Earthquake active fault monitoring and early warning method and system

Abstract

The invention discloses an earthquake active fault monitoring and early warning method and system. The early warning method comprises the following steps: taking an average deformation rate as a first characteristic, and taking an accumulated deformation quantity as a second characteristic; obtaining the position change and the speed change of the monitoring area based on the GNSS data, extracting a third feature according to the position change, and extracting a fourth feature according to the speed change; extracting a fifth feature based on level observation data, and extracting a sixth feature based on fault gas measurement data; constructing and training a first deep learning model and a second deep learning model, inputting the first feature, the second feature, the third feature and the fourth feature into the first deep learning model to obtain a first result, and inputting the fifth feature and the sixth feature into the second deep learning model to obtain a second result; and constructing and training a third deep learning model. And inputting the first result and the second result into a third deep learning model to obtain stratum early warning information.

Description

Technical Field

[0001] The present invention relates to the field of stratum early warning technology, and in particular to a method and system for monitoring and early warning of earthquake active faults. Background Art

[0002] A large number of studies at home and abroad have shown that active faults are the root cause of earthquakes and the culprit of earthquake disasters. The surface rupture and dislocation of active fault earthquakes cause serious direct damage to ground buildings (structures). Superimposed on the damage caused by earthquake vibrations, the ground damage and disaster level along the active faults are aggravated, showing that active faults have a significant control effect on the distribution of severe earthquake disaster zones. Practical experience in earthquake disaster prevention at home and abroad has shown that monitoring the spatial position of active faults, scientifically evaluating their seismic hazards, and reasonably avoiding or taking effective engineering measures on this basis are effective ways to reduce earthquake disaster risks and reduce disaster losses. However, the current existing technology lacks long-term monitoring and early warning methods and systems for stratum activities. Summary of the invention

[0003] In view of the above-mentioned prior art, the present invention provides a method and system for monitoring and early warning of earthquake active faults, mainly to solve the technical problems existing in the above-mentioned background technology.

[0004] To achieve the above object, the technical solution of the embodiment of the present invention is implemented as follows:

[0005] The first aspect of the present invention discloses a method for monitoring and early warning of earthquake active faults, the method comprising the following steps:

[0006] Acquire synthetic aperture radar data, GNSS data, leveling observation data, and fault gas measurement data;

[0007] Obtaining an average deformation rate and a cumulative deformation amount of each pixel based on the synthetic aperture radar data, taking the average deformation rate as a first feature, and taking the cumulative deformation amount as a second feature;

[0008] Obtaining the position change and speed change of the monitoring area based on the GNSS data, and extracting the third feature according to the position change and the fourth feature according to the speed change;

[0009] Extracting a fifth feature based on the leveling observation data and extracting a sixth feature based on the fault gas measurement data;

[0010] Constructing and training a first deep learning model and a second deep learning model, inputting the first feature, the second feature, the third feature, and the fourth feature into the first deep learning model to obtain a first result, and inputting the fifth feature and the sixth feature into the second deep learning model to obtain a second result;

[0011] Construct and train a third deep learning model. Input the first result and the second result into the third deep learning model to obtain formation warning information.

[0012] Optionally, obtaining an average deformation rate of each pixel based on the synthetic aperture radar data specifically includes:

[0013] Acquire a synthetic aperture radar image, and preprocess the synthetic aperture radar image;

[0014] Based on the preprocessed synthetic aperture radar image, a series of interferograms are generated;

[0015] Analyzing the time series of the interference pattern using the SBAS-I nSAR method, and extracting deformation time series data of each pixel point;

[0016] Obtaining deformation values ​​of each pixel at multiple time points in deformation time series data of each pixel;

[0017] The deformation values ​​are fitted using the least square method to form a deformation line, and the slope of the deformation line is the average deformation rate.

[0018] Optionally, obtaining the accumulated deformation amount of each pixel point based on the synthetic aperture radar data specifically includes:

[0019] Acquire a synthetic aperture radar image, and preprocess the synthetic aperture radar image;

[0020] Based on the preprocessed synthetic aperture radar image, a series of interferograms are generated;

[0021] Analyzing the time series of the interference pattern using the SBAS-I nSAR method, and extracting deformation time series data of each pixel point;

[0022] The deformation value of each pixel from the beginning of time to the end of time is obtained in the deformation time series data of each pixel, and the cumulative deformation amount is obtained based on the deformation value from the beginning of time to the end of time.

[0023] Optionally, multiple GNSS receivers are set up in the monitoring area, wherein the monitoring area includes a fault zone, and the GNSS receivers include at least one fixed receiver and three mobile receivers, and the mobile receivers collect data for no less than three effective time periods at each measuring point.

[0024] Optionally, the location change of the monitoring area is obtained based on GNSS data, specifically including:

[0025] Obtaining the coordinate position of each GNSS receiver at a time node of each measuring point through the GNSS receiver, and plotting the change of the coordinate position over time into a time series graph;

[0026] The position change of each receiver is extracted from the time series, expressed as the coordinate change in the easting, northing and vertical directions;

[0027] The coordinate differences between adjacent time nodes are calculated, and the final position change is obtained based on multiple coordinate differences.

[0028] Optionally, the speed change of the monitoring area is obtained based on the GNSS data, specifically including: after obtaining the multiple coordinate differences, fitting the coordinate differences by the least squares method to obtain a coordinate difference line, and calculating the slope of the coordinate difference line, which is the speed change.

[0029] Optionally, a level time series is obtained from the level observation data, and the level time series is used as the fifth feature.

[0030] Optionally, a gas time series is obtained from fault gas measurement data, and the gas time series is used as the sixth feature.

[0031] The second aspect of the present invention discloses a seismic active fault monitoring and early warning system, the system is used to implement a seismic active fault monitoring and early warning method as described in any one of claims 1 to 8, characterized in that the system includes: a level, a radon meter, a first data acquisition terminal, a second data acquisition terminal and an Internet of Things platform, the level is used to collect leveling observation data, the radon meter is used to collect fault gas measurement data, the first data acquisition terminal is used to receive data from a synthetic aperture radar and GNSS data, the second data acquisition terminal is used to receive leveling observation data and fault gas measurement data, and the Internet of Things platform is configured to receive data from the first data acquisition terminal and the second data acquisition terminal, and to generate formation warning information according to the first deep learning model, the second deep learning model, and the third deep learning model.

[0032] The beneficial effects of the present invention are as follows: by acquiring synthetic aperture radar data, GNSS data, leveling observation data and fault gas measurement data, the average deformation rate is obtained from the synthetic aperture radar data as the first feature, the cumulative deformation amount is obtained as the second feature, the position change and speed change of the monitoring area are obtained from the GNSS data, and the third feature is extracted from the position change, the fourth feature is extracted from the speed change, the level time series is obtained from the leveling observation data as the fifth feature, and the gas time series is obtained from the fault gas measurement data as the sixth feature, and the first feature, the second feature, the third feature and the fourth feature are input into the first deep learning model to obtain the formation deformation analysis result based on the radar situation, the fifth feature and the sixth feature are input into the second deep learning model to obtain the formation deformation analysis result based on the environmental situation, and finally the formation deformation analysis result based on the radar situation and the formation deformation analysis result based on the environmental situation are input into the third deep learning model to finally obtain the formation deformation result, and through multiple data sources such as GNSS, SAR, leveling observation and fault gas measurement, the crust movement and earthquake precursor information can be more comprehensively reflected. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of a method for monitoring and early warning of earthquake active faults in an embodiment of the present application;

[0034] Figure 2 This is a schematic diagram of the module composition of an earthquake active fault monitoring and early warning system in an embodiment of the present application.

[0035] Description of Figure Numbers:

[0036] A level meter 1, a radon detector 2, a first data collection terminal 3, a second data collection terminal 4 and an Internet of Things platform 5. DETAILED DESCRIPTION

[0037] The technical solution of the present invention is further elaborated in detail below in conjunction with the drawings and specific embodiments of the specification. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. In the following description, the expression "some embodiments" is related to a subset of all possible embodiments, but it should be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0038] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0039] It should be understood that the present invention can be implemented in different forms and should not be interpreted as being limited to the embodiments proposed herein. On the contrary, providing these embodiments will make the disclosure thorough and complete, and the scope of the present invention will be fully conveyed to those skilled in the art. And the purpose of the terms used herein is only to describe specific embodiments and is not intended to be a limitation of the present invention. When used herein, the singular forms of "one", "one" and "said / the" are also intended to include plural forms, unless the context clearly indicates another way. It should also be understood that the terms "compose" and / or "include" when used in this specification determine the presence of the features, integers, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0040] It should also be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "inside", "outside", "left", "right" and similar expressions used herein are for illustrative purposes only and are not intended to be the only implementation method.

[0041] In order to fully understand the present invention, a detailed structure will be proposed in the following description to illustrate the technical solution proposed by the present invention. The optional embodiments of the present invention are described in detail as follows, but in addition to these detailed descriptions, the present invention may also have other implementations.

[0042] Please refer to the attached Figure 1 The first aspect of the present invention discloses a method for monitoring and early warning of earthquake active faults, the method comprising the following steps:

[0043] S1. Acquire synthetic aperture radar data, GNSS data, leveling observation data and fault gas measurement data;

[0044] S2. Obtaining an average deformation rate and a cumulative deformation amount of each pixel based on the synthetic aperture radar data, taking the average deformation rate as a first feature, and taking the cumulative deformation amount as a second feature;

[0045] S3. Obtaining the position change and speed change of the monitoring area based on the GNSS data, and extracting the third feature according to the position change, and extracting the fourth feature according to the speed change;

[0046] S4, extracting the fifth feature based on the leveling observation data, and extracting the sixth feature based on the fault gas measurement data;

[0047] S5. Build and train a first deep learning model and a second deep learning model, input the first feature, the second feature, the third feature, and the fourth feature into the first deep learning model to obtain a first result, and input the fifth feature and the sixth feature into the second deep learning model to obtain a second result;

[0048] S6. Construct and train a third deep learning model. Input the first result and the second result into the third deep learning model to obtain formation warning information.

[0049] This embodiment obtains synthetic aperture radar data, GNSS data, leveling observation data and fault gas measurement data, obtains the average deformation rate from the synthetic aperture radar data as the first feature, obtains the cumulative deformation amount as the second feature, obtains the position change and speed change of the monitoring area from the GNSS data, and extracts the third feature from the position change, extracts the fourth feature from the speed change, obtains the leveling time series from the leveling observation data as the fifth feature, obtains the gas time series from the fault gas measurement data as the sixth feature, and inputs the first feature, the second feature, the third feature and the fourth feature into the first deep learning model to obtain the formation deformation analysis result based on the radar condition, inputs the fifth feature and the sixth feature into the second deep learning model to obtain the formation deformation analysis result based on the environmental condition, and finally inputs the formation deformation analysis result based on the radar condition and the formation deformation analysis result based on the environmental condition into the third deep learning model to finally obtain the formation deformation result.

[0050] In one possible implementation, the type structures of the first deep learning model, the second deep learning model, and the third deep learning model are all different. For example, the first deep learning model adopts a convolutional recurrent neural network, and the second deep learning model adopts a long short-term memory network (LSTM) model, which can handle long-term dependencies and is suitable for time series prediction of seismic waveform data, which is just suitable for level time series and gas time series, while the third learning model adopts an autoencoder structure, which can realize anomaly detection.

[0051] In a possible implementation, obtaining an average deformation rate of each pixel based on the synthetic aperture radar data specifically includes:

[0052] S201, acquiring synthetic aperture radar images, which are usually acquired by satellites, such as the Sentine 1-1 satellite of the European Space Agency or other commercial SAR satellites, and preprocessing the synthetic aperture radar images, wherein the preprocessing process includes geometric correction and atmospheric correction, wherein the geometric correction is used to eliminate the influence of terrain, and the atmospheric correction is to correct the atmospheric delay effect using an atmospheric model or auxiliary data (such as meteorological data),

[0053] S202, generating a series of interference patterns based on the preprocessed synthetic aperture radar image;

[0054] S203, using the SBAS-I nSAR method to analyze the time series of the interference pattern, and extracting deformation time series data of each pixel point;

[0055] S204, obtaining the deformation value of each pixel at multiple time points in the deformation time series data of each pixel, for example, for pixel D1, the deformation values ​​u1, u2, u3, u4 at time points t1, t2, t3, t4 are obtained;

[0056] S205, fitting the deformation values ​​using the least square method to form a deformation line, wherein the slope of the deformation line is the average deformation rate.

[0057] In a possible implementation manner, obtaining the accumulated deformation amount of each pixel point based on the synthetic aperture radar data specifically includes:

[0058] Acquire a synthetic aperture radar image, and preprocess the synthetic aperture radar image;

[0059] Based on the preprocessed synthetic aperture radar image, a series of interferograms are generated;

[0060] Analyzing the time series of the interference pattern using the SBAS-I nSAR method, and extracting deformation time series data of each pixel point;

[0061] The deformation value of each pixel from the beginning of time to the end of time is obtained in the deformation time series data of each pixel, and the cumulative deformation value is obtained based on the deformation value from the beginning of time to the end of time. For example, for pixel D1, the deformation values ​​u1, u2, u3, u4 at time points t1, t2, t3, and t4 are obtained, and the initial value is set to u0, and the cumulative deformation value is u0+(u2-u2)+(u3-u2)+(u4-u3).

[0062] In a possible implementation, the Beidou satellite positioning system (compatible with other GNSS systems) is used. A plurality of ground-based augmentation stations based on domestic Beidou satellite positioning are set up in the monitoring area, each of which is provided with a GNSS receiver, and the monitoring area constructed by connecting a plurality of GNSS receivers to each other includes a fault zone, and the GNSS receivers include at least one fixed receiver and three mobile receivers, and the mobile receivers collect data for no less than three effective periods at each measuring point, and the effective data for each period is no less than 23.5 hours.

[0063] Furthermore, the location change of the monitoring area is obtained based on the GNSS data, specifically including:

[0064] Obtaining the coordinate position of each GNSS receiver at a time node of each measuring point through the GNSS receiver, and plotting the change of the coordinate position over time into a time series graph;

[0065] The position change of each receiver is extracted from the time series, expressed as the coordinate change in the easting, northing and vertical directions;

[0066] The coordinate differences between adjacent time nodes are calculated through the differential algorithm, and the final position change is obtained based on multiple coordinate differences.

[0067] Furthermore, the speed change of the monitoring area is obtained based on the GNSS data, specifically including: after obtaining the multiple coordinate differences, fitting the coordinate differences by the least squares method to obtain a coordinate difference line, and calculating the slope of the coordinate difference line, which is the speed change.

[0068] Furthermore, the leveling time series is obtained from the leveling observation data, and the leveling time series is used as the fifth feature, wherein the technical requirements for leveling observation are implemented according to the second-class leveling measurement method. The accuracy of the monitoring instrument should meet its reliability requirements. The error requirement for settlement observation is ﹤±0.5mm, and the closing error requirement for leveling measurement is ﹤4mm (L is the length of the measuring section route, in km; when the measuring section length is less than 0.1km, it is calculated as 0.1km). The settlement observation is carried out using the American Trimble D INI03 high-precision digital level and indium steel barcode leveling rod. The error of the instrument per kilometer of leveling measurement is ±0.3mm.

[0069] Furthermore, existing studies have shown that seismic active fault gas radon is the most likely of the many fluid components generated inside the earth to migrate to the surface in large quantities and be released at a certain point on the surface. Its abnormal concentration and flux can well reflect the seismic activity and fault zone activity. Therefore, the greater the rupture intensity of the active fault and the more cracks produced by the rupture, the more radon gas is generated. The high-value measurement points of radon gas concentration in active faults are mainly near the fracture point and are basically consistent with the fault direction. The geochemical characteristics of soil gas in seismic active faults can reflect their activity. Therefore, using reliable equipment to conduct periodic comparative measurements of fault escaping gas may capture abnormal deformation of active faults and precursor information on the occurrence of activity.

[0070] The instrument used for the fault gas escaping measurement is the KJD-2000R radon detector produced by Sichuan Xinxinda Measurement and Control Technology Co., Ltd. The gas extraction flow rate during measurement is 1L / min; the measurement unit is Bq / m3 (indicates the number of radon atoms that decay in every 1m3 of gas every 1 second); the sampling rate is 300s, that is, one value is read every 5 minutes; the distance between points on the measuring line is mostly 20m, and the distance between a few adjacent measuring points will increase or decrease slightly due to site limitations; 3 data are read at each measuring point during measurement, and the second result or the average of the second and third measurement results is the radon concentration at that point (because the first measurement value may be unstable), and the reading is accurate to 0.1Bq / m3, and finally the gas time series is obtained from the fault gas measurement data, and the gas time series is used as the sixth feature.

[0071] The second aspect of the present invention discloses a seismic active fault monitoring and early warning system, the system is used to implement a seismic active fault monitoring and early warning method as described in any one of claims 1 to 8, characterized in that the system includes: a level meter 1, a radon meter 2, a first data acquisition terminal 3, a second data acquisition terminal 4 and an Internet of Things platform 5, the level meter 1 is used to collect leveling observation data, the radon meter 2 is used to collect fault gas measurement data, the first data acquisition terminal 3 is used to receive data from a synthetic aperture radar and GNSS data, the second data acquisition terminal 4 is used to receive leveling observation data and fault gas measurement data, and the Internet of Things platform 5 is configured to receive data from the first data acquisition terminal 3 and the second data acquisition terminal 4, and generate formation warning information according to the first deep learning model, the second deep learning model and the third deep learning model.

[0072] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. The protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for monitoring and early warning of earthquake active faults, characterized in that: The early warning method comprises the following steps: Acquire synthetic aperture radar data, GNSS data, leveling observation data, and fault gas measurement data; Obtaining an average deformation rate and a cumulative deformation amount of each pixel based on the synthetic aperture radar data, taking the average deformation rate as a first feature, and taking the cumulative deformation amount as a second feature; Obtaining the position change and speed change of the monitoring area based on the GNSS data, and extracting the third feature according to the position change and the fourth feature according to the speed change; Extracting a fifth feature based on the leveling observation data and extracting a sixth feature based on the fault gas measurement data; Constructing and training a first deep learning model and a second deep learning model, inputting the first feature, the second feature, the third feature, and the fourth feature into the first deep learning model to obtain a first result, and inputting the fifth feature and the sixth feature into the second deep learning model to obtain a second result; Construct and train a third deep learning model. Input the first result and the second result into the third deep learning model to obtain formation warning information.

2. A method for monitoring and early warning of seismic active faults according to claim 1, characterized in that: Obtaining an average deformation rate of each pixel based on the synthetic aperture radar data specifically includes: Acquire a synthetic aperture radar image, and preprocess the synthetic aperture radar image; Based on the preprocessed synthetic aperture radar image, a series of interferograms are generated; Analyzing the time series of the interference pattern using the SBAS-InSAR method, and extracting deformation time series data of each pixel point; Obtaining deformation values ​​of each pixel at multiple time points in deformation time series data of each pixel; The deformation values ​​are fitted using the least square method to form a deformation line, and the slope of the deformation line is the average deformation rate.

3. A method for monitoring and early warning of earthquake active faults according to claim 2, characterized in that: Obtaining the accumulated deformation amount of each pixel point based on the synthetic aperture radar data specifically includes: Acquire a synthetic aperture radar image, and preprocess the synthetic aperture radar image; Based on the preprocessed synthetic aperture radar image, a series of interferograms are generated; Analyzing the time series of the interference pattern using the SBAS-InSAR method, and extracting deformation time series data of each pixel point; The deformation value of each pixel from the beginning of time to the end of time is obtained in the deformation time series data of each pixel, and the cumulative deformation amount is obtained based on the deformation value from the beginning of time to the end of time.

4. A method for monitoring and early warning of earthquake active faults according to claim 3, characterized in that: A plurality of GNSS receivers are arranged in a monitoring area, wherein the monitoring area includes a fault zone, and the GNSS receivers include at least one fixed receiver and three mobile receivers, wherein the mobile receivers collect data for no less than three effective periods at each measuring point.

5. A method for monitoring and early warning of seismic active faults according to claim 4, characterized in that: The location changes of the monitoring area are obtained based on GNSS data, including: Obtaining the coordinate position of each GNSS receiver at a time node of each measuring point through the GNSS receiver, and plotting the change of the coordinate position over time into a time series graph; The position change of each receiver is extracted from the time series, expressed as the coordinate change in the easting, northing and vertical directions; The coordinate differences between adjacent time nodes are calculated, and the final position change is obtained based on multiple coordinate differences.

6. A method for monitoring and early warning of seismic active faults according to claim 5, characterized in that: Obtaining the speed change of the monitoring area based on GNSS data specifically includes: after obtaining the multiple coordinate differences, fitting the coordinate differences by least square method to obtain a coordinate difference line, and calculating the slope of the coordinate difference line, which is the speed change.

7. The method for monitoring and early warning of seismic active faults according to claim 1, characterized in that: A level time series is obtained from the level observation data, and the level time series is used as the fifth feature.

8. The method for monitoring and early warning of seismic active faults according to claim 1, characterized in that: A gas time series is obtained from fault gas measurement data, and the gas time series is used as the sixth feature.

9. A seismic active fault monitoring and early warning system, the system being used to implement a seismic active fault monitoring and early warning method as claimed in any one of claims 1 to 8, characterized in that: The system comprises: A level meter, a radon meter, a first data acquisition terminal, a second data acquisition terminal and an Internet of Things platform, wherein the level meter is used to collect leveling observation data, the radon meter is used to collect fault gas measurement data, the first data acquisition terminal is used to receive data from a synthetic aperture radar and GNSS data, the second data acquisition terminal is used to receive leveling observation data and fault gas measurement data, and the Internet of Things platform is configured to receive data from the first data acquisition terminal and the second data acquisition terminal, and to generate formation warning information according to a first deep learning model, a second deep learning model and a third deep learning model.

Images