Landslide disaster intelligent early warning method based on water and soil loss monitoring

Through the neural network model combined with soil erosion indicators, a slope-one model was established, which solved the problem of unconsidered linkage between soil erosion and landslide disasters in slope monitoring and early warning in the existing technology, improved the prediction accuracy and accuracy, and achieved more To meet the actual landslide disaster warning.

CN119049226APending Publication Date: 2024-11-29CHINA MCC5 GROUP CORP LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411148813.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing slope monitoring and early warning fails to effectively consider the internal linkage between soil erosion and landslide disasters, resulting in insufficient accuracy and accuracy of early warnings.

Method used

Using intelligent landslide disaster warning method based on neural network model, data samples are generated through indoor model experiments and numerical simulations, model is established based on slope geological information, neural network is trained, and real-time warning is performed using on-site monitoring data, and soil erosion indicators such as rainfall are considered Infiltration amount and runoff sediment content.

Benefits of technology

The accuracy and accuracy of predictions have been improved, and one slope and one model has been built to fit the actual slope early warning model, making full use of on-site monitoring resources, and enhancing the perfection and economic benefits of the monitoring system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049226A_ABST
    Figure CN119049226A_ABST
Patent Text Reader

Abstract

The invention discloses a landslide disaster intelligent early warning method based on water and soil loss monitoring. The method comprises the following steps: acquiring basic information characteristics of a side slope; an indoor model test: carrying out tests of the slope indoor model under different rainfall scenes to obtain test data; numerical simulation: simulating dynamic response conditions of the numerical analysis model in different rainfall scenes to obtain simulation data; neural network training is conducted through the sample data set, and the neural network model capable of mapping the implicit relation among the rainfall infiltration amount, the water and soil loss degree, the slope displacement, the slope safety coefficient and the instability range is established; field monitoring; and disaster early warning. The beneficial effects of the invention are that a one-slope one-model idea is provided, and a data sample is established for a specific slope to construct an early warning model, so that the prediction accuracy and precision can be effectively improved; and an internal linkage rule of water and soil loss and landslide disasters is considered, so that the prediction model better fits an actual slope.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of geological disaster early warning technology, and specifically relates to an intelligent early warning method for landslide disasters based on soil and water loss monitoring, which is mainly applied to soil slopes under rainfall conditions, and provides early warning and forecasting of landslide disasters based on soil and water loss monitoring of the slope body. Background Art

[0002] The southwestern mountainous area is one of the areas with the highest incidence of landslide disasters. At the same time, it is also a serious area of ​​soil erosion in the country. Landslide disasters and soil erosion interact with each other and have a certain linkage relationship. Relevant investigations and studies have shown that soil erosion and landslide disasters are closely correlated in time and space. The correlation is manifested in the microscopic aspect that the development of landslide disasters is the inheritance and development of the specific form of soil erosion. On the macroscopic aspect, it is manifested in that the zoning of the degree of soil erosion has a certain correspondence with the development density of landslide disasters, and within a certain spatial scale, the degree of soil erosion can approximately represent the degree of development of landslide disasters in the area. Summary of the invention

[0003] Existing slope monitoring and early warning fail to take into account the inherent linkage between soil erosion and landslide disasters. Based on this, the present application provides an intelligent early warning method for landslide disasters based on soil erosion monitoring. Through indoor model tests and numerical simulations, tests and numerical simulation analysis are carried out on the early warning slopes to generate a large number of data samples. The sample information includes the safety factor, apparent displacement, soil erosion index (rainfall infiltration and runoff sediment content), and range of instability of the slope. A neural network model is established and trained based on the geological information of the slope, model calculation parameters, and sample information to construct a neural network model. Monitoring equipment is deployed on site, and based on the real-time monitoring data on site (slope displacement, rainfall infiltration, and runoff sediment content), the prediction results are obtained according to the constructed neural network model, so as to issue an early warning.

[0004] The purpose of this application is achieved through the following technical solutions:

[0005] An intelligent early warning method for landslide disasters based on soil and water loss monitoring includes the following steps:

[0006] S01, obtaining basic information characteristics of the slope;

[0007] S02, indoor model test: According to the basic information characteristics of the slope, the indoor model of the slope is designed based on similarity theory, and the indoor model of the slope is tested under different rainfall scenarios. The test data of the slope during the test are obtained as the sample data set;

[0008] S03, numerical simulation: according to the basic information characteristics of the slope, a numerical analysis model is established to simulate the dynamic response of the numerical analysis model under different rainfall scenarios, and the simulated data of the slope in the numerical analysis process is obtained as a sample data set;

[0009] S04, neural network model establishment: using the sample data set to train the neural network, a neural network model is established that can map the implicit relationship between rainfall infiltration, soil erosion, slope displacement, slope safety factor and instability range;

[0010] S05, on-site monitoring;

[0011] S06, disaster warning;

[0012] The order of steps S02 and S03 can be adjusted forward and backward.

[0013] Furthermore, the slope object is a soil slope, and the soil strength complies with the Mohr-Coulomb criterion.

[0014] Furthermore, the basic information characteristics of the slope include geological structure characteristic information of the slope and physical and mechanical information of the soil.

[0015] Furthermore, the geological structural characteristic information includes the slope geometry, geological composition and soil particle gradation; the soil physical and mechanical information includes density, shear strength, internal friction angle and permeability coefficient.

[0016] Furthermore, the test data in the indoor model test include apparent displacement, soil erosion index and range of instability, and the soil erosion index includes rainfall infiltration and runoff sediment content.

[0017] Furthermore, the simulation data in the numerical simulation include apparent displacement, soil erosion index and range of instability, and the soil erosion index includes rainfall infiltration and runoff sediment content.

[0018] Furthermore, in establishing the neural network model, the sample data set is divided, one part is used as a training set, and the other part is used as a test set. First, the training set is used to train the neural network model to establish the neural network model. Secondly, the test set is used to test and calculate the neural network model to obtain the test relative error. Finally, by changing the structure of the neural network model and repeating the training and testing steps, the neural network structure with the smallest relative error is searched, that is, the optimal neural network model is obtained.

[0019] Furthermore, the sample data set of numerical simulation is used as the training set, and the sample data set of indoor model test is used as the test set.

[0020] Furthermore, in the on-site monitoring, a location is selected on the slope, soil and water loss monitoring equipment and geological disaster monitoring equipment are arranged, monitoring work is carried out, and on-site monitoring data are collected.

[0021] Furthermore, the monitoring data in the on-site monitoring include slope displacement, rainfall infiltration and runoff sediment content.

[0022] Beneficial effects of this application:

[0023] 1. The model proposes the idea of ​​one slope one model, establishes data samples for a specific slope, and constructs an early warning model, which can effectively improve the accuracy and precision of the prediction.

[0024] 2. Different from the traditional slope displacement monitoring and early warning, the soil erosion index is introduced, and the inherent linkage between soil erosion and landslide disasters is taken into account, making the prediction model more in line with the actual slope.

[0025] 3. The model makes full use of on-site monitoring resources and applies soil and water loss monitoring content to slope monitoring, making the monitoring system more complete and giving full play to the economic benefits of on-site monitoring.

[0026] The aforementioned main scheme of the present application and its further options can be freely combined to form multiple schemes, all of which are schemes that can be adopted and claimed for protection in the present application; and in the present application, (non-conflicting options) can also be freely combined with each other and with other options. After understanding the scheme of the present application, those skilled in the art can understand that there are many combinations based on the prior art and common knowledge, all of which are technical schemes to be protected by the present application, and they are not exhaustively listed here. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flowchart of this application.

[0028] Figure 2 It is a schematic diagram of the neural network model of this application. DETAILED DESCRIPTION

[0029] The following non-limiting examples are provided to illustrate the present application.

[0030] Embodiment 1:

[0031] refer to Figure 1 and Figure 2 As shown, an intelligent early warning method for landslide disasters based on soil and water loss monitoring is shown. The slope object is a soil slope, and the soil strength meets the Mohr-Coulomb criterion.

[0032] This method uses numerical simulation, indoor experiments, and field monitoring to provide early warning for landslide disasters based on soil erosion monitoring. This method believes that continuous soil erosion occurs on the slope during continuous rainfall, and that soil erosion and landslide disasters have an inherent chain development law. Therefore, landslide disasters are warned and forecasted by monitoring soil erosion.

[0033] This method adopts the idea of ​​one slope and one model. The actual slope has extremely complex geological conditions, so it is difficult to find two identical or even similar slopes in reality. Therefore, the existing monitoring technology uses a simple early warning model to achieve monitoring and early warning of all slopes. This idea is not very reasonable. Based on this, this application proposes the idea of ​​one slope and one model, establishes a data sample for a specific slope, and constructs an early warning model, so as to achieve long-term early warning monitoring of the slope.

[0034] The method comprises the following steps.

[0035] S01, after determining the actual slope object, firstly, the basic information characteristics of the slope are obtained according to the geological survey report. The basic information characteristics of the slope include the geological structure characteristics information of the slope and the physical and mechanical information of the soil. The geological structure characteristics information includes the geometric shape of the slope, geological composition and particle grading of the soil, and the physical and mechanical information of the soil includes gravity, shear strength, internal friction angle and permeability coefficient.

[0036] S02, indoor model test: According to the basic information characteristics of the slope, the indoor model of the slope is designed based on similarity theory, and the indoor model of the slope is tested under different rainfall scenarios. The test data of the slope during the test is obtained as a sample data set.

[0037] The test data in the indoor model test include apparent displacement, soil erosion index and range of instability. The soil erosion index includes rainfall infiltration and runoff sediment content. The sample data set of the indoor model test is used as the test set when the neural network model is established.

[0038] S03, numerical simulation: According to the basic information characteristics of the slope, a numerical analysis model is established to simulate the dynamic response of the numerical analysis model under different rainfall scenarios, and the simulated data of the slope during the numerical analysis process is obtained as a sample data set.

[0039] The simulation data in the numerical simulation include apparent displacement, soil erosion index and range of instability. The soil erosion index includes rainfall infiltration and runoff sediment content. The sample data set of the numerical simulation is used as the training set when the neural network model is established.

[0040] The order of steps S02 and S03 can be adjusted forward and backward.

[0041] S04, neural network model establishment: use the sample data set to train the neural network and establish a neural network model that can map the implicit relationship between rainfall infiltration, soil erosion, slope displacement, slope safety factor and instability range.

[0042] In the establishment of the neural network model, the sample data set is divided, the simulated data is used as the training set, and the experimental data is used as the test set. First, the training set is used to train the neural network model to establish the neural network model. Secondly, the test set is used to test and calculate the neural network model to obtain the test relative error. Finally, by changing the structure of the neural network model and repeating the training and testing steps, the neural network structure with the smallest relative error is searched, that is, the optimal neural network model is obtained.

[0043] S05, On-site monitoring: In on-site monitoring, representative locations are selected on the slopes, soil and water loss monitoring equipment and geological disaster monitoring equipment are arranged, monitoring work is carried out, and on-site monitoring data are collected. The monitoring data in on-site monitoring include slope displacement, rainfall infiltration and runoff sediment content.

[0044] S06, Disaster Warning: Import the monitoring data collected on site into the neural network model, and issue early warning for landslide disasters through the neural network model.

[0045] The aforementioned basic examples and their further selected examples can be freely combined to form multiple embodiments, all of which are embodiments that can be adopted and claimed for protection in this application. In the scheme of this application, each selected example can be arbitrarily combined with any other basic examples and selected examples.

[0046] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent early warning method for landslide disasters based on soil and water loss monitoring, characterized in that: The steps include: S01, obtaining basic information characteristics of the slope; S02, indoor model test: According to the basic information characteristics of the slope, the indoor model of the slope is designed based on similarity theory, and the indoor model of the slope is tested under different rainfall scenarios. The test data of the slope during the test are obtained as the sample data set; S03, numerical simulation: according to the basic information characteristics of the slope, a numerical analysis model is established to simulate the dynamic response of the numerical analysis model under different rainfall scenarios, and the simulated data of the slope in the numerical analysis process is obtained as a sample data set; S04, neural network model establishment: using the sample data set to train the neural network, a neural network model is established that can map the implicit relationship between rainfall infiltration, soil erosion, slope displacement, slope safety factor and instability range; S05, on-site monitoring; S06, disaster warning; The order of steps S02 and S03 can be adjusted forward and backward.

2. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 1 is characterized in that: The slope object is a soil slope, and the soil strength conforms to the Mohr-Coulomb criterion.

3. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 1 is characterized in that: The basic information characteristics of the slope include the geological structure characteristic information of the slope and the physical and mechanical information of the soil.

4. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 3 is characterized in that: The geological structure characteristic information includes the slope geometry, geological composition and soil particle gradation; the soil physical and mechanical information includes density, shear strength, internal friction angle and permeability coefficient.

5. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 1 is characterized in that: The test data in the indoor model test include apparent displacement, soil erosion index and range of instability. The soil erosion index includes rainfall infiltration and runoff sediment content.

6. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 1 or 5, characterized in that: The simulation data in the numerical simulation include apparent displacement, soil erosion index and range of instability. The soil erosion index includes rainfall infiltration and runoff sediment content.

7. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 1 is characterized in that: In establishing the neural network model, the sample data set is divided into one part as a training set and the other part as a test set. First, the training set is used to train the neural network model to establish the neural network model. Secondly, the test set is used to test and calculate the neural network model to obtain the test relative error. Finally, by changing the structure of the neural network model and repeating the training and testing steps, the neural network structure with the smallest relative error is searched, that is, the optimal neural network model is obtained.

8. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 7 is characterized in that: The sample data set of numerical simulation is used as the training set, and the sample data set of indoor model test is used as the testing set.

9. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 1 is characterized in that: In the on-site monitoring, a location is selected on the slope, soil and water loss monitoring equipment and geological disaster monitoring equipment are arranged, monitoring work is carried out, and on-site monitoring data are collected.

10. The intelligent early warning method for landslide disasters based on soil and water loss monitoring according to claim 9 is characterized in that: The monitoring data in the on-site monitoring include slope displacement, rainfall infiltration and runoff sediment content.

Citation Information

Cited By

  • Multi-physics coupling slope monitoring method and system based on unmanned aerial vehicle technology

    CN120894718A