A method and system for predicting land subsidence risk

Abstract

The present invention discloses a method and system for predicting the danger of ground subsidence, which relates to the technical field of geological disaster prediction. The monitoring area is divided, a plurality of divided areas are obtained, the historical ground subsidence soil layer pressure data, water flow data and vegetation index data of the divided areas are collected, and the data are divided at equal time intervals; each pressure settlement prediction model based on the historical ground subsidence soil layer pressure data, water flow data and vegetation index data is established; the weight of each settlement prediction model is calculated using a preset fuzzy hierarchical analysis, a combined settlement prediction model is established according to the weight, and the ground subsidence is predicted according to the combined settlement prediction model. The method and system for predicting the danger of ground subsidence provided by the present invention can accurately predict ground subsidence, thereby reducing the harm caused by ground subsidence to human life.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster prediction, and more specifically to a method and system for predicting the danger of ground subsidence. Background Art

[0002] Land subsidence is an irreparable permanent environmental and resource loss. The prominent problem of land subsidence has become an important factor restricting regional development.

[0003] At present, with the rapid development of urbanization, human demand for natural resources is increasing. The continuous exploitation of natural resources such as coal, oil, natural gas, and groundwater has made the problem of ground subsidence more and more serious. The high-speed and uneven subsidence has caused great harm to human life. The problem of ground subsidence has become a global geological problem. Therefore, the prediction and analysis of ground subsidence has important research value and practical significance for national space planning, disaster prevention and mitigation, etc.

[0004] Therefore, how to accurately predict ground subsidence and thus reduce the harm caused by subsidence to human life is an issue that technical personnel in this field urgently need to solve. Summary of the invention

[0005] In view of this, the present invention provides a method and system for predicting the danger of ground subsidence, which accurately predicts ground subsidence by studying vegetation index data, water flow data, and soil pressure data, thereby reducing the harm caused by ground subsidence to human life.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A method for predicting land subsidence hazard, comprising:

[0008] Step 1: Divide the monitoring area to obtain multiple divided areas, collect historical ground subsidence soil pressure data, historical ground subsidence water flow data and historical ground subsidence soil layer internal vegetation index data of the divided areas, and divide them into equal time intervals;

[0009] Step 2: Establish a pressure settlement prediction model based on historical ground subsidence soil layer pressure data; a water flow settlement prediction model based on historical ground subsidence water flow data; and a vegetation settlement prediction model based on historical ground subsidence soil layer internal vegetation index data;

[0010] Step 3: Calculate the weight of each settlement prediction model using a preset fuzzy hierarchical analysis method, establish a combined settlement prediction model based on the weight, and predict the ground settlement based on the combined settlement prediction model;

[0011] Step 4: Using SAR images, use PS-InSAR technology to obtain long-term ground subsidence information in the monitoring area, and verify the combined subsidence prediction model. If it does not meet the accuracy requirements, return to step 1 for recalibration;

[0012] Step 5: Use the combined settlement model verified and calibrated in step 4 to perform future prediction simulations and output the results.

[0013] Preferably, the division also includes separately dividing the area where soil fractures or cracks occur.

[0014] Preferably, the historical ground subsidence soil layer pressure data include the thickness and distribution of the compression layer at different burial depths of the soil layer; the historical ground subsidence water flow data include water flow extraction volume, precipitation and water flow level; the historical ground subsidence soil layer internal vegetation index includes root diameter, single root tensile strength and the number of roots on the shear collapse surface.

[0015] Preferably, the step 3 specifically includes:

[0016] The settlement prediction models are divided into five levels according to the pairwise comparison criteria;

[0017] Each level is represented by triangular fuzzy numbers to construct a fuzzy judgment matrix;

[0018] The fuzzy hierarchical analysis method is used to calculate the weights of each settlement prediction model.

[0019] Preferably, the pressure settlement prediction model in step 2 specifically includes:

[0020] Read the thickness of the compressed layer at different burial depths of the soil layer in the divided area, and establish the functional relationship between the thickness and the burial depth, as shown below:

[0021]

[0022] Where D is the burial depth of the soil layer, H i is the thickness of soil compression layer;

[0023] Establish a length prediction model and a width prediction model for the soil layers in the divided areas respectively, and draw the overall distribution of the resources inside the soil layers at different thicknesses and different burial depths based on the length prediction model and the width prediction model;

[0024] The length prediction model and width prediction model of the soil layer in the divided area are as follows:

[0025]

[0026]

[0027] Among them, D is the burial depth of the soil layer, H is the thickness of the resource layer inside the soil layer, Ly is the length of the soil layer, and Wy is the width of the soil layer.

[0028] Preferably, the water flow settlement prediction model in step 2 specifically includes that the dynamic change of groundwater level is affected by precipitation and the amount of local groundwater extraction. In years with large precipitation, the extraction amount is small and the groundwater level rises; in years with small precipitation, the extraction amount is large and the groundwater level shows a corresponding downward trend. The water flow settlement prediction model constructed in this way is:

[0029] y=b0+b1x1+b2x2

[0030] Among them, x1 is the precipitation, x2 is the water flow extraction, y is the water flow level, and b0, b1 and b2 are constants.

[0031] Preferably, the step 4 specifically includes: inputting the long-term land subsidence information obtained in the monitoring area into the combined subsidence prediction model for comparative verification, selecting RMSE, R 2 As a parameter to determine the simulation accuracy of the model, the feasibility of the model is verified. If it does not meet the accuracy requirements, return to step 1 to calibrate the model.

[0032] A land subsidence hazard prediction system, characterized by comprising:

[0033] The data collection module is used to divide the monitoring area, obtain multiple divided areas, collect historical ground subsidence soil layer pressure data, historical ground subsidence water flow data and historical ground subsidence soil layer internal vegetation index data of the divided areas, and divide them into equal time intervals;

[0034] Model building module, used to establish a pressure settlement prediction model based on historical ground settlement soil pressure data; a water flow settlement prediction model based on historical ground settlement water flow data; a vegetation settlement prediction model based on historical ground settlement soil internal vegetation index data;

[0035] A model combination module is used to calculate the weights of each settlement prediction model using a preset fuzzy hierarchical analysis, establish a combined settlement prediction model according to the weights, and predict the ground settlement according to the combined settlement prediction model;

[0036] A model verification module is used to obtain long-term ground subsidence information in the monitoring area through SAR images using PS-InSAR technology, and verify the combined subsidence prediction model. If it does not meet the accuracy requirements, it will be returned to the data collection module for recalibration;

[0037] The output module is used to perform future prediction simulations using the verified and calibrated combined settlement prediction model and output the results.

[0038] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a method and system for predicting the danger of ground subsidence, which reduces the amount of calculation of the model at one time by dividing the areas where soil fractures or cracks occur separately and predicting them by area and time period; and introduces vegetation index data, constructs three prediction models, calculates the weight of each prediction model, improves the accuracy of ground subsidence prediction, and can predict the occurrence of ground subsidence disasters in advance, thereby reducing the harm caused by ground subsidence to human life. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0040] Figure 1 A schematic diagram of the steps of the present invention is provided. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, an embodiment of the present invention discloses a method for predicting the danger of land subsidence, comprising:

[0043] Step 1: Divide the monitoring area, obtain multiple divided areas, collect historical ground subsidence soil pressure data, historical ground subsidence water flow data and historical ground subsidence soil layer internal vegetation index data in the divided areas, and divide them into equal time intervals;

[0044] Step 2: Establish a pressure settlement prediction model based on historical ground subsidence soil layer pressure data; a water flow settlement prediction model based on historical ground subsidence water flow data; and a vegetation settlement prediction model based on historical ground subsidence soil layer internal vegetation index data;

[0045] Step 3: Use the preset fuzzy hierarchical analysis to calculate the weight of each settlement prediction model, establish a combined settlement prediction model based on the weight, and predict the ground settlement based on the combined settlement prediction model;

[0046] Step 4: Using SAR images, use PS-I nSAR technology to obtain long-term ground subsidence information in the monitoring area, and verify the combined subsidence prediction model. If it does not meet the accuracy requirements, return to step 1 for recalibration;

[0047] Step 5: Use the combined settlement model verified and calibrated in step 4 to perform future prediction simulations and output the results.

[0048] In a specific embodiment, step 1 also includes preprocessing operations, which include: data cleaning, data filling, data feature expansion, data resampling, data normalization and data integration.

[0049] In a specific embodiment, in step 4, the PS-InSAR technology is used to obtain long-term ground subsidence information, which is completed through GAMMA, SARPROZ, and StaMPS software.

[0050] Among them, obtaining time-series ground subsidence information specifically includes: determining the main image based on the time-space baseline and Doppler information, and aligning and resampling other images as secondary images based on the main image; performing differential interference on each pair of main and secondary images to generate a time-series interference map and selecting PS points based on the amplitude deviation method; performing time-series analysis to obtain surface deformation information, comparing the results with leveling monitoring data and measured GPS data and verifying their accuracy, and obtaining ground subsidence information within the historical time period.

[0051] In a specific embodiment, the combined prediction model can obtain the limit value of ground subsidence based on big data, obtain various current data, bring them into each prediction model, and predict the limit value of ground subsidence according to the weight of each prediction model, thereby completing the prediction of ground subsidence by the combined prediction model.

[0052] In a specific embodiment, the division further includes separately dividing the area where soil fractures or cracks occur.

[0053] In a specific embodiment, the historical ground subsidence soil layer pressure data include the thickness and distribution of the compression layer at different burial depths of the soil layer; the historical ground subsidence water flow data include water flow extraction, precipitation and water flow level; the historical ground subsidence soil layer internal vegetation index includes root diameter, single root tensile strength and the number of roots on the shear collapse surface.

[0054] In a specific embodiment, step 3 specifically includes:

[0055] The settlement prediction models are divided into five levels according to the pairwise comparison criteria;

[0056] Each level is represented by triangular fuzzy numbers to construct a fuzzy judgment matrix;

[0057] The fuzzy hierarchical analysis method is used to calculate the weights of each settlement prediction model.

[0058] In a specific embodiment, the pressure drop prediction model in step 2 specifically includes:

[0059] Read the thickness of the compressed layer at different burial depths of the soil layer in the divided area, and establish the functional relationship between the thickness and the burial depth, as shown below:

[0060]

[0061] Where D is the burial depth of the soil layer, H i is the thickness of soil compression layer;

[0062] Establish a length prediction model and a width prediction model for the soil layers in the divided areas respectively, and draw the overall distribution of the resources inside the soil layers at different thicknesses and different burial depths based on the length prediction model and the width prediction model;

[0063] The length prediction model and width prediction model of the soil layer in the divided area are as follows:

[0064]

[0065]

[0066] Among them, D is the burial depth of the soil layer, H is the thickness of the resource layer inside the soil layer, Ly is the length of the soil layer, and Wy is the width of the soil layer.

[0067] In a specific embodiment, the water flow settlement prediction model in step 2 specifically includes the influence of precipitation on the dynamic change of groundwater level and the amount of local groundwater extraction. In years with large precipitation, the extraction amount is small and the groundwater level rises; in years with small precipitation, the extraction amount is large and the groundwater level shows a downward trend accordingly. The water flow settlement prediction model constructed in this way is:

[0068] y=b0+b1x1+b2x2

[0069] Among them, x1 is the precipitation, x2 is the water flow extraction, y is the water flow level, and b0, b1 and b2 are constants.

[0070] In a specific embodiment, step 4 specifically includes: inputting the long-term land subsidence information obtained in the monitoring area into the combined subsidence prediction model for comparative verification, selecting RMSE, R 2 As a parameter to determine the simulation accuracy of the model, the feasibility of the model is verified. If it does not meet the accuracy requirements, return to step 1 to calibrate the model.

[0071] A land subsidence hazard prediction system, characterized by comprising:

[0072] The data collection module is used to divide the monitoring area, obtain multiple divided areas, collect historical ground subsidence soil pressure data, historical ground subsidence water flow data and historical ground subsidence soil layer internal vegetation index data in the divided areas, and divide them into equal time intervals;

[0073] Model building module, used to establish a pressure settlement prediction model based on historical ground settlement soil pressure data; a water flow settlement prediction model based on historical ground settlement water flow data; a vegetation settlement prediction model based on historical ground settlement soil internal vegetation index data;

[0074] The model combination module is used to calculate the weight of each settlement prediction model using a preset fuzzy hierarchical analysis, establish a combined settlement prediction model according to the weight, and predict the ground settlement according to the combined settlement prediction model;

[0075] The model verification module is used to obtain the long-term ground subsidence information of the monitoring area through SAR images using PS-I nSAR technology, and verify the combined subsidence prediction model. If it does not meet the accuracy requirements, it will be returned to the data collection module for recalibration;

[0076] The output module is used to perform future prediction simulations using the verified and calibrated combined settlement prediction model and output the results.

[0077] In a specific embodiment, the vegetation settlement prediction model specifically includes:

[0078] Obtain the cohesion of the soil layer in the divided area, the inclination angle after the root system is deformed, the internal friction angle of the soil, and the shear collapse surface area of ​​the root-soil complex;

[0079] The effective cohesion of rooted soil is calculated according to the following formula:

[0080]

[0081] Where c is the effective cohesion of rooted soil, c s is the cohesion of soil layer in the division area; r is the inclination angle of the root system after deformation; is the internal friction angle; A s is the shear collapse surface area; d i is the i-th root diameter of the collapse surface; n i is the number of roots corresponding to the i-th root diameter of the collapse surface; T i is the single root tensile strength corresponding to the i-th root diameter of the collapse surface.

[0082] A convolutional neural network was established based on the ground subsidence data of historically divided areas and the effective cohesion of rooted soil, thereby constructing a vegetation subsidence prediction model.

[0083] Furthermore, based on the water-root-soil coupling mechanism, the root diameter and the single root tensile strength were measured by vernier caliper and pull-out test, respectively, to measure the characteristic indicators of the vegetation root system, including the root diameter d, the single root tensile strength T and the number of roots on the shear collapse surface n. The root diameter was measured by vernier caliper, and the single root tensile strength was obtained by pull-out test.

[0084] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0085] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may 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 predicting land subsidence risk, characterized in that: include: Step 1: Divide the monitoring area to obtain multiple divided areas, collect historical ground subsidence soil pressure data, historical ground subsidence water flow data and historical ground subsidence soil layer internal vegetation index data of the divided areas, and divide them into equal time intervals; Step 2: Establish a pressure settlement prediction model based on historical ground subsidence soil layer pressure data; a water flow settlement prediction model based on historical ground subsidence water flow data; and a vegetation settlement prediction model based on historical ground subsidence soil layer internal vegetation index data; Step 3: Calculate the weight of each settlement prediction model using a preset fuzzy hierarchical analysis method, establish a combined settlement prediction model based on the weight, and predict the ground settlement based on the combined settlement prediction model; Step 4: Using SAR images, use PS-InSAR technology to obtain long-term ground subsidence information in the monitoring area, and verify the combined subsidence prediction model. If it does not meet the accuracy requirements, return to step 1 for recalibration; Step 5: Use the combined settlement prediction model verified and calibrated in step 4 to perform future prediction simulations and output the results.

2. A method for predicting land subsidence risk according to claim 1, characterized in that: The division also includes separately dividing the area where soil fractures or cracks occur.

3. A method for predicting land subsidence risk according to claim 1, characterized in that: include: The historical ground subsidence soil layer pressure data include the thickness and distribution of the compression layer at different burial depths of the soil layer; the historical ground subsidence water flow data include water flow extraction, precipitation and water flow level; the historical ground subsidence soil layer internal vegetation index includes root diameter, single root tensile strength and the number of roots on the shear collapse surface.

4. A method for predicting land subsidence risk according to claim 1, characterized in that: The step 3 specifically includes: The settlement prediction models are divided into five levels according to the pairwise comparison criteria; Each level is represented by triangular fuzzy numbers to construct a fuzzy judgment matrix; The fuzzy hierarchical analysis method is used to calculate the weights of each settlement prediction model.

5. A method for predicting land subsidence risk according to claim 3, characterized in that: The pressure settlement prediction model in step 2 specifically includes: Read the thickness of the compressed layer at different burial depths of the soil layer in the divided area, and establish the functional relationship between the thickness and the burial depth, as shown below: Where D is the burial depth of the soil layer, is the thickness of soil compression layer; Establish a length prediction model and a width prediction model for the soil layers in the divided areas respectively, and draw the overall distribution of the resources inside the soil layers at different thicknesses and different burial depths based on the length prediction model and the width prediction model; The length prediction model and width prediction model of the soil layer in the divided area are as follows: Among them, D is the burial depth of the soil layer, H is the thickness of the resource layer inside the soil layer, Ly is the length of the soil layer, and Wy is the width of the soil layer.

6. A method for predicting land subsidence risk according to claim 3, characterized in that: The water flow settlement prediction model described in step 2 specifically includes the influence of precipitation on the dynamic change of groundwater level and the amount of local groundwater extraction. In years with heavy precipitation, the extraction amount is small and the groundwater level rises. In years with light precipitation, the extraction amount is large and the groundwater level shows a downward trend accordingly. The water flow settlement prediction model constructed in this way is: in, is the precipitation, is the water flow extraction volume, y is the water flow level, , as well as is a constant.

7. A method for predicting land subsidence risk according to claim 1, characterized in that: The step 4 specifically includes: inputting the long-term land subsidence information of the monitoring area into the combined subsidence prediction model for comparison and verification, selecting RMSE, R 2 As a parameter to determine the simulation accuracy of the model, the feasibility of the model is verified. If it does not meet the accuracy requirements, return to step 1 to calibrate the model.

8. A land subsidence hazard prediction system, characterized in that: include: The data collection module is used to divide the monitoring area, obtain multiple divided areas, collect historical ground subsidence soil layer pressure data, historical ground subsidence water flow data and historical ground subsidence soil layer internal vegetation index data of the divided areas, and divide them into equal time intervals; Model building module, used to establish a pressure settlement prediction model based on historical ground settlement soil pressure data; a water flow settlement prediction model based on historical ground settlement water flow data; a vegetation settlement prediction model based on historical ground settlement soil internal vegetation index data; A model combination module is used to calculate the weights of each settlement prediction model using a preset fuzzy hierarchical analysis, establish a combined settlement prediction model according to the weights, and predict the ground settlement according to the combined settlement prediction model; A model verification module is used to obtain long-term ground subsidence information in the monitoring area through SAR images using PS-InSAR technology, and verify the combined subsidence prediction model. If it does not meet the accuracy requirements, it will be returned to the data collection module for recalibration; The output module is used to perform future prediction simulation using the verified and calibrated combined settlement prediction model and output the results.

Images