KR100982448B1 - Ground subsidence prediction system and predicting method using the same - Google Patents

Abstract

PURPOSE: A ground subsidence prediction system and a predicting method using the same are provided to accurately predict land subsidence through more than one geospatial correlation analysis. CONSTITUTION: A land subsidence information database(100) supplies an occurrence position, a topography, a geological feature, and earth usage data. An geospatial correlation integrated analysis unit(200) analyzes the geospatial correlation based on the land subsidence information database. A setting unit(300) set an estimation rate of the land subsidence and estimates the integrated estimation.

Description

Ground Subsidence Prediction System and Predicting Method using the same}

The present invention relates to a ground subsidence occurrence prediction system and a method for predicting subsidence occurrence using the same as a result of "development of GIS-based national geographic information system commercialization technology" hosted by the Ministry of Knowledge Economy.

Anthracite coal emerged as the only energy resource in Korea in the 1940s, when energy resources were raised for national economic rehabilitation and public stability. As a result of the promotion of pelleting in 1948, up to 24.2 million tonnes were produced in 1988. However, since the mid-1980s, the tendency to prefer oil and gas fuels has decreased, resulting in a decrease in coal production (Coal Rationalization Project, 2000). These phenomena increased waste coal mines, and ground subsidence, water quality and soil pollution caused by acidic wastewater, and natural damage are occurring in the area around waste coal mines. In particular, it is a disaster phenomenon that poses a great risk in terms of existing living environment and stability, and accurate prediction of this can realize the effect of excluding this risk in advance. will be.

Predictive research is needed because the subsidence that occurs to the human life and infrastructure is irregular in the earth system. However, due to the development of advanced technologies and equipment, real-world geospatial data associated with ground subsidence are becoming more and more massive and complex. Therefore, GIS is needed to effectively find information related to geological phenomena and correlate with geospatial elements with complex characteristics.

  1 is a conceptual diagram illustrating a general geographic information system (GIS). Referring to this, the GIS is a powerful tool for efficient storage, management, and integrated analysis of massive geospatial data, and serves to support expert decision making. However, GIS lacks the ability to predict the spatial relevance related to the occurrence of events or to predict future occurrences inherent in large multidimensional data. Therefore, various fields related to geospatial use probabilities, statistics, and pattern recognition techniques using GIS to efficiently find potential information, spatial correlations and patterns in the spatial database, and predict future occurrences of events.

Various existing analytical methods have been used to calculate correlation coefficients or weights through the first-order analysis between dependent variables (event data) and independent variables (input data), and to prepare event occurrence predictions using them. Verification is performed to determine the predictive accuracy of the results obtained through each technique. However, even if the result shows the maximum prediction accuracy, there is a limit to the accuracy improvement depending on the type and characteristic of the technique.

Therefore, there is a limit in making accurate prediction about the occurrence of subsidence of a specific area by using such a geographic information system, and this limit acts as a deterrent to securing and developing the stability of human life and infrastructure.

The present invention has been made to solve the above problems, the object of the present invention is to use likelihood ratio analysis, WOE analysis using the location information of ground subsidence and the data such as terrain, geology, borehole data, mines and land use It is to provide a predictive system for performing accurate ground subsidence prediction analysis by allowing geospatial correlation analysis to be performed at least once using an analysis unit capable of performing logistic regression and artificial neural network analysis. .

As a means for solving the above problems, the present invention provides a ground subsidence prediction system, comprising: a ground subsidence information database for providing geological information such as ground subsidence occurrence location, topography, geology, land use; A geospatial correlation integrated analysis unit for integrated analysis of geospatial correlations based on the geological information provided from the geologic subsidence information database; A ground subsidence generation prediction diagram providing unit based on the information analyzed by the geospatial correlation integrated analysis unit, the ground subsidence occurrence prediction diagram and the integrated prediction diagram; A verification unit quantitatively verifying the accuracy of the ground subsidence prediction using the ground subsidence position; It is possible to provide a ground subsidence occurrence prediction system using a geospatial correlation integrated method characterized in that it comprises a ;;

In addition, the present invention, the ground subsidence information database, geological disaster database of the geospatial location information of the ground subsidence using the ground subsidence distribution map of the analysis region; Terrain database of slope factors extracted from topographic maps of the analysis area; Geological database of geological factors extracted from geological maps of the target area; Borehole database of groundwater depth and permeability factors extracted from borehole data in the target area; A gangdo database, which is a spatial database of gangdo chart data such as gangway depth and distance from the gang; And a soil use database extracted from the land use of the analysis target area; and to provide a ground subsidence occurrence prediction system using the integrated geospatial correlation method.

In addition, in the system according to the present invention, the geospatial correlation integrated analysis unit, the ground subsidence location in the ground subsidence information database that provides all the factors related to the ground subsidence as a dependent variable, topography, soil, clinical, geological And a likelihood ratio analysis unit for predicting the occurrence probability of ground subsidence by analyzing correlations between factors related to the observed ground subsidence using data such as land use and independent variables; A weight of evidence (WOE) analysis unit for analyzing weights of grades between ground subsidence and related factors; A logistic regression analysis unit for analyzing the logistic regression correlation coefficients of the independent data and the dependent data as input data; The ground subsidence generation region and the non-occurrence region may be made by including any one or more of the artificial neural network analysis unit for analyzing the weight of each factor through the back propagation algorithm.

In addition, the geospatial correlation integrated analysis unit, the first to n-th order prediction including any one or more of the likelihood ratio analysis unit, WOE (weight of evidence) analysis unit, logistic regression analysis unit, artificial neural network analysis unit The n-th predictive analyzing unit has the same configuration as the (n-1) th predictive analyzing unit, and includes the n-th order result value provided by the (n-1) th predictive analyzing unit. As an independent variable of the predictive analysis unit, it is possible to construct a ground subsidence occurrence prediction system using the integrated geospatial correlation method, which is characterized by re-analyzing the ground subsidence occurrence dependent variable of the analysis target area (where n is 2 or more). Natural numbers).

Using the system described above, it is possible to perform ground subsidence prediction performed by the following steps.

Specifically, the first step of constructing a spatial database of information on the causes of ground subsidence of the analysis target area; Integrating and analyzing the geospatial correlations between the factors related to the subsidence location established in the spatial database; A third step of preparing a predictive map and an integrated prediction diagram of ground subsidence using the analysis information of the second step; A fourth step of verifying the quantitative accuracy of the ground subsidence predictions and the integrated predictions derived in steps 2 and 3; 5 steps for deriving all the results; it is possible to implement a ground subsidence prediction method using the integrated geospatial correlation.

In this case, the second step is the likelihood ratio analysis and the WOE (weight of evidence) using the location of the ground subsidence as a dependent variable, and the topographic, geological, borehole data, mines and land use data as independent variables. Analysis), logistic regression analysis, or neural network analysis by performing one or more of the analysis can be formed in the step of analyzing the ground subsidence prediction.

In addition, using the ground subsidence prediction predicted in step 2 as an independent variable, and using the dependent variable input in step 2, likelihood ratio analysis, WOE (weight of evidence) analysis, logistic regression analysis, and artificial neural network analysis It is also possible to implement the ground subsidence generation integrated prediction method using the geospatial correlation integrated technique, characterized in that to perform at least two times the analysis of the ground subsidence prediction degree by performing any one or more of the analysis.

According to the present invention, likelihood ratio analysis, WOE analysis, logistic regression analysis, neural network analysis, and other various model analysis are performed by using the location information of the ground subsidence, topography, geology, borehole data, mine map and land use data. By performing the analysis using the analysis unit that can be at least two times, there is an effect that can perform a predictive analysis of accurate ground subsidence occurrence.

1 is an exemplary diagram conceptually showing a configuration of a conventional GIS.
Figure 2a is a block diagram of the ground subsidence occurrence prediction system using the integrated geospatial correlation method according to the present invention, Figures 2b and 2c is a configuration example of the ground subsidence occurrence database according to the present invention, Figure 2d The topographic map of the ground subsidence predicted area is shown.
Figure 2e shows the result of building all the factors related to ground subsidence into a spatial database.
3A and 3B illustrate the configuration of the geospatial correlation integrated analysis unit of the present system. Figure 3c shows the flow of the ground subsidence prediction analysis using the integrated geospatial correlation analysis, Figure 3d is a prediction diagram of ground subsidence occurrence using the likelihood ratio, Figure 3e is the likelihood ratio of factors related to the prediction of ground subsidence occurrence And weight of evidence.
FIG. 3F illustrates the ground subsidence prediction using the weight of evidence, and FIG. 3G illustrates the logistic regression correlation coefficients of the factors related to the ground subsidence. Figure 3h shows the ground subsidence prediction using logistic regression analysis.
FIG. 3I is a table illustrating artificial neural network weights of ground subsidence occurrence and related factors, and FIG. 3J illustrates a prediction diagram of ground subsidence occurrence using artificial neural network.
3K illustrates the integrated likelihood ratio and the integrated weight of evidence of ground subsidence related factors.
3L shows the integrated neural network weights of the ground subsidence-related factors (ground subsidence occurrence prediction maps prepared through the primary analysis likelihood ratio, WOE, logistic regression, and artificial neural network analysis).
FIG. 4A shows the results of integrated subsidence occurrence prediction (secondary analysis) using the likelihood ratio, and FIG. 4B shows the results of the integrated subsidence prediction (secondary analysis) using the weight of evidence, FIG. 4C. Shows the results of the integrated prediction of subsidence occurrence (secondary analysis) using logistic regression analysis. 4D shows the results of integrated prediction of ground subsidence (secondary analysis) using an artificial neural network.
FIG. 5A shows the accuracy of the ground subsidence prediction map of the first order single analysis, and FIG. 5B shows the accuracy of the ground subsidence integrated prediction map of the geospatial correlation analysis. This study compares the accuracy of the ground subsidence prediction and the integrated prediction in the second integrated analysis.

Hereinafter, with reference to the accompanying drawings will be described in detail the configuration and operation according to the present invention. In the description with reference to the accompanying drawings, the same components are given the same reference numerals regardless of the reference numerals, and duplicate description thereof will be omitted. The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

The present invention provides a ground subsidence prediction system using an integrated geospatial correlation method, and in particular, analysis of any one or more of likelihood ratio analysis, WOE (weight of evidence) analysis, logistic regression analysis, artificial neural network, and other various model analysis The main purpose of this study is to analyze the predictability of ground subsidence or to increase the accuracy of the prediction by repeating the first analysis.

Referring to Figure 2a, which shows the configuration of the ground subsidence occurrence prediction system (hereinafter referred to as 'main system') using the geospatial correlation integrated technique according to the present invention.

The present system according to the present invention is a geospatial correlation on the basis of the ground subsidence information database 100 for providing geological information such as the ground subsidence occurrence location, topography, geology, land use data, geological information provided from the ground subsidence information database The geospatial correlation integrated analysis unit 200 for integrated analysis of the relationship, the geospatial correlation integrated analysis unit 300 to prepare the ground subsidence occurrence prediction and integrated prediction diagram, the ground subsidence generation location It may be configured to include a verification unit 400 for quantitatively verify the accuracy of the ground subsidence prediction and integrated prediction using the result derived unit 500 for comparing the results.

In particular, the ground subsidence information database 100 may be provided with a geological disaster database (110) that is a spatial database of the location information of the ground subsidence using the ground subsidence distribution map of the analysis target area, and further, the topographic map of the analysis target area Geological database 130 using the geographic database 120 or geological data extracted from the geological map of the analysis region using the information on the slope of the terrain using the geography database 130, land use that provides information on the land use of the analysis region Database 140, the borehole database 150 for the drilling of the borehole data, such as the groundwater depth and permeability coefficient of the analysis target area, the spatial database of the gangdodo data such as the depth of the analysis target area and the distance from the gang The mine can also be configured to include a database 160 and the like.

An example of the formation of a database for predicting ground subsidence occurrence for a specific region using the present system described above with reference to FIGS. 2C and 2D will be described.

Figure 2c is a classification of the factors constituting the database formed in Figure 2b, Figure 2d is an embodiment to which the present system is applied to show the topographic map of the analysis target area, the specific location is Samma coal mine area located in Magyori, Gangwon-do. The Samma Coal Mine was a medium-sized coal mine that produced 2.43 million tons from 1960 to 1989. It was closed in 1989. In this area, ground subsidence has occurred due to the neglect of abandoned coal mines. In addition, ground subsidence continues as the Taebaek-Gyedo Route 38 extension is underway. This area is in latitude 37˚ 14 '26˝? 37˚ 15 '24˝N, Hardness 129˚ 02' 40˝? Located between 129˚ 03 '30'E.

Referring to the process of building a spatial database using the information on the application area (GIS) as follows.

The data collected for the ground subsidence prediction analysis in the Magyo area are ground subsidence distribution, topographical map, geological map, land use, borehole data, and mines (Fig. 2d, 2e). As a result of a survey by the Coal Industry Rationalization Project in 1999, the ground subsidence of the study area was reported as 21 places (Fig. 2d). Ground subsidences were categorized into training (70%, 7,259 grids) and verification (30%, 3,110 grids) and used for predictive analysis and verification of prediction. The slope was extracted from a DEM created from a 1: 5,000 digital topographic map. For geological data, we used 1: 50,000 scale bracken geological map published by Korea Institute of Geoscience and Mineral Resources. Groundwater depth and permeability coefficients were extracted from borehole data from the soil stability survey conducted by the Coal Rationalization Project. Factors for the degree of gangway were extracted by using the 1: 1,200 numerical mine map developed by the Coal Rationalization Project. In order to build the depth, the depth from the ground surface was calculated by subtracting the altitude above sea level in the DEM of the study area. The horizontal distance from the gang derived from the numerical mine map was calculated by buffering the gangway as a factor for horizontally predicting the risk range of ground subsidence. All factors related to ground subsidence were constructed with a spatial database as shown in Figure 4, and the grid size of the study area was set to 1m × 1m considering the scale of input data, and the number of grids in the study area was 1,742 × 1,207. The lattice number is 2,102,594.

Hereinafter, a configuration of the present system and a method for implementing a ground subsidence prediction using the same will be described with reference to FIGS. 3A to 3L.

3A and 3B show the configuration of the geospatial correlation integrated analysis unit of the present system (FIG. 2A).

The geospatial correlation integrated analysis unit 200 according to the present invention specifically uses the ground subsidence location information provided by the ground subsidence geological information database as a dependent variable, and uses terrain, geology, borehole data, mines and land use. The likelihood ratio analysis unit 210 predicts the probability of subsidence by analyzing correlations between factors related to the occurrence of subsidence, using data, etc. as independent variables, and analyzes the weights of grades between subsidence occurrence and related factors. WOE (weight of evidence) analysis unit 220, logistic regression analysis unit 230 for analyzing the logistic regression correlation coefficient of the independent variable as the input data and the dependent variable as the event data, the ground subsidence occurrence area and subsidence subsidence area Including at least one of the artificial neural network analysis unit 240 for analyzing the weight for each factor through the back-propagation algorithm, or further includes a variety of other model analysis unit It may be the lure. Of course, it is preferable to form in a structure including all of these analysis.

Alternatively, as shown in FIG. 3B, in another preferred embodiment according to the present invention, in addition to the above-described one geospatial correlation integrated analysis unit 200, an n-th predictive analysis of the result value analyzed by the first prediction unit is performed. It may be configured as an independent variable of, and further comprises an n-th prediction unit for performing the analysis by inputting the dependent variable used in the first prediction unit as a dependent variable again. The predictive analysis unit of this structure may be formed at least twice (where n is a natural number of two or more).

For example, the first to nth order predictive analysis unit including any one or more of the likelihood ratio analysis unit, the WOE (weight of evidence) analysis unit, logistic regression analysis unit, artificial neural network analysis unit, other model analysis unit When formed, the n-th predictive analyzing unit has the same configuration as the (n-1) th predictive analyzing unit, and the result value provided by the (n-1) th predictive analyzing unit is independent of the n-th order analyzing. It can be formed into a structure in which the dependent variable input from the (n-1) prediction part is input again as the dependent variable and the analysis is performed again (n is a natural number of two or more).

Figure 3c shows a flow chart for performing the ground subsidence prediction analysis using the integrated air pipe correlation integrated analysis using the present system according to the present invention.

As shown, the ground subsidence predictive analysis using this system, geospatial correlation integration uses the first analyzed ground subsidence prediction using a variety of single techniques as a new independent variable (input data). Perform secondary analysis with dependent variables (event data). This technique can be used to predict various geological phenomena (geological, natural disasters, geological resources, environmental pollution) as well as GIS, and can be applied to predicting all events occurring in geospatial space. The primary analysis method for the prediction of lipid phenomena may be various techniques based on probability, statistics, and data mining (see FIG. 1). In the present invention, ground sedimentation occurrence integrated predictive analysis using GIS, likelihood ratio, weight of evidence, logistic regression, artificial neural network, and the like was performed in the same procedure.

In other words, first, research areas and analysis areas are selected (for example, Magwi-ri, Gangwon-do), and then a spatial database is constructed using GIS. In this case, the location of the ground subsidence among the values provided from the spatial database is used as a dependent variable, and the terrain, geology, borehole data, mines, and land use data are input as independent variables to the geospatial correlation integrated analysis unit. .

When the geospatial correlation integrated analysis unit is formed of a single primary predictive analysis unit as shown in FIG. 3A or an nth integrated predictive unit formed in a complex form as shown in FIG. 3B, a single predictive analysis or repetitive analysis is performed as described above. Predictive analysis will be performed, and then the accuracy of the ground subsidence prediction and integrated predictions analyzed by various likelihood ratios, weight of evidence, logistic regression, artificial neural network, and various model analysis units will be verified. do.

Then, the verified ground subsidence prediction can be derived more accurate results through the comparison of accuracy.

Hereinafter, a detailed analysis method of the geospatial correlation integrated analysis unit will be described in detail.

1) Likelihood Ratio Analysis Part-Likelihood Ratio Analysis

The application of the likelihood ratio technique reveals the correlation between the factors related to the observed ground subsidence occurrence and predicts the subsidence occurrence region through the likelihood ratio of each factor. The likelihood ratio is the ratio of the ground subsidence area of each factor based on the conditional probability principle. If the likelihood ratio is greater than 1, the ground subsidence probability is high. If it is less than 1, it means that the ground subsidence probability is low. For calculating the likelihood ratio, the input data slope, gang depth, distance from the gang, depth of groundwater, and permeability coefficient were classified into 10 grades with equal area by grade. The calculated likelihood ratio (FIG. 3e) was assigned to each class of factors, and GIS overlapping analysis was used to calculate the subsidence Hazard Index by likelihood ratio (SHILR) as shown in {Equation 1}.

{Equation 1}

Ground subsidence prediction predicted using the risk index was graded in the top 5%, 10%, 15% 70% for visual interpretation (Fig. 3d; ground subsidence occurrence prediction using the likelihood ratio). The ground subsidence in the upper 30% (Very high – Medium) of the risk index was found to be 96%. The minimum value of the risk index was 1.14, the maximum value was 37.22, the average value was 7.00, and the standard deviation was 4.90.

2) WOE analysis part-weight of evidence analysis method

The weight of evidence technique takes a natural log of the likelihood ratio and presents the correlation between ground subsidence occurrence and factor ratings with positive and negative weights. The studentized value, C / S (C), is the optimal cutoff value that divides the values of factors related to ground subsidence, with positive weight and (-) based on the grade with the highest C / S (C) value. ) And dividing by weight. If the weight has a value of zero, there is no correlation. If the weight has a value of zero, it has a negative correlation, and if it has a positive value, it shows a positive correlation. The calculated weight for each factor is as shown in FIG. 3E.

The subdivided weights were assigned to the grades of each factor, followed by overlapping analysis, and the ground subsidence hazard index (SHIwoe) was obtained as shown in {Equation 2}.

{Equation 2}

The ground subsidence prediction predicted using the risk index was ranked in the top 5%, 10%, 15% 70% for visual interpretation (Fig. 3f). The ground subsidence in the upper 30% of the risk index (Very high Medium) was 94%. The minimum value of the risk index was -4.11, the maximum value was 9.28, the average value was -2.62, and the standard deviation was 2.00.

3) Logistic Regression Analysis-Logistic Regression Analysis

In order to accurately grasp the relationship between the independent variable (input data) and the dependent variable (event data), a regression analysis is required to obtain a regression or prediction equation representing the regularity between the two variables. Logistic regression is appropriate because ground subsidence is classified as 0 and 1. The logistic correlation coefficient is a statistical value that shows how much the dependent variable changes according to the change of the independent variable.If the value is positive, the greater the value of the independent variable is, the greater the probability of ground subsidence is; The higher the value, the greater the probability that no ground subsidence will occur.

  The ground subsidence regression equation was derived as shown in {Equation 3} using the logistic regression coefficient shown in Table 3, and the probability of subsidence occurrence was calculated as {Equation 4}. The factor multiplied by each factor is the weight of the factors, and the ground subsidence risk index was calculated using {Equation 3} and {Equation 4}.

{Equation 3}

{Equation 4}

The ground subsidence prediction predicted using the risk index was ranked in the top 5%, 10%, 15% 70% for visual interpretation (Fig. 3H). The ground subsidence in the upper 30% of the risk index (Very high – Medium) was 98%. The minimum value of the risk index was 0, the maximum value was 0.60252, the average value was 0.00377, and the standard deviation was 0.02167.

4) Artificial Neural Network Analysis Part-Prediction of Ground Subsidence Using Artificial Neural Network

The backpropagation algorithm used in the artificial neural network precisely recognizes the subsidence area and the non-occurrence area to the neural network, and trains the subsidence area based on the input data of this location. Through this, the artificial neural network calculates the ground subsidence occurrence weight for the output layer. In order to calculate the weight, the artificial neural network structure was set to 7 × 14 × 1, and the relative weight of the factors was calculated by setting the maximum number of repetitions before reaching the target error to 5,000 times and the learning rate to 0.01 (FIG. 3I). The subsidence risk index for the entire study area was calculated by assigning the calculated weight to each factor.

The ground subsidence prediction predicted using the risk index was ranked in the top 5%, 10%, 15% 70% for visual interpretation (Fig. 3j). The ground subsidence in the upper 30% of the risk index (Very high – Medium) was 95%. The minimum value of the risk index was 0.00599, the maximum value was 0.98832, the average value was 0.36971, and the standard deviation was 0.23159.

5) Geospatial Correlation Integration Method-Second Analysis

If the first analysis using the respective analysis units of 1) to 4) is defined as the first analysis, thereafter, as another embodiment according to the present invention, it is repeatedly predicted using the above-described first analysis results. We will explain how to analyze (this is called 'secondary analysis' for convenience).

Precipitation analysis of ground subsidence occurrence using the existing single technique was performed by making prediction maps through primary analysis and verifying and comparing their accuracy. However, the geospatial correlation integration uses the same prediction method as the new independent variable (input data) and conducts the second analysis using the same technique. Also, the predictive value of the secondary analysis can be applied as a new independent variable (input data) for the third analysis. By repeating this, n-th analysis is performed using n new independent variables (input data). This is a new concept that has not been approached in previous research. The geospatial correlation analysis analyzes the ground subsidence prediction with the procedures of i) input of raw data, ii) generation of intermediate results or new input data, and v) derivation of high-level results.

를, 기법 fm에 적용하여 예측도

을 도출한다.{Equation 5} is a general formula for predictive drawing using a single method of primary analysis, which is a raw independent variable related to ground subsidence (raw input data).

Is applied to the technique fm

To derive

{Equation 5}

은 지공간 상관관계 통합분석에서 종속변수(지반침하 발생 위치)의 새로운 독립변수(입력자료)로 사용되어, 통합기법의 {식 6}을 이용하여 통합 예측도를 작성하게 된다.

은 2차 분석 {식 6}의 독립변수(입력자료)

로 설정하고 기법 fm에 적용하여 통합예측도

을 도출한다.Prediction diagram of {Equation 5}

Is used as a new independent variable (input data) of the dependent variable (ground subsidence location) in the integrated geospatial correlation analysis, and the integrated prediction map is made using the equation (6) of the integrated technique.

Is an independent variable (input) of the secondary analysis {Equation 6}

And predictive integration using technique fm

To derive

{Equation 6}

은 n차 분석의 독립변수(입력자료)

로 설정하고, 기법 fm에 적용하여 n차 통합예측도

을 도출한다.

Is an independent variable (input data) of n-th analysis

N-th integrated prediction by applying to technique fm

To derive

{Equation 7}

The new geospatial correlation analysis analyzes the relationship between likelihood ratio, weight of evidence, logistic regression, prediction of ground subsidence occurrence derived from artificial neural network and ground subsidence in the same way (FIG. 3k), The integrated risk index of ground subsidence was calculated by deriving the weight of evidence (Figure 3k), logistic regression equation (Equation 8), and the weight of artificial neural network (Figure 3l).

{Equation 8}

Ground sedimentation occurrence predictions, prepared using the integrated risk index, were ranked in the top 5%, 10%, 15% and 70% for visual interpretation (see FIG. 4).

FIG. 4A illustrates the results of the integrated prediction of the subsidence occurrence using the likelihood ratio (secondary analysis), and FIG. 4B illustrates the results of the integrated prediction of the subsidence occurrence using the weight of evidence (secondary analysis), FIG. 4C. Shows the results of integrated prediction of ground subsidence (secondary analysis) using logistic regression. 4D illustrates the integrated prediction map (secondary analysis) of ground subsidence occurrence using artificial neural networks.

The minimum value of the integrated risk index calculated by the second-order analysis likelihood ratio method is 0.03, the maximum value is 50.09, the average value is 4.00, and the standard deviation is 10.05. -0.78, the maximum value is 9.93, the average value is -6.45, and the standard deviation is 3.70.The minimum value of the integrated risk index calculated by the secondary analysis logistic regression technique is 0.00003, the maximum value is 0.86384, the average value is 0.00344, and the standard deviation is 0.02442, the minimum value of the integrated risk index calculated by the secondary analysis neural network method was 0.02272, the maximum value was 0.93418, the average value was 0.28487, and the standard deviation was 0.25106.

6) Result delivery department

The ground subsidence risk index calculated using the single technique and the integrated geospatial correlation technique is an estimate and needs to be verified to know the quantitative accuracy of the prediction. To this end, Success Rate Curve (SRC) and AUC (Area Under the Curve) methods were used (see FIGS. 5A and 5B). FIG. 5A shows the accuracy of the ground subsidence prediction map of the first single analysis, FIG. 5B shows the accuracy of the geospatial correlation integrated prediction map, and FIG. 5C shows the ground of the first single analysis and the second integrated analysis. This is a comparison of the accuracy of settlement prediction.

The X-axis of the SRC is a rating of the risk index value in the upper percentage, and the Y-axis is a rating value in the cumulative percent of ground subsidence. If the value of the X-axis is 1% when the value of the X-axis is 1%, it means that all the ground subsidences occur within the pixels of the upper 1% of the ground subsidence risk index, and that their predictions are correct. The AUC method was used for quantitative verification rather than SRC. This method is to calculate the area under the SRC. By multiplying the X-axis and the Y-axis by 1: 1, the area under the SRC can be obtained. The larger the area, the higher the accuracy of the prediction.

  Therefore, as a result of comparing the accuracy of the predicted and integrated predicted data of the first-order single analysis and the second-order geospatial correlation integrated analysis, the accuracy of the integrated predicted map increased from 0.31% to 2.25%. Therefore, it can be seen that the secondary geospatial correlation integrated analysis method has better predictive performance than the first-order single analysis method.

In the foregoing detailed description of the present invention, specific examples have been described. However, various modifications are possible within the scope of the present invention. The technical idea of the present invention should not be limited to the embodiments of the present invention but should be determined by the equivalents of the claims and the claims.

100: subsidence database
110: Geological Disaster Database
120: topographic database
130: Geological Data Database
140: land use database
150: borehole database
160: mine chart database
200: geospatial correlation interleaver
210: likelihood ratio analysis unit
220: weight of evidence analysis unit
230: logistic regression analysis unit
240: artificial neural network analysis unit
300: ground subsidence prediction map creation unit
400: ground subsidence prediction map verification unit
500: result comparison section

Claims (7)

In the ground subsidence prediction system,
A subsidence information database providing geological information such as subsidence location, topography, geology, and land use data; A geospatial correlation integrated analysis unit for integrated analysis of geospatial correlations based on the geological information provided from the geologic subsidence information database; A ground subsidence occurrence prediction drawing based on the information analyzed by the geospatial correlation integrated analysis unit; A verification unit which verifies quantitative accuracy of ground subsidence using the ground subsidence prediction map; A result derivation unit for comparing all the result diagrams analyzed in the prediction system; Consists of including
The ground subsidence information database,
Geological disaster database using geospatial subsidence distribution map of geothermal sedimentation to spatial database; A terrain database for information on the slope of the terrain using the topographic map of the analysis region; A geological database that spatializes geological data extracted from geological maps of the analysis region; a borehole database that spatializes borehole data such as groundwater depth and permeability coefficient of the analysis region; A gangdo database for spatial data of gangdo chart data such as gangway depth and distance from gangs of the analysis target area; And a land use database for providing information on land use of the region to be analyzed; Ground subsidence occurrence prediction system using an integrated geospatial correlation method characterized in that it comprises one or more of.
delete The method according to claim 1,
The geospatial correlation integrated analysis unit,
The location of occurrence of ground subsidence provided in the ground subsidence information database is used as a dependent variable, and topographical, geological, borehole data, mines and land use data are used as independent variables.
A likelihood ratio analysis unit for predicting the probability of ground subsidence by analyzing correlations between observed landslides;
A weight of evidence (WOE) analysis unit that analyzes weights of grades related to landslides;
A logistic regression analysis unit for analyzing the logistic regression correlation coefficients of the independent data and the dependent data as input data;
An artificial neural network analysis unit for analyzing the weighting factors of the ground subsidence and the subsurface subsidence by the backpropagation algorithm;
Ground subsidence occurrence prediction system using an integrated geospatial correlation method characterized in that it comprises any one or more of.
The method according to claim 3,
The geospatial correlation integrated analysis unit,
Consisting of the first to n-th predictive analysis unit comprising any one or more of the likelihood ratio analysis unit, WOE (weight of evidence) analysis unit, logistic regression analysis unit, artificial neural network analysis unit;
The nth predictive analyzing unit has the same configuration as the (n-1) th predictive analyzing unit, and the resultant value provided by the (n-1) th predictive analyzing unit is an independent variable of the nth order analyzing unit, and (n) -1) Input the dependent variable input from the prediction unit as the dependent variable
Ground subsidence occurrence prediction system using the integrated geospatial correlation method characterized in that the analysis again (n is a natural number of two or more).
A first step of constructing a spatial database of geological information on the ground subsidence factors of the analysis region;
Analyze the integration of geospatial correlations by inputting the geological information constructed in the spatial database, using the location information of the occurrence of subsidence as a dependent variable, and independent variables such as topography, geology, borehole data, mines, and land use data. As such, the step of analyzing one or more of the likelihood ratio analysis, the weight of evidence (WOE) analysis, the logistic regression analysis, and the artificial neural network analysis may be performed first to analyze the ground subsidence prediction predictive level;
A third step of generating a landslide prediction map and integrated prediction map by using the analysis information of the two steps; Four steps of quantitatively verifying the accuracy of the three-stage landslide occurrence prediction map and integrated prediction map; Step 5 and step 5 to derive the results for comparing all the results in steps 3 and 4, including,
The third step;
The subsurface sedimentation predictability derived by performing one or more of the likelihood ratio analysis, the weight of evidence analysis, the logistic regression analysis, and the neural network analysis is the independent variable of the second analysis. The process of analyzing the ground subsidence integrated prediction by performing one or more of the likelihood ratio analysis, the WOE (weight of evidence) analysis, the logistic regression analysis, and the neural network analysis using the dependent variable input from the analysis as a dependent variable Ground subsidence prediction method using a geospatial correlation integrated method characterized in that to be performed at least twice. delete delete

Images