Mine geological disaster dynamic identification method and system based on multi-source data

Abstract

The invention discloses a mine geological disaster dynamic identification method and system based on multi-source data, and relates to the technical field of geological disaster identification. The method comprises the following specific steps: acquiring multi-source monitoring data of a target area to form a multi-source heterogeneous data set; performing spatial registration and data fusion on the multi-source heterogeneous data set to obtain fused data; acquiring information features of the mine geological disaster from the fused data; inputting the information features into a mine geological disaster hidden danger recognition model, performing recognition analysis on the geological disaster hidden danger of the target area, and determining a mine geological disaster recognition result; and monitoring the mine geological disasters in real time, feeding back the real-time monitoring data to the mine geological disaster hidden danger recognition model, and updating the mine geological disaster recognition result in real time. According to the method, a mine geological disaster dynamic identification framework based on multi-source data is constructed, and the accuracy and the working efficiency of multi-source geological disaster information processing are improved, so that effective support is provided for disaster prevention and reduction of geological disasters.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster identification, and in particular to a method and system for dynamic identification of mining geological disasters based on multi-source data. Background Art

[0002] With the continuous intensification of mineral resource exploitation and large-scale human engineering activities, various types of geological disasters such as glaciers, collapses, landslides, ground subsidence and ground fissures have been induced, and they show a trend of high frequency and chain generation, posing a serious threat to mine safety production. Early identification of large-scale and efficient geological disasters is an important prerequisite for disaster prevention and mitigation, and is also a technical guarantee for engineering safety. The existing technology for identifying geological disasters in mines based on a small amount of observation data cannot achieve comprehensive, error-free and accurate predictions. Traditional methods for identifying geological hazards mainly rely on manual judgment, which has the problems of strong subjectivity and poor accuracy, and it is difficult to achieve multi-dimensional and cross-scale dynamic tracking, which cannot meet the needs of disaster prediction and early warning. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for dynamic identification of mining geological disasters based on multi-source data to solve the problems raised in the background technology.

[0004] To achieve the above object, the present invention provides the following solutions: On the one hand, a method for dynamic identification of mining geological hazards based on multi-source data is provided, and the specific steps include the following:

[0005] Acquire multi-source monitoring data of the target area to form a multi-source heterogeneous data set;

[0006] Performing spatial registration and data fusion on the multi-source heterogeneous data sets to obtain fused data;

[0007] Acquiring information characteristics of mining geological hazards from the fused data;

[0008] Inputting the information features into a mine geological hazard hidden danger identification model, identifying and analyzing the geological hazard hidden dangers in the target area, and determining the mine geological hazard identification results;

[0009] The mine geological disasters are monitored in real time, and the real-time monitoring data is fed back to the mine geological disaster hazard identification model to update the mine geological disaster identification results in real time.

[0010] Preferably, the multi-source monitoring data includes mine remote sensing image data, ground displacement observation data, meteorological observation data, InSAR data and geophysical data.

[0011] Preferably, it also includes preprocessing operations on the multi-source heterogeneous data sets, including data cleaning, data dimensionality reduction, and normalization processing; wherein data cleaning includes removing duplicates, processing missing values, correcting format errors and processing outliers, integrating the cleaned multi-source heterogeneous data, and unifying the format and structure of the multi-source heterogeneous data; and using principal component analysis to perform data dimensionality reduction.

[0012] Preferably, different preprocessing operation modes are adopted for different monitoring data. Specifically, for the mine remote sensing image data, Fourier transform algorithm is adopted to eliminate noise, histogram minimum removal algorithm is adopted to perform radiation correction, FLASSH atmospheric correction algorithm is adopted to perform atmospheric correction, and affine transformation algorithm is adopted to perform spatial registration to obtain orthophoto data; for the ground displacement observation data, differential interferometry short baseline set timing analysis technology is adopted to generate a sequence interference diagram of multiple main images, and deformation sequence data is obtained through spatial filtering, phase unwrapping, noise phase estimation and removal operations; for the InSAR data, a statistical filtering algorithm is adopted to remove noise points, ground control points are adopted to correct distortion, a feature matching algorithm is adopted to perform point cloud registration, a Gaussian filtering algorithm is adopted to remove unnecessary details, and a RANSAC segmentation algorithm is adopted to remove ground points to obtain preprocessed InSAR data.

[0013] Preferably, the specific steps of performing data fusion on the multi-source heterogeneous data sets are:

[0014] Extracting a target field from multi-source heterogeneous data, wherein the multi-source heterogeneous data includes one or more fields;

[0015] Preprocessing the multi-source heterogeneous data based on the target field;

[0016] The preprocessed multi-source heterogeneous data is converted into a data format of a graph data structure to obtain multi-source data in a unified format; and a multi-source data association algorithm based on adaptive constraint relaxation is used to associate the multi-source data to achieve multi-source data fusion, thereby achieving trajectory association.

[0017] Preferably, the specific steps of constructing the mine geological disaster hidden danger identification model are:

[0018] Construct a convolutional neural network model, divide the preprocessed multi-source monitoring data into training samples and test samples, train the convolutional neural network model using the deep learning framework and training samples, and verify the accuracy of the convolutional neural network model using the test samples;

[0019] The convolutional neural network model includes an input layer, a convolutional layer C1, a pooling layer S2, a convolutional layer C3, a pooling layer S4, a convolutional layer C5, a pooling layer S6, a fully connected layer F7, a fully connected layer F8 and an output layer. The convolutional layer extracts features from the input data, the pooling layer performs dimensionality reduction on the data after convolution by the convolutional layer, and further uses an adaptive pooling method to extract features. The fully connected layer sets multiple neurons, connects all features output by the pooling layer, sets the ReLU function as the activation function, and the output layer is fully connected to the neurons of the fully connected layer F8, and the output result is classified into two categories.

[0020] On the other hand, a dynamic identification system for mining geological disasters based on multi-source data is provided, including a multi-source data acquisition module, a data fusion module, a feature extraction module, a disaster identification module, and a real-time update module; wherein,

[0021] The multi-source data acquisition module is used to acquire multi-source monitoring data of the target area to form a multi-source heterogeneous data set;

[0022] The data fusion module is used to perform spatial registration and data fusion on the multi-source heterogeneous data sets to obtain fused data;

[0023] The feature extraction module obtains information features of mining geological disasters from the fused data;

[0024] The disaster identification module is used to input the information features into the mine geological disaster hidden danger identification model, identify and analyze the geological disaster hidden dangers in the target area, and determine the mine geological disaster identification results;

[0025] The real-time update module is used to monitor mining geological disasters in real time, feed back real-time monitoring data to the mining geological disaster hidden danger identification model, and update the mining geological disaster identification results in real time.

[0026] Preferably, it further comprises a data preprocessing module, which is connected to the multi-source data acquisition module and is used to perform preprocessing operations on the acquired multi-source data.

[0027] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0028] 1. A dynamic identification framework for mining geological hazards based on multi-source data has been constructed, which greatly improves the accuracy and work efficiency of multi-source geological disaster information processing, thereby providing effective support for geological disaster prevention and mitigation work;

[0029] 2. By combining mine remote sensing image data, ground displacement observation data, meteorological observation data, InSAR data and geophysical data, the three-dimensional identification and analysis of geological disaster hazards are realized, and the full use of multi-source data and high-tech can reduce the omission and misjudgment of geological disaster hazards;

[0030] 3. Based on the convolutional neural network, dynamic identification of mine geological disasters is carried out to effectively improve the efficiency of disaster identification, with good use effect and easy promotion and use. In addition, the real-time monitoring data is fed back to the mine geological disaster hazard identification model, and the mine geological disaster identification results are updated in real time. The positive feedback form is used to ensure the accuracy of dynamic identification of mine geological disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] 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 will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0032] Figure 1 is a flow chart of the method of the present invention;

[0033] Figure 2 It is a system structure diagram of the present invention. DETAILED DESCRIPTION

[0034] 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.

[0035] The purpose of this invention is to provide a method for dynamic identification of mining geological hazards based on multi-source data. Figure 1 As shown, the specific steps include the following:

[0036] S1, obtain multi-source monitoring data of the target area to form a multi-source heterogeneous data set;

[0037] S2, perform spatial registration and data fusion on multi-source heterogeneous data sets to obtain fused data;

[0038] S3, obtain information characteristics of mining geological hazards from fused data;

[0039] S4, inputting the information features into the mine geological hazard hidden danger identification model, identifying and analyzing the geological hazard hidden dangers in the target area, and determining the mine geological hazard identification results;

[0040] S5. Monitor mining geological disasters in real time, feed the real-time monitoring data back to the mining geological disaster hazard identification model, and update the mining geological disaster identification results in real time.

[0041] Furthermore, in step S1, the multi-source monitoring data includes mine remote sensing image data, ground displacement observation data, meteorological observation data, InSAR data and geophysical data. Synthetic aperture radar interferometry (InSAR) technology is a geodetic technology based on earth observation by satellite-borne sensors that has developed rapidly in the past three decades. Its basic principle is to obtain elevation information and deformation information of the surface by phase interference of two radar images of the same area. Among them, multi-baseline InSAR is a technology that uses multiple remote sensing images to obtain surface deformation information. It uses multiple baselines and is suitable for long-term monitoring. Compared with conventional measurements, InSAR technology has the characteristics of wider measurement range, higher accuracy, all-weather, all-day, and high efficiency, and has become one of the commonly used geodetic technologies.

[0042] Furthermore, it also includes preprocessing operations on multi-source heterogeneous data sets, including data cleaning, data dimensionality reduction, and normalization processing; among which, data cleaning includes removing duplicates, processing missing values, correcting format errors and processing outliers, integrating the cleaned multi-source heterogeneous data, and unifying the format and structure of the multi-source heterogeneous data; using principal component analysis to perform data dimensionality reduction.

[0043] It should be noted that different preprocessing operation methods are used for different monitoring data. Specifically, for mine remote sensing image data, the Fourier transform algorithm is used to eliminate noise, the histogram minimum removal algorithm is used for radiation correction, the FLASSH atmospheric correction algorithm is used for atmospheric correction, and the affine transformation algorithm is used for spatial registration to obtain orthophoto data; for ground displacement observation data, the differential interferometry short baseline set timing analysis technology is used to generate a sequence interference map of multiple main images, and the deformation sequence data is obtained after spatial filtering, phase unwrapping, noise phase estimation and removal operations; for InSAR data, the statistical filtering algorithm is used to remove noise points, the ground control points are used to correct distortion, the feature matching algorithm is used for point cloud registration, the Gaussian filtering algorithm is used to remove unnecessary details, and the RANSAC segmentation algorithm is used to remove ground points to obtain the preprocessed InSAR data.

[0044] Furthermore, the specific steps for data fusion of multi-source heterogeneous data sets are as follows:

[0045] S21, extracting target fields from multi-source heterogeneous data, wherein the multi-source heterogeneous data includes one or more fields;

[0046] S22. Preprocess multi-source heterogeneous data based on the target field;

[0047] S23, converting the preprocessed multi-source heterogeneous data into a data format of a graph data structure to obtain multi-source data in a unified format; and using a multi-source data association algorithm based on adaptive constraint relaxation to associate the multi-source data to achieve fusion of the multi-source data, thereby achieving trajectory association.

[0048] Furthermore, the specific steps of constructing a mine geological disaster hazard identification model are as follows:

[0049] Construct a convolutional neural network model, divide the preprocessed multi-source monitoring data into training samples and test samples, train the convolutional neural network model using the deep learning framework and training samples, and verify the accuracy of the convolutional neural network model using the test samples;

[0050] The convolutional neural network model includes an input layer, a convolutional layer C1, a pooling layer S2, a convolutional layer C3, a pooling layer S4, a convolutional layer C5, a pooling layer S6, a fully connected layer F7, a fully connected layer F8 and an output layer. The convolutional layer extracts features from the input data, the pooling layer performs dimensionality reduction processing on the data after convolution by the convolutional layer, and further uses an adaptive pooling method to extract features, the fully connected layer sets multiple neurons, connects all features output by the pooling layer, sets the ReLU function as the activation function, and the output layer is fully connected to the neurons of the fully connected layer F8, and the output result is classified into two categories.

[0051] On the other hand, a dynamic identification system for mining geological hazards based on multi-source data is provided. Figure 2 As shown, it includes a multi-source data acquisition module, a data fusion module, a feature extraction module, a disaster identification module, and a real-time update module; among them,

[0052] A multi-source data acquisition module is used to acquire multi-source monitoring data of the target area to form a multi-source heterogeneous data set;

[0053] The data fusion module is used to perform spatial registration and data fusion on multi-source heterogeneous data sets to obtain fused data;

[0054] A feature extraction module, which obtains information features of mining geological disasters from the fused data;

[0055] The disaster identification module is used to input information features into the mine geological disaster hidden danger identification model, identify and analyze the geological disaster hidden dangers in the target area, and determine the mine geological disaster identification results;

[0056] The real-time update module is used to monitor mining geological disasters in real time, feed back the real-time monitoring data to the mining geological disaster hazard identification model, and update the mining geological disaster identification results in real time.

[0057] Furthermore, it also includes a data preprocessing module, which is connected to the multi-source data acquisition module and is used to perform preprocessing operations on the acquired multi-source data.

[0058] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for dynamic identification of mining geological hazards based on multi-source data, characterized in that: The specific steps include the following: Acquire multi-source monitoring data of the target area to form a multi-source heterogeneous data set; Performing spatial registration and data fusion on the multi-source heterogeneous data sets to obtain fused data; Acquiring information characteristics of mining geological hazards from the fused data; Inputting the information features into a mine geological hazard hidden danger identification model, identifying and analyzing the geological hazard hidden dangers in the target area, and determining the mine geological hazard identification results; The mine geological disasters are monitored in real time, and the real-time monitoring data is fed back to the mine geological disaster hazard identification model to update the mine geological disaster identification results in real time.

2. According to the method for dynamic identification of mine geological disasters based on multi-source data described in claim 1, the multi-source monitoring data includes mine remote sensing image data, ground displacement observation data, meteorological observation data, InSAR data and geophysical data.

3. The method for dynamic identification of mining geological hazards based on multi-source data according to claim 2 further comprises preprocessing the multi-source heterogeneous data sets, including data cleaning, data dimension reduction, and normalization processing; wherein, Data cleaning includes removing duplicates, processing missing values, correcting format errors and processing outliers, integrating the cleaned multi-source heterogeneous data, and unifying the format and structure of multi-source heterogeneous data; using principal component analysis to reduce data dimensionality.

4. According to the method for dynamic identification of mining geological disasters based on multi-source data described in claim 3, different preprocessing operation modes are adopted for different monitoring data. Specifically, for the mining remote sensing image data, Fourier transform algorithm is adopted to eliminate noise, histogram minimum removal algorithm is adopted to perform radiation correction, FLASSH atmospheric correction algorithm is adopted to perform atmospheric correction, and affine transformation algorithm is adopted to perform spatial registration to obtain orthophoto data; for the ground displacement observation data, differential interferometry short baseline set timing analysis technology is adopted to generate a sequence interference diagram of multiple main images, and deformation sequence data is obtained through spatial filtering, phase unwrapping, noise phase estimation and removal operations; for the InSAR data, a statistical filtering algorithm is adopted to remove noise points, ground control points are adopted to correct distortion, a feature matching algorithm is adopted to perform point cloud registration, a Gaussian filtering algorithm is adopted to remove unnecessary details, and a RANSAC segmentation algorithm is adopted to remove ground points to obtain preprocessed InSAR data.

5. The method for dynamic identification of mining geological hazards based on multi-source data according to claim 1 is characterized in that: The specific steps of data fusion for the multi-source heterogeneous data sets are: Extracting a target field from multi-source heterogeneous data, wherein the multi-source heterogeneous data includes one or more fields; Preprocessing the multi-source heterogeneous data based on the target field; The preprocessed multi-source heterogeneous data is converted into a data format of a graph data structure to obtain multi-source data in a unified format; and a multi-source data association algorithm based on adaptive constraint relaxation is used to associate the multi-source data to achieve multi-source data fusion, thereby achieving trajectory association.

6. The method for dynamic identification of mining geological hazards based on multi-source data according to claim 1 is characterized in that: The specific steps of constructing the mine geological disaster hidden danger identification model are: Construct a convolutional neural network model, divide the preprocessed multi-source monitoring data into training samples and test samples, train the convolutional neural network model using the deep learning framework and training samples, and verify the accuracy of the convolutional neural network model using the test samples; The convolutional neural network model includes an input layer, a convolutional layer C1, a pooling layer S2, a convolutional layer C3, a pooling layer S4, a convolutional layer C5, a pooling layer S6, a fully connected layer F7, a fully connected layer F8 and an output layer. The convolutional layer extracts features from the input data, the pooling layer performs dimensionality reduction on the data after convolution by the convolutional layer, and further uses an adaptive pooling method to extract features. The fully connected layer sets multiple neurons, connects all features output by the pooling layer, sets the ReLU function as the activation function, and the output layer is fully connected to the neurons of the fully connected layer F8, and the output result is classified into two categories.

7. A dynamic identification system for mining geological hazards based on multi-source data, characterized in that: It includes multi-source data acquisition module, data fusion module, feature extraction module, disaster identification module and real-time update module; among them, The multi-source data acquisition module is used to acquire multi-source monitoring data of the target area to form a multi-source heterogeneous data set; The data fusion module is used to perform spatial registration and data fusion on the multi-source heterogeneous data sets to obtain fused data; The feature extraction module obtains information features of mining geological disasters from the fused data; The disaster identification module is used to input the information features into the mine geological disaster hidden danger identification model, identify and analyze the geological disaster hidden dangers in the target area, and determine the mine geological disaster identification results; The real-time update module is used to monitor mining geological disasters in real time, feed back real-time monitoring data to the mining geological disaster hidden danger identification model, and update the mining geological disaster identification results in real time.

8. The mining geological disaster dynamic identification system based on multi-source data according to claim 7 is characterized in that: It also includes a data preprocessing module, which is connected to the multi-source data acquisition module and is used to perform preprocessing operations on the acquired multi-source data.

Images