Seismic source parameter-based strong earthquake risk assessment method

Abstract

The invention discloses a strong earthquake risk assessment method based on focus parameters, and relates to the technical field of earthquake risk analysis. The method comprises the following specific steps: acquiring real-time seismic waveform data received by a seismic station located in a strong earthquake risk area; constructing a seismic source mechanism inversion model based on the real-time seismic waveform data and the seismic phase file; calculating focus parameters in real time through the focus mechanism inversion model; and inputting the seismic source parameters into a trained seismic magnitude prediction model, and outputting seismic magnitude prediction information. The physical process of the seismic source and the propagation characteristics of seismic waves are analyzed through the method, the strong earthquake risk can be evaluated more scientifically and more accurately, an important basis is provided for disaster prevention and reduction, and the possible influence and damage degree of the earthquake are predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of earthquake risk analysis, and in particular to a method for evaluating strong earthquake risk based on earthquake source parameters. Background Art

[0002] Earthquake is one of the natural disasters that seriously threatens the safety of human life and property. Due to the extremely complex process of earthquake gestation and occurrence, science is not yet able to accurately predict the occurrence of earthquakes, and earthquake early warning can be an acceptable, relatively practical and effective means of disaster reduction. Strong earthquake hazard assessment refers to the probability assessment of strong earthquake events that may occur in a specific area in the future to determine the earthquake risk level in the area. This assessment process involves a comprehensive analysis of historical records of earthquake activity, geological structure, fault activity and other aspects, with the aim of providing a scientific basis to guide related construction and planning activities. At present, the provincial and municipal earthquake bureaus and monitoring centers mainly process earthquake data manually, which often requires a lot of manpower and material resources. Although some automatic earthquake data processing technologies have been developed, the common defect of these technologies is the lack of universality and flexibility, that is, they are often only applicable to a few regions and specific data types. Summary of the invention

[0003] The purpose of the present invention is to provide a method for evaluating the risk of strong earthquakes based on earthquake source parameters to solve the problems raised in the background technology.

[0004] To achieve the above object, the present invention provides the following solution: a method for assessing the risk of strong earthquakes based on earthquake source parameters, the specific steps of which include the following:

[0005] Acquire real-time seismic waveform data received by seismic stations located in areas at risk of strong earthquakes;

[0006] Based on the real-time seismic waveform data and seismic phase files, construct a focal mechanism inversion model;

[0007] Real-time calculation of earthquake source parameters through the earthquake source mechanism inversion model;

[0008] The earthquake source parameters are input into a trained earthquake magnitude prediction model to output earthquake magnitude prediction information.

[0009] Preferably, the method further comprises filtering the real-time seismic waveform data using a wavelet coefficient denoising model to obtain filtered real-time seismic waveform data.

[0010] Preferably, the specific steps of determining the strong earthquake risk area are:

[0011] Obtain seismic literature data in the study area and determine the Curie point depth model for the land area;

[0012] Determine the earthquake frequency distribution map of the study area based on historical earthquake catalog data;

[0013] Determine the earthquake frequency superposition Curie point depth map of the study area according to the land Curie point depth model and the earthquake frequency distribution map;

[0014] Based on the seismic frequency superposition Curie surface depth map of the study area, it is determined that if Ms in the study area is greater than a preset threshold, it is determined to be the strong earthquake risk area.

[0015] Preferably, the steps of constructing the focal mechanism inversion model are:

[0016] Identify P-waves and S-waves in filtered real-time seismic waveform data;

[0017] Build a real-time focal mechanism inversion model and set the focal model parameters;

[0018] Initialize the focal mechanism parameters and focal mechanism type according to the prior information, and use the seismic waveform data preprocessed by the full waveform inversion algorithm to fit and optimize the focal mechanism parameters;

[0019] The focal mechanism parameters are adjusted iteratively, and the fitting and focal mechanism parameter optimization are repeated until the best fitting result is reached or convergence occurs.

[0020] Preferably, the earthquake source parameters include fault plane length, fault width, fault strike, fault dip and inclination, fault slip direction, fault offset and fault rupture propagation speed.

[0021] Preferably, in the earthquake magnitude prediction model, the input is the earthquake source parameters and the output is the earthquake magnitude prediction information. The specific steps for constructing the earthquake magnitude prediction model are: inputting the earthquake source parameters into a convolutional neural network model for training and using a particle swarm algorithm for optimization to obtain the earthquake magnitude prediction model.

[0022] Preferably, the earthquake magnitude prediction model includes a feature extraction network, a rapid magnitude estimation network and an adaptive decision network.

[0023] Preferably, the feature extraction network is composed of a CNN convolutional network, a mask layer, a bidirectional LSTM layer, two Attention layers and a Concatenate layer connected in sequence; the magnitude rapid estimation network and the adaptive decision network are both composed of three fully connected layers connected in sequence.

[0024] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: obtaining real-time seismic waveform data received by seismic stations located in strong earthquake risk areas; constructing a source mechanism inversion model based on the real-time seismic waveform data and seismic phase files; calculating source parameters in real time through the source mechanism inversion model; inputting the source parameters into a trained earthquake magnitude prediction model, and outputting earthquake magnitude prediction information, which can accurately and quickly identify strong earthquakes. By analyzing the physical process of the source and the propagation characteristics of seismic waves through the method of the present invention, the risk of strong earthquakes can be assessed more scientifically and accurately, providing an important basis for disaster prevention and mitigation, and predicting the possible impact and degree of damage of earthquakes. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] 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.

[0026] Figure 1 is a flow chart of the method of the present invention;

[0027] Figure 2 It is a system structure diagram of the present invention. DETAILED DESCRIPTION

[0028] 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.

[0029] The purpose of this invention is to provide a method for assessing the risk of strong earthquakes based on earthquake source parameters. Figure 1 As shown, the specific steps include the following:

[0030] S1. Acquire real-time seismic waveform data received by seismic stations located in strong earthquake risk areas;

[0031] S2. Construct a focal mechanism inversion model based on real-time seismic waveform data and seismic phase files;

[0032] S3, real-time calculation of earthquake source parameters through earthquake source mechanism inversion model;

[0033] S4. Input the earthquake source parameters into the trained earthquake magnitude prediction model and output the earthquake magnitude prediction information.

[0034] Furthermore, the specific steps of determining the strong earthquake risk area in step S1 are:

[0035] S11. Obtain seismic literature data in the study area and determine the Curie point depth model for the land area;

[0036] S12. Determine the earthquake frequency distribution map of the study area based on historical earthquake catalog data;

[0037] S13. Determine the earthquake frequency superposition Curie point depth map of the study area based on the land Curie point depth model and earthquake frequency distribution map;

[0038] S14. Based on the earthquake frequency superposition Curie surface depth map of the study area, it is determined that if Ms in the study area is greater than a preset threshold, it is determined to be a strong earthquake risk area.

[0039] A strong earthquake is an earthquake with a magnitude of 6 or greater. An earthquake with a magnitude of 8 or greater is called a giant earthquake. Magnitude refers to the size of an earthquake, usually represented by the letter M. The larger the earthquake, the larger the magnitude number. For every 1.0 magnitude difference, the energy released by the earthquake differs by about 32 times, and for every 2.0 magnitude difference, the energy differs by about 1,000 times.

[0040] Furthermore, the invention also includes filtering the real-time seismic waveform data using a wavelet coefficient denoising model to obtain filtered real-time seismic waveform data. The value of the wavelet threshold has an important influence on the final denoising effect. The wavelet threshold should be able to separate the noise to the greatest extent. However, as the number of wavelet decomposition layers increases, the wavelet coefficient region where the noise is located should gradually decrease, because most of the wavelet coefficients decomposed at a higher level are from the original seismic signal. Therefore, the invention constructs a self-adjusting wavelet threshold based on the number of decomposition layers.

[0041] The wavelet denoising method is an algorithm based on multi-resolution analysis of wavelet transform. Its basic idea is to remove the wavelet coefficients corresponding to the noise in each frequency band according to the characteristics that the wavelet decomposition coefficients of noise and signal in different frequency bands have different intensity distributions, retain the wavelet decomposition coefficients of the original signal, and then perform wavelet reconstruction on the processed coefficients to obtain a pure signal.

[0042] Furthermore, in step S2, the steps for constructing the focal mechanism inversion model are:

[0043] S21, identifying P waves and S waves in the filtered real-time seismic waveform data;

[0044] S22, build a real-time focal mechanism inversion model and set the focal model parameters;

[0045] S23, initializing focal mechanism parameters and focal mechanism types according to prior information, fitting the seismic waveform data preprocessed by the full waveform inversion algorithm, and optimizing the focal mechanism parameters;

[0046] S24, iteratively adjust the focal mechanism parameters, repeat the fitting and focal mechanism parameter optimization until the best fitting result is reached or convergence occurs.

[0047] Furthermore, in step S3, the source parameters calculated in real time by the source mechanism inversion model include fault plane length, fault width, fault strike, fault dip and inclination, fault slip direction, fault offset and fault rupture propagation speed.

[0048] Furthermore, in the earthquake magnitude prediction model, the input is the earthquake source parameters and the output is the earthquake magnitude prediction information. The specific steps of constructing the earthquake magnitude prediction model are: inputting the earthquake source parameters into the convolutional neural network model for training and using the particle swarm algorithm for optimization to obtain the earthquake magnitude prediction model.

[0049] Furthermore, the earthquake magnitude prediction model includes a feature extraction network, a magnitude rapid estimation network, and an adaptive decision network. The feature extraction network is composed of a CNN convolutional network, a mask layer, a bidirectional LSTM layer, two Attention layers, and a Concatenate layer connected in sequence; the magnitude rapid estimation network and the adaptive decision network are both composed of three fully connected layers connected in sequence.

[0050] Embodiment 2 of the present invention provides a method for assessing the risk of a strong earthquake based on earthquake source parameters, such as Figure 2 As shown, it includes a data acquisition module, a model building module, a parameter inversion module, and a prediction and evaluation module; among which,

[0051] A data acquisition module, used to acquire real-time seismic waveform data received by seismic stations located in strong earthquake risk areas;

[0052] Model building module, used to build focal mechanism inversion model based on real-time seismic waveform data and seismic phase files;

[0053] Parameter inversion module, used to calculate source parameters in real time through the focal mechanism inversion model;

[0054] The prediction and evaluation module is used to input the earthquake source parameters into the trained earthquake magnitude prediction model and output the earthquake magnitude prediction information.

[0055] 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 assessing strong earthquake hazard based on earthquake source parameters, characterized in that: The specific steps include the following: Acquire real-time seismic waveform data received by seismic stations located in areas at risk of strong earthquakes; Based on the real-time seismic waveform data and seismic phase files, construct a focal mechanism inversion model; Real-time calculation of earthquake source parameters through the earthquake source mechanism inversion model; The earthquake source parameters are input into a trained earthquake magnitude prediction model to output earthquake magnitude prediction information.

2. A method for assessing strong earthquake risk based on earthquake source parameters according to claim 1, characterized in that: The method also includes filtering the real-time seismic waveform data using a wavelet coefficient denoising model to obtain filtered real-time seismic waveform data.

3. A method for assessing strong earthquake risk based on earthquake source parameters according to claim 1, characterized in that: The specific steps for determining the strong earthquake risk area are: Obtain seismic literature data in the study area and determine the Curie point depth model for the land area; Determine the earthquake frequency distribution map of the study area based on historical earthquake catalog data; Determine the earthquake frequency stacking Curie point depth map of the study area according to the land Curie point depth model and the earthquake frequency distribution map; Based on the seismic frequency superposition Curie surface depth map of the study area, it is determined that if Ms in the study area is greater than a preset threshold, it is determined to be the strong earthquake risk area.

4. A method for assessing strong earthquake risk based on earthquake source parameters according to claim 2, characterized in that: The steps of constructing the focal mechanism inversion model are as follows: Identify P-waves and S-waves in filtered real-time seismic waveform data; Build a real-time focal mechanism inversion model and set the focal model parameters; Initialize the focal mechanism parameters and focal mechanism type according to the prior information, and use the seismic waveform data preprocessed by the full waveform inversion algorithm to fit and optimize the focal mechanism parameters; The focal mechanism parameters are adjusted iteratively, and the fitting and focal mechanism parameter optimization are repeated until the best fitting result is reached or convergence occurs.

5. A method for assessing strong earthquake risk based on earthquake source parameters according to claim 1, characterized in that: The earthquake source parameters include fault plane length, fault width, fault strike, fault dip and inclination, fault slip direction, fault offset and fault rupture expansion speed.

6. A method for assessing strong earthquake risk based on earthquake source parameters according to claim 1, characterized in that: In the earthquake magnitude prediction model, the input is the earthquake source parameters and the output is the earthquake magnitude prediction information. The specific steps of constructing the earthquake magnitude prediction model are: inputting the earthquake source parameters into the convolutional neural network model for training and using the particle swarm algorithm for optimization to obtain the earthquake magnitude prediction model.

7. A method for assessing strong earthquake risk based on earthquake source parameters according to claim 1, characterized in that: The earthquake magnitude prediction model includes a feature extraction network, a magnitude rapid estimation network and an adaptive decision network.

8. A method for assessing strong earthquake risk based on earthquake source parameters according to claim 7, characterized in that: The feature extraction network is composed of a CNN convolutional network, a mask layer, a bidirectional LSTM layer, two Attention layers and a Concatenate layer connected in sequence; the magnitude rapid estimation network and the adaptive decision network are both composed of three fully connected layers connected in sequence.

Images