method based on GB-InSAR monitoring and deformation region expansion, and a method flow is as shown in Figure 5 The method specifically comprises the following steps.

[0031] Step one, design deformation threshold

[0032] Continuous real-time monitoring is performed on the scene by using GB-InSAR, and the cumulative deformation of the monitored scene in the current measurement period is obtained. According to the current measurement period, the maximum cumulative deformation value S max in the monitored scene is obtained, and N deformation thresholds are designed.

[0033] The number and value of the deformation threshold can be determined by experience or by numerical simulation. In this embodiment, the following design is adopted.

[0034] When S max ≤100mm, the deformation threshold S k =10·k, k=1~N, and the down-rounding is represented.

[0035] When S max >100mm, the deformation threshold S

[0036] Step two, obtain the area-time curve

[0037] The deformation time curve based on single-point monitoring can well identify the landslide danger state of the point in the slope deformation process. However, after the slope has been damaged once, the deformation speed of the single point may slow down, but the deformation region may still continue to expand. The area-time curve method can more accurately identify the overall deformation trend of the landslide surface.

[0038] For the cumulative deformation in the current measurement period, for each deformation threshold S k , the total area A k,i of the pixel points whose cumulative deformation at each time t i in the measurement period exceeds S k is calculated, and the collection of all A k,i in the measurement period is the area-time curve A k (t). In the landslide development process, A k (t) is an increasing curve, and the increasing speed of A k (t) is the expansion speed of the deformation region.

[0039] Step three, calculate the tangent angle of the area-time curve

[0040] For each deformation threshold S kCorresponding area-time curve A k (t), calculate the area-time curve A k (t) from the initial time t1 to each time t i The average speed B k，i (Formula (1)).

[0041]

[0042] In the way of dividing the area by the average speed, the ordinate A k,i of the area-time curve is transformed into T k，i with the same dimension as the abscissa time t (Formula (2)).

[0043]

[0044] Calculate the tangent angle (slope) α i at time t k,i (Formula (3)), α k,i is a dimensionless natural number.

[0045]

[0046] In formula (3), Δt is the time window in the tangent angle calculation, which can be set as a fixed value according to the deformation characteristics of the monitoring target, or adaptively determined by the degree of change of the tangent angle curve under different time windows Δt.

[0047] Step four, obtain the early warning decision tangent angle

[0048] At time t i , each deformation threshold S k corresponds to a tangent angle α k,i , and N deformation thresholds can obtain N tangent angles. The maximum tangent angle α i is taken as the early warning decision tangent angle (Formula (4)).

[0049]

[0050] Step five, generate early warning decision results

[0051] According to the type of landslide, set four tangent angle thresholds (θ1, θ2, θ3, θ4) to associate the tangent angle with the early warning level (Table 1), so that the early warning decision results can be generated according to the value of the decision tangent angle α i .

[0052] Generally, the tangent angle threshold setting range is: θ1 ∈ [35°, 45°], θ2 ∈ [40°, 50°], θ3 ∈ [60°, 80°], θ4 ∈ [70°, 90°], and θ1 < θ2 < θ3 < θ4 is satisfied.

[0053] Table 1. Correspondence between tangent angle and warning level

[0054]

[0055] Case

[0056] Taking the slope of the Jianshan Iron Mine in Shanxi Province as an example, ground-based interferometric radar was used to monitor the slope deformation process, acquiring deformation time curves for 3016 pixels within the scene. The early warning calculation process is as follows:

[0057] 1) The maximum cumulative deformation value in the current scene is 3103.9 mm. 40 deformation thresholds are designed (10, 20, 30, ..., 100, 200, 300, ..., 3100 mm), and 40 corresponding area time-series curves are obtained. Figure 1 ).

[0058] 2) Based on formulas (1) to (3), using a 5-hour time window, calculate... Figure 1 The tangent angle time curve corresponding to each area time curve is used to obtain multiple tangent angle time curves. Figure 2 ).

[0059] 3) Figure 2 Each moment corresponds to multiple tangent angle values. The maximum value is selected as the warning decision tangent angle, resulting in the warning decision tangent angle time curve. Figure 3 ).

[0060] 4) Tangent angle thresholds θ1, θ2, θ3, and θ4 are set to 40°, 50°, 75°, and 85° respectively. These four thresholds are substituted into Table 1 for early warning decisions. The early warning decisions for all times are as follows: Figure 4 As shown.

[0061] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A landslide early warning method based on GB-InSAR monitoring and deformation area expansion, characterized in that, Comprise: Step 1, using GB-InSAR to continuously monitor the scene in real time, and obtain the cumulative deformation of the monitoring scene in the current measurement period; According to the maximum value in the cumulative deformation, design multiple deformation thresholds; Step 2, obtain the area-time curve: For each deformation threshold, real-time calculate the total area of the pixel points whose cumulative deformation exceeds the deformation threshold at the current time; each time corresponds to a total area of pixel points exceeding the deformation threshold, and with the passage of time, an area-time curve is formed; Step 3, calculate the tangent angle of the area-time curve corresponding to each deformation threshold at the current time respectively; Step 4, the maximum of the tangent angle of the area-time curve corresponding to each deformation threshold is the early warning decision tangent angle, which is matched with the set early warning level to perform early warning; In step 1, the following method is used to design N a deformation threshold: Maximum accumulated deformation value in current measurement period monitoring scenario S max ≤ 100 mm, , deformation threshold , denotes rounding down; Maximum accumulated deformation value in current measurement period monitoring scenario S max > 100 mm, , deformation threshold .

2. The landslide warning method based on GB-InSAR monitoring and deformation area expansion of claim 1, wherein, In step 3, the deformation threshold The corresponding area-time curve The tangent angle calculation method is as follows: First, the area-time curve is calculated The average velocity at each time is calculated Then, the ordinate of the area-time curve is converted to the same dimension as the abscissa time t by dividing the area by the average velocity ; The slope of the line at time is the tangent angle of the area-time curve at that time.

3. The landslide warning method based on GB-InSAR monitoring and deformation area expansion of claim 2, wherein, tangent angle is: wherein, i.e. the time window for the tangent angle calculation.

4. The landslide warning method based on GB-InSAR monitoring and deformation area expansion of claim 3, wherein, The According to the deformation characteristics of the monitoring target, the setting is a fixed value, or through different time windows The variation degree of the lower tangent angle curve is adaptively determined.

5. The landslide warning method based on GB-InSAR monitoring and deformation area expansion of claim 1, wherein, The step 4, the warning level is 4 kinds: attention level , warning level , alert level And alarm level ; wherein, , , , , while satisfying ; The tangent angle of the early warning decision.

Images