A method and system for constructing a seismic rupture fault model and a coseismic displacement field

Abstract

The application provides a method and system for constructing a seismic rupture fault model and a coseismic displacement field, and relates to the technical field of seismic data processing.The method comprises the following steps: acquiring GNSS data, strong seismograph data and accelerometer data of a preset area, and performing fusion processing to obtain multi-source seismic data of the preset area; performing seismic waveform inversion on the multi-source seismic data to obtain relevant crust parameters of the preset area; combining the relevant crust parameters, calculating a Green function using a layered plane dislocation model according to the multi-source seismic data to obtain a seismic rupture fault model of the preset area; performing inversion on the seismic rupture fault model through a gradient algorithm to obtain a seismic rupture surface of the preset area; and performing analysis on the seismic rupture surface to obtain a coseismic displacement field.The application combines multi-source data fusion and algorithms, effectively reduces the influence of noise interference and data errors, and accurately reveals the earthquake occurrence mechanism and surface deformation characteristics.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of seismic data processing, in particular to a method and system for constructing a seismic rupture fault model and a coseismic displacement field. BACKGROUND

[0002] Due to the huge impact of earthquake disasters on human society and the natural environment, establishing a seismic rupture fault model and a coseismic displacement field is of great significance for understanding the mechanism of earthquake occurrence, assessing earthquake disasters, conducting earthquake early warning, and earthquake prevention and disaster reduction. Among them, the seismic rupture fault model can reveal the spatial and temporal distribution of underground rock rupture during an earthquake, and the coseismic displacement field reflects the displacement of the ground surface when an earthquake occurs.

[0003] In related technologies, the seismic rupture surface inversion method mainly relies on single seismic wave data, such as body wave or surface wave data. These data are easily disturbed by noise during inversion, resulting in low accuracy and reliability of the inversion results. In addition, the existing method for constructing a coseismic displacement field is mostly based on theoretical model calculation, lacking effective combination with actual observation data, and unable to accurately reflect the true displacement of the ground surface when an earthquake occurs. SUMMARY

[0004] The problem solved by the present application is how to improve the accuracy of the seismic rupture fault model and the coseismic displacement field.

[0005] To solve the above problems, the present application provides a method and system for constructing a seismic rupture fault model and a coseismic displacement field.

[0006] In a first aspect, the present application provides a method for constructing a seismic rupture fault model and a coseismic displacement field, comprising:

[0007] Obtaining GNSS data, strong motion instrument data and accelerometer data of a preset area;

[0008] Fusing the GNSS data, the strong motion instrument data and the accelerometer data to obtain multi-source seismic data of the preset area;

[0009] Performing seismic waveform inversion on the multi-source seismic data to obtain related crust parameters of the preset area;

[0010] Using a layered plane dislocation model to calculate Green's function according to the multi-source seismic data in combination with the related crust parameters to obtain a seismic rupture fault model of the preset area;

[0011] Inverting the seismic rupture fault model by a gradient algorithm to obtain a seismic rupture surface of the preset area;

[0012] The seismic rupture surface is analyzed to obtain a coseismic displacement field.

[0013] Optionally, the GNSS data, the strong motion data and the accelerometer data of the preset area are acquired, including:

[0014] The surface position and displacement information of the preset area are determined through satellite positioning technology, and the surface position and the displacement information are taken as the GNSS data;

[0015] The ground acceleration, ground speed and ground displacement of the preset area are determined through strong motion recorder data, and the ground acceleration, the ground speed and the ground displacement are taken as the strong motion data;

[0016] The acceleration change value of the ground and the building in the preset area is determined through an acceleration sensor, and the acceleration change value is taken as the accelerometer data.

[0017] Optionally, the GNSS data, the strong motion data and the accelerometer data are fused to obtain multi-source seismic data of the preset area, including:

[0018] The GNSS data, the strong motion data and the accelerometer data are respectively subjected to time alignment processing and spatial coordinate conversion processing according to the same time reference and spatial reference, to obtain the GNSS data, the strong motion data and the accelerometer data eliminating time-space differences;

[0019] The GNSS data, the strong motion data and the accelerometer data eliminating time-space differences are respectively assigned weights through a weighted data fusion filtering algorithm;

[0020] The multi-source seismic data are obtained by fusing the weights corresponding to the GNSS data, the strong motion data and the accelerometer data eliminating time-space differences.

[0021] Optionally, the multi-source seismic data are subjected to seismic waveform inversion to obtain related crust parameters of the preset area, including:

[0022] The theoretical seismic waveform of the preset area is generated through forward modeling according to the multi-source seismic data and a preset underground medium model;

[0023] The error between the theoretical seismic waveform and an observed waveform of the preset area is determined by comparing the theoretical seismic waveform with the observed waveform;

[0024] The related crust parameters are obtained through iterative optimization according to the error.

[0025] Optionally, the combining the related crust parameters, calculating the Green function according to the multi-source seismic data using a layered plane dislocation model, and obtaining a seismic rupture fault model of the preset area, comprises:

[0026] inputting the related crust parameters into the layered plane dislocation model, and setting an internal earth layering model of the preset area according to the related crust parameters;

[0027] discretizing the internal earth layering model to obtain a plurality of sub-layers;

[0028] sequentially calculating propagation data of seismic waves in the sub-layers by a Thomson-Haskell propagation algorithm, and then obtaining a Green function value of each of the sub-layers by inverse Hankel transformation;

[0029] determining total deformation of the seismic waves in the crust by linear superposition;

[0030] obtaining the seismic rupture fault model according to the total deformation combined with the Green function value.

[0031] Optionally, the inverting the seismic rupture fault model by a gradient algorithm to obtain a seismic rupture surface of the preset area, comprises:

[0032] obtaining model data according to the seismic rupture fault model;

[0033] performing geometric correction on the model data, and performing grid division on a fault of the preset area to obtain a fault structure of the preset area;

[0034] performing model calculation according to the fault structure, and then inverting according to a calculation result of the model calculation to obtain the seismic rupture surface.

[0035] Optionally, the performing model calculation according to the fault structure, and then inverting according to a calculation result of the model calculation to obtain the seismic rupture surface, comprises:

[0036] performing model calculation according to the fault structure to obtain a seismic fault dip angle, a slip gradient loss function, a sliding gradient loss function, a fault slip compensation point source, a loss function maximum gradient, a sliding model displacement field, and a sub-area sliding gradient of the preset area, and calculating a fault slip compensation point source, a loss function maximum gradient, a sliding model displacement field, and a sub-area sliding gradient;

[0037] taking the seismic fault dip angle, the slip gradient loss function, the sliding gradient loss function, the sliding compensation point source, the loss function maximum gradient, the sliding model displacement field, and the sub-area sliding gradient as the calculation result, and inverting the calculation result by an analytical algorithm of surface deformation caused by internal shear and tensile faults to obtain the seismic rupture surface.

[0038] Optionally, the analyzing the seismic rupture surface comprises:

[0039] discretizing the seismic rupture surface into a plurality of discrete point doublet sources;

[0040] determining the coseismic displacement field according to the discrete point doublet sources.

[0041] Optionally, the determining the coseismic displacement field according to the discrete point doublet sources comprises:

[0042] obtaining a general equation in a wave number domain according to distribution and parameters of the discrete point doublet sources through a Hankel customized transform;

[0043] determining a special solution satisfying source and boundary conditions according to the general equation through a Thomson-Haskell propagator algorithm;

[0044] converting the special solution into a solution in a spatial domain through an inverse Hankel;

[0045] performing linear superposition operation according to the solution in the spatial domain to obtain seismic variation data;

[0046] obtaining the coseismic displacement field according to the seismic variation data.

[0047] In a second aspect, the present application provides a system for constructing a seismic rupture fault model and a coseismic displacement field, comprising:

[0048] a data acquisition unit configured to acquire GNSS data, strong motion data and accelerometer data of a preset region;

[0049] a data fusion unit configured to perform fusion processing on the GNSS data, the strong motion data and the accelerometer data to obtain multi-source seismic data of the preset region;

[0050] a first inversion unit configured to perform seismic waveform inversion on the multi-source seismic data to obtain relevant crust parameters of the preset region;

[0051] a fault model construction unit configured to combine the relevant crust parameters and calculate a Green function using a layered plane dislocation model according to the multi-source seismic data to obtain a seismic rupture fault model of the preset region;

[0052] a second inversion unit configured to perform inversion on the seismic rupture fault model through a gradient algorithm to obtain a seismic rupture surface of the preset region;

[0053] A displacement field construction unit is configured to analyze the seismic rupture surface to obtain a co-seismic displacement field.

[0054] The earthquake rupture fault model and the construction method and system of the co-seismic displacement field can eliminate the limitations of a single data source through fusion processing of multiple data sources, for example, high-precision displacement information of GNSS data combined with high-frequency seismic waveform information of strong motion instrument and accelerometer data, can more comprehensively reflect the dynamic changes of the ground and underground when the earthquake occurs, fully play the advantages of each data source, make up for the shortcomings of a single data source, and more comprehensively reflect the dynamic changes of the ground and underground during the earthquake process. Secondly, the estimation of the crustal parameters is further optimized through the seismic waveform inversion technology, which provides a more accurate physical basis for subsequent model calculation. Moreover, the Green function calculation based on the layered plane dislocation model considers the complex layered structure of the crust, which can more realistically simulate the propagation and deformation effect of seismic waves in multi-layer medium, thereby improving the accuracy of the earthquake rupture fault model. Finally, the gradient algorithm is used to invert the earthquake rupture fault model, and the co-seismic displacement field is calculated by combining the analytical algorithm, which further improves the fitting accuracy and reliability of the model. The method of combining multi-source data fusion with advanced algorithms can not only effectively reduce the influence of noise interference and data error, but also more accurately reveal the earthquake mechanism and ground deformation characteristics, and further provide more accurate and reliable data support for seismology research, earthquake disaster assessment, and earthquake prevention and disaster reduction. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 The method flowchart of the construction method of the earthquake rupture fault model and the co-seismic displacement field in the embodiment of the present application is shown in the figure;

[0056] Figure 2 The working structure diagram of the strong motion instrument in the embodiment of the present application is shown in the figure;

[0057] Figure 3 The weighted data fusion filtering algorithm diagram in the embodiment of the present application is shown in the figure;

[0058] Figure 4 The principle information diagram for obtaining related crustal parameters in the embodiment of the present application is shown in the figure;

[0059] Figure 5 The flowchart of customizing the Green algorithm in the embodiment of the present application is shown in the figure;

[0060] Figure 6 The principle diagram of the customizing Green algorithm for solving the response of the seismic source in the medium in the embodiment of the present application is shown in the figure;

[0061] Figure 7 The working flowchart of the earthquake rupture surface inversion calculation in the embodiment of the present application is shown in the figure;

[0062] Figure 8 A hierarchical model data processing schematic diagram is used for the formation model in the embodiment of the present application.

[0063] Figure 9 A structure schematic diagram of a system for constructing a seismic rupture fault model and a coseismic displacement field. DETAILED DESCRIPTION

[0064] In order to make the above objectives, features and advantages of the present application more apparent, specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms, and should not be interpreted as being limited to the embodiments set forth herein, rather, these embodiments are provided to make the present application more thorough and complete. It should be understood that the drawings and embodiments of the present application are only for illustrative purposes, and are not intended to limit the scope of protection of the present application.

[0065] It should be understood that each step described in the method embodiments of the present application can be performed in different orders, and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0066] The term "comprising" and variations thereof as used herein are open-ended, that is "including but not limited to"; the term "based on" is "based, at least in part, on"; and the term "optional" means "optional implementation". Related definitions of other terms will be given in the description below.

[0067] It should be noted that the modification of "one" or "multiple" mentioned in the present application is illustrative rather than limiting, and those skilled in the art should understand that, unless otherwise explicitly indicated in the context, it should be understood as "one or more".

[0068] The names of the messages or information exchanged between the devices in the embodiments of the present application are only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0069] In combination Figure 1 As shown, in order to solve the problems of the related art described above, the embodiment provides a method for constructing a seismic rupture fault model and a coseismic displacement field, comprising:

[0070] GNSS data, strong motion data and accelerometer data of a preset area are acquired.

[0071] Specifically, by deploying high-precision Global Navigation Satellite System (GNSS) monitoring stations, strong motion instruments and accelerometers in the preset area, real-time collection of ground motion data before and after the earthquake is realized. GNSS data can provide high-precision ground displacement information, strong motion instrument data is used to record the propagation characteristics of seismic waves, and accelerometer data is used to capture the acceleration changes during the earthquake process; these data are the basis for building the earthquake rupture fault model and coseismic displacement field, and provide accuracy and reliability for subsequent analysis.

[0072] The GNSS data, the strong motion instrument data and the accelerometer data are fused to obtain multi-source seismic data of the preset area.

[0073] Specifically, the acquired GNSS data, strong motion instrument data and accelerometer data are fused, and the fusion process includes data preprocessing (such as filtering, denoising), time alignment and data format unification. Through fusion processing, different types of data are integrated into a complete multi-source seismic data set, so as to more comprehensively reflect the ground motion characteristics during the earthquake process. The data fusion method of the embodiment can fully exert the advantages of each type of data and improve the accuracy and reliability of subsequent analysis.

[0074] The multi-source seismic data are subjected to seismic waveform inversion to obtain related crust parameters of the preset area.

[0075] Specifically, the related crust parameters of the preset area are calculated by using the fused multi-source seismic data through seismic waveform inversion technology. The seismic waveform inversion is a method based on the theory of seismic wave propagation, which inverts the velocity structure, density distribution and other parameters in the crust by fitting the difference between the observed seismic waveform and the theoretical waveform.

[0076] In combination with the related crust parameters, the Green function is calculated according to the multi-source seismic data using a layered plane dislocation model to obtain the earthquake rupture fault model of the preset area.

[0077] Specifically, in combination with the inverted crust parameters and multi-source seismic data, the Green function is calculated using a layered plane dislocation model; the Green function is used to describe the influence of dislocation on the fault surface on ground motion during the earthquake rupture process. By calculating the Green function, the earthquake rupture fault model of the preset area can be constructed, which can describe the dislocation distribution and rupture propagation characteristics on the fault surface during the earthquake rupture process in detail.

[0078] The earthquake rupture fault model is inverted by a gradient algorithm to obtain the earthquake rupture surface of the preset area.

[0079] Specifically, a gradient algorithm is used to invert the earthquake rupture fault model and optimize the model parameters to better fit the observed data. The gradient algorithm is a highly efficient optimization method that iteratively adjusts model parameters to gradually reduce the discrepancy between model predictions and actual observations. The earthquake rupture surface obtained through inversion clearly demonstrates the dislocation distribution and rupture propagation path along the fault plane during the earthquake rupture process, providing key input for subsequent coseismic displacement field calculations.

[0080] The earthquake rupture surface is analyzed to obtain the co-seismic displacement field.

[0081] Specifically, the inverted earthquake rupture surface is analyzed to calculate the ground coseismic displacement field caused by the earthquake rupture process. The coseismic displacement field reflects the instantaneous displacement changes of the ground during an earthquake and is an important basis for assessing earthquake impacts and disaster risks. By analyzing the earthquake rupture surface, the ground displacement amount and direction at different locations can be accurately calculated, resulting in a complete distribution map of the coseismic displacement field.

[0082] The earthquake rupture fault model and the construction method and system of the co-seismic displacement field of the present invention eliminate the limitations of a single data source through the fusion processing of multi-source data. For example, the high-precision displacement information of GNSS data is combined with the high-frequency seismic waveform information of strong motion detectors and accelerometer data, which can more comprehensively reflect the dynamic changes of the surface and underground when an earthquake occurs, give full play to the advantages of each data source, and make up for the shortcomings of a single data source, thereby more comprehensively reflecting the dynamic changes of the surface and underground during the earthquake process. Secondly, the estimation of crustal parameters is further optimized through seismic waveform inversion technology, providing a more accurate physical basis for subsequent model calculations. In addition, the Green's function calculation based on the layered plane dislocation model takes into account the complex layered structure of the crust, which can more realistically simulate the propagation and deformation effects of seismic waves in multi-layer media, thereby improving the accuracy of the earthquake rupture fault model. Finally, the earthquake rupture fault model is inverted by a gradient algorithm and combined with an analytical algorithm to calculate the co-seismic displacement field, further improving the fitting accuracy and reliability of the model. The method of combining multi-source data fusion with advanced algorithms can not only effectively reduce the impact of noise interference and data errors, but also more accurately reveal the earthquake occurrence mechanism and surface deformation characteristics. Furthermore, it can also provide more accurate and reliable data support for many tasks such as seismological research, earthquake disaster assessment, and earthquake prevention and mitigation.

[0083] Optionally, obtaining GNSS data, strong motion detector data, and accelerometer data of a preset area includes:

[0084] Determining the surface position and displacement information of the preset area by satellite positioning technology, and using the surface position and the displacement information as the GNSS data;

[0085] determining ground acceleration, ground velocity and ground displacement of the preset area through strong motion instrument recording data, and taking the ground acceleration, the ground velocity and the ground displacement as the strong motion instrument data;

[0086] determining acceleration change values of the ground and the building in the preset area through an acceleration sensor, and taking the acceleration change values as the accelerometer data.

[0087] Specifically, in the process of obtaining GNSS data, strong motion instrument data and accelerometer data of the preset area, first, the satellite positioning technology (such as GPS or Beidou system) is used to monitor the surface position and displacement information of the preset area in real time. Through the high-precision GNSS receiver, the accurate position change of the ground before and after the earthquake is obtained, and these data not only include horizontal displacement, but also involve vertical displacement, which provides important ground motion information for the construction of the earthquake rupture fault model. Among them, the GNSS data is the surface position and displacement information obtained by satellite positioning technology, which can provide high-precision ground deformation data. These data include displacement observation values of the station in the east, north and sky directions, which can be used to monitor the ground displacement caused by earthquakes, plate movement and other crustal deformation phenomena. GNSS data has the characteristics of high precision, high spatial and temporal resolution and all-weather observation, and is an important data source indispensable in seismology research and crustal deformation monitoring. For example, in earthquake monitoring, GNSS data can help determine the range and degree of ground deformation caused by earthquakes. Secondly, the acceleration, velocity and displacement data of the ground when the earthquake occurs are recorded through the strong motion instrument, which reflects the propagation characteristics of seismic waves on the ground, including the intensity, frequency and duration of seismic motion. Strong motion instrument data has the characteristics of high sampling rate and wide frequency band, and can record high-frequency signals in the earthquake process. Strong motion instruments are usually installed in earthquake-prone areas or key monitoring points, and can capture the dynamic changes in the propagation process of seismic waves. These data are crucial for analyzing the propagation characteristics of seismic waves and the impact of earthquakes on ground structures.

[0088] In the preferred embodiments of the present application, the GNSS data, the strong motion instrument data and the accelerometer data of the preset area are obtained by Figure 2As shown, the working structure of the strong motion seismograph, as the key equipment for seismic data acquisition in the invention; the main function of the seismograph is to record various seismic waves generated when the earthquake occurs, including body waves (P waves and S waves) and surface waves. The working structure of the seismograph is divided into several key parts, each part has a clear function and data flow, the sensor part of the seismograph is responsible for detecting the vibration of seismic waves, these sensors usually include multiple direction detectors to ensure that three-dimensional information of seismic waves can be fully captured. The raw data collected by the sensor is then transmitted to the data processing unit, which is responsible for amplifying, filtering and digitizing the signal to improve the quality and availability of the data. The data processed by the data processing unit is stored in the data storage module, at the same time, these data can also be transmitted in real time to the central processing system or remote monitoring center through the data transmission module. The data storage module not only saves the original seismic waveform data, but also may include the characteristic data after preliminary processing, such as the arrival time, amplitude and frequency of seismic waves, etc. The power management module in it, the module ensures that the seismograph can run stably under various environmental conditions. The power management module includes battery, solar panel or other energy supply equipment to adapt to different deployment scenarios and work requirements. The communication interface is used for data exchange and communication with external devices. Through the wireless communication module, the seismograph can send data to the nearby base station or satellite, realize the remote transmission and real-time monitoring of data. This structure diagram not only helps users and professionals understand the working mechanism of the seismograph, but also provides important background information for multi-source data fusion and seismic data processing in the invention. Through efficient data acquisition and transmission, the seismograph provides high-quality raw data for the establishment of the earthquake rupture fault model and the construction of the coseismic displacement field, ensuring the accuracy and reliability of the invention method. Finally, the acceleration sensor is used to monitor the acceleration change value of the ground and buildings in the preset area, wherein the accelerometer data is the acceleration change of the ground or building recorded by the acceleration sensor when the earthquake occurs. These data can capture the inertial force caused by the earthquake in real time, reflecting the intensity and dynamic characteristics of the ground motion. Accelerometer usually has high sensitivity and high sampling rate, can record seismic signals from low frequency to high frequency, is an important data source for studying ground motion characteristics and structural response. However, acceleration data is not intuitive, and it is usually subjected to two simple numerical integrations to obtain the displacement change result caused by the earthquake. The error of the accelerometer is cumulative, and the integration will inevitably further amplify the influence of the noise. In order to reduce the quality decline of the displacement change result caused by non-physical drift, it is necessary to analyze and fuse with other data.

[0089] Specifically, the control module is the "brain" of the entire system, responsible for receiving and processing data from various sensors; receiving data (such as seismic signals and temperature information) from sensors (such as seismic detection sensors and temperature measurement modules), executing corresponding control strategies and sending instructions to other modules (such as communication modules and alarm modules). For example, when a seismic signal triggers, it can quickly respond by activating a safety blocker to cut off power supply, preventing equipment damage or danger. The seismic detection sensor is mainly responsible for detecting the vibration of seismic waves, including body waves (P waves and S waves) and surface waves, and transmitting the detected raw seismic data to the control module for processing. The temperature measurement module is used to monitor the temperature changes of the external or internal environment, providing temperature information for the control module; and sending temperature data to the control module. The safety blocker will execute the operation of cutting off or restoring power supply according to the instructions of the control module when detecting a seismic signal, in order to prevent equipment damage or danger, thereby protecting equipment and personnel safety. The communication module is responsible for data exchange and communication with external devices (such as base stations or satellites); through the wireless communication module, seismic data can be transmitted in real time to a remote monitoring center, realizing remote monitoring and data analysis. The alarm module will issue an alarm when detecting a seismic signal or other abnormal conditions, reminding relevant personnel to take emergency measures. The power supply module provides stable power support for the entire system. The power supply module can include batteries, solar panels or other energy devices to adapt to different deployment scenarios and work requirements. The load is the device or system being monitored, which can be a building, a bridge or other important structure. The purpose of the seismograph is to monitor the performance of these structures during an earthquake to assess their safety. Each of the above modules works together to achieve the collection, processing, transmission and safety protection of seismic data.

[0090] In optional embodiments, through the comprehensive collection of multi-source data, the dynamic changes of the ground and buildings during an earthquake are fully captured, providing rich information for earthquake research and disaster assessment. GNSS data provides high-precision ground displacement information, which helps to determine the range of ground deformation caused by earthquakes; strong motion instrument data can record the propagation characteristics of seismic waves in detail, providing key input for seismic waveform inversion; and accelerometer data focuses on the seismic performance evaluation of buildings, which can help identify potential damage to buildings caused by earthquakes. The comprehensive use of these data not only improves the accuracy and reliability of earthquake monitoring, but also provides a solid data foundation for the construction of earthquake rupture fault models and coseismic displacement fields, thereby providing scientific basis for rapid response and disaster mitigation measures for earthquake disasters.

[0091] Optionally, the fusion processing of the GNSS data, the strong motion instrument data and the accelerometer data to obtain the multi-source seismic data of the preset region comprises:

[0092] The GNSS data, the strong motion data and the accelerometer data are respectively time-aligned and spatially converted according to the same time reference and spatial reference, to obtain the GNSS data, the strong motion data and the accelerometer data with eliminated time and space differences;

[0093] The GNSS data, the strong motion data and the accelerometer data with eliminated time and space differences are respectively assigned weights through a weighted data fusion filtering algorithm;

[0094] The GNSS data, the strong motion data and the accelerometer data with eliminated time and space differences are fused according to the corresponding weights, to obtain the multi-source seismic data.

[0095] Specifically, when fusing the GNSS data, the strong motion data and the accelerometer data, first, these data need to be time-aligned and spatially converted according to the same time reference and spatial reference; this process not only ensures the consistency of data from different sources in time and space, but also eliminates errors caused by time synchronization differences and spatial position differences of data acquisition equipment. Secondly, the GNSS data, the strong motion data and the accelerometer data with eliminated time and space differences are assigned weights through a weighted data fusion filtering algorithm; the assignment of weights is based on the accuracy, reliability and contribution to seismic analysis of data. For example, the GNSS data has high accuracy in displacement measurement and may be assigned a higher weight; while the strong motion data has an advantage in seismic waveform recording and will also be assigned a corresponding weight according to its importance. Finally, different data sources are fused according to these weights, to obtain multi-source seismic data. In the data fusion stage, the GNSS data, the strong motion data and the accelerometer data are fused through a weighted data fusion filtering algorithm.

[0096] The method of this embodiment comprehensively considers the accuracy and reliability of each data source by assigning appropriate weights to different data sources. Compared with the traditional accelerometer double integration method, the fused high-frequency displacement observation and high-frequency velocity observation data have higher accuracy. This is because although accelerometer data can provide high-frequency information, the integration process is prone to cumulative errors, while GNSS data can provide high-precision displacement information, and fusion can effectively compensate for the shortcomings of a single data source. In the fusion to obtain the regional co-seismic displacement field stage, the fusion algorithm automatically selects the appropriate calculation method according to the input stratum model type and performs data fusion to form the final ground surface deformation result. According to the different stratum models, there are mainly two calculation paths: when the stratum model adopts a homogeneous model, the algorithm adopts an analytical method to calculate the surface deformation caused by spatial internal shear and tensile faults. This method is based on the assumption of homogeneous medium and considers the interaction between uniform seismic source and ground surface to calculate the ground surface deformation under different fault characteristics. The data fusion algorithm can provide accurate seismic deformation prediction results under different geological conditions by intelligently selecting appropriate models and methods combined with the complexity of stratum structure.

[0097] In a preferred embodiment of the present application, time alignment is achieved through Timestamp Alignment, which reads the timestamps in the data files and aligns the data from different sensors to a unified time reference. If the data sampling rates are different, interpolation methods can be used for alignment. There is also a Time Delay Estimation-TDE method that calculates the time delay between two signals and aligns the data to the same time reference. Spatial coordinate transformation is achieved through coordinate transformation matrices, rotation matrices, and translation vectors, which convert data from one coordinate system to another. This method is based on linear algebra and is suitable for simple coordinate transformations. Alternatively, a seven-parameter transformation is used to convert one coordinate system to another through seven parameters (three translation parameters, three rotation parameters, and one scale parameter). This method is suitable for accurate conversion between different coordinate systems. Time synchronization and spatial coordinate transformation eliminate the temporal and spatial differences between data, allowing all data to be analyzed in the same reference frame. This process not only improves the compatibility of data, but also lays the foundation for subsequent fusion processing.

[0098] In another preferred embodiment of the present application, the fusion algorithm is combined with the stratum model to calculate the co-seismic displacement field. The stratum model is used to simulate the propagation of seismic waves in the earth's crust, and the fusion algorithm is used to calculate the surface deformation caused by the interaction between the seismic waves and the ground surface. This method is based on the assumption of a homogeneous medium and considers the interaction between uniform seismic sources and the ground surface to calculate the ground surface deformation under different fault characteristics. The data fusion algorithm can provide accurate seismic deformation prediction results under different geological conditions by intelligently selecting appropriate models and methods combined with the complexity of stratum structure. Figure 3The illustrated weighted data fusion filtering algorithm fuses GNSS data, strong motion data, and accelerometer data. By combining the advantages of the two data sources, the algorithm significantly improves the reliability and accuracy of displacement change analysis in earthquake monitoring. The input of the algorithm includes time-aligned GNSS displacement observation data and accelerometer data in the NEU (North, East, Up) direction. Accelerometer data provides high-frequency acceleration information, but the integration process is prone to cumulative errors, resulting in a decrease in the accuracy of displacement data. Although GNSS data has high accuracy, its sampling frequency is low. Through the weighted data fusion filtering algorithm, the two types of data can be effectively fused to overcome their respective shortcomings. Specifically, the algorithm first predicts the displacement and velocity based on the output of the accelerometer. This prediction takes advantage of the high-frequency characteristics of the accelerometer, allowing it to quickly respond to changes in seismic waves. Subsequently, the predicted values are updated based on GNSS observations. This updating process utilizes the high-precision characteristics of GNSS data to correct the predicted values, resulting in more accurate displacement and velocity data. As Figure 3 The iterative process of the algorithm is shown in the figure. At each time step, the algorithm continuously performs prediction and updating until the next GNSS observation value appears. Where w(k+1) represents the input signal, which is the external input at time step (k+1). Γ(k+1,k) represents the input transition matrix, which is used to convert the input signal w(k+1) into the influence on the state vector. Φ(k+1,k) represents the state transition matrix, which is used to convert the current state vector x(k) into the state vector x(k+1) at the next time step. x(k) represents the current state vector, which is the system state at time step k. x(k+1) represents the state vector at the next time step, which is calculated from the current state vector x(k) and the input signal w(k+1) through the state transition matrix and the input transition matrix. H(k+1) represents the output transition matrix, which is used to convert the state vector x(k+1) at the next time step into the output signal z(k+1). z(k+1) represents the output signal, which is the system output at time step (k+1). This iterative mechanism ensures the real-time and accuracy of the data, allowing the algorithm to dynamically adapt to changes in seismic waves. The weighted data fusion filtering algorithm balances the high-frequency characteristics of accelerometer data and the high-precision characteristics of GNSS data by adjusting the weights, ultimately outputting fused high-frequency displacement data and velocity data in the NEU direction. This algorithm not only improves the accuracy and reliability of the data, but also provides high-quality displacement data support for the establishment of the earthquake rupture fault model and the construction of the coseismic displacement field in this embodiment.

[0099] In optional embodiments, by time alignment and spatial coordinate conversion, the spatio-temporal differences between different data sources are eliminated, ensuring the consistency and comparability of the data, and providing a unified data framework for subsequent analysis. The use of weighted data fusion filtering algorithm further optimizes the integration process of the data, by reasonably allocating weights, fully utilizing the advantages of each data source, and improving the accuracy and reliability of the fused data.

[0100] Optionally, the seismic waveform inversion of the multi-source seismic data obtains the related crust parameters of the preset region, including:

[0101] Forward modeling is performed according to the multi-source seismic data and a preset underground medium model to generate theoretical seismic waveforms of the preset region;

[0102] The error between the theoretical seismic waveforms and the observed waveforms is determined by comparing the theoretical seismic waveforms with the observed waveforms of the preset region;

[0103] The related crust parameters are obtained by iterative optimization according to the error.

[0104] Specifically, the seismic waveform inversion technique is a geophysical method that infers the structure of the underground medium and the characteristics of the seismic source through seismic waveform data. The core of seismic waveform inversion is to use the propagation characteristics of seismic waves in the underground medium. Seismic waves start from the source, propagate through different media, and reach the seismograph on the ground. The waveform data recorded by the seismograph contains various information about the propagation of seismic waves, such as wave propagation time, amplitude, and frequency. Through these waveform data, the structure of the underground medium and the characteristics of the seismic source can be inverted. The goal of seismic waveform inversion is to find an underground medium model that makes the predicted seismic waveforms consistent with the actual observed seismic waveforms as much as possible. First, based on multi-source seismic data and a preset underground medium model, forward modeling is performed to generate theoretical seismic waveforms. Error evaluation is performed to calculate the difference between the theoretical seismic waveforms and the actual observed waveforms, usually using the least squares method or other optimization objective functions to quantify this difference, and parameter updating is performed to adjust the underground medium model and the seismic source parameters to reduce the difference between the theoretical waveforms and the observed waveforms. Iterative optimization is performed to repeat the above steps until an optimal underground medium model and seismic source parameters are found, which minimizes the difference between the theoretical waveforms and the observed waveforms. Seismic waveform inversion technology combines seismic waveform data and mathematical models to provide important parameters of the underground medium structure and the characteristics of the seismic source. These parameters include crustal thickness, density, Poisson's ratio, P-wave velocity, and S-wave velocity. Through seismic waveform inversion, these parameters can be more accurately estimated, thereby optimizing the crustal velocity model. Further parameter optimization enables the model to more accurately reflect the characteristics of the seismic rupture process and the surface deformation.

[0105] In a preferred embodiment of the present application, in combination Figure 4 Figure 1 shows a schematic diagram of the principle of obtaining relevant crustal parameters in a method for establishing a seismic rupture fault model and constructing a coseismic displacement field by fusing multi-source observation data, which shows the principle and information flow of obtaining the principle of crustal parameters. Specifically, Figure 4 The three-dimensional coordinate system (X, Y, Z) is used to represent the spatial coordinate system of the seismic activity area. The X and Y axes define the horizontal direction, and the Z axis is vertically upward, representing the depth. The hypocenter represents the initial position of the earthquake, located inside the crust; the seismic wave propagates in all directions from the hypocenter. The epicenter represents the vertical projection point of the hypocenter on the ground surface; it is also commonly referred to as the "epicenter position" in earthquake monitoring and reporting. The hypocenter depth represents the vertical distance from the hypocenter to the ground surface. The epicentral distance represents the straight-line distance from a certain point on the ground surface (such as an observation station or sensor location) to the epicenter; it is used to evaluate the time and intensity of seismic waves arriving at different locations. The crust is composed of different geological layers, each layer has different physical properties (such as density, velocity, etc.); different layers will affect the propagation speed and path of seismic waves. P-wave (P-wave) and S-wave (S-wave) velocity; P-wave (P-wave) is the fastest wave in seismic waves, which can propagate in solids, liquids and gases. S-wave (S-wave) propagates slower and can only propagate in solids. This part is the key part of obtaining relevant crustal parameters in the present application, and its main function is to obtain the crustal physical property parameters of the region according to the input central latitude and longitude of the seismic area. Specifically, the central latitude and longitude of the receiving seismic area are obtained, and the crustal parameter results are output through seismic waveform inversion technology. These parameters include the thickness, density, Poisson's ratio, P-wave velocity and S-wave velocity of the crust, and other key data, which provide necessary basic information for subsequent seismic rupture fault model establishment and coseismic displacement field calculation.

[0106] In an optional embodiment, theoretical seismic waveforms are generated by forward modeling and compared with actual observed waveforms, which can quantitatively evaluate the difference between the model and the actual seismic process. The iterative optimization process based on the error further improves the accuracy of the inversion results, so that the obtained crustal parameters can more accurately reflect the actual geological conditions. Not only provides reliable crustal parameters for the construction of the seismic rupture fault model, but also enhances the understanding of the seismic propagation mechanism, which helps to improve the accuracy of earthquake disaster prediction and evaluation.

[0107] Optionally, in combination with the relevant crustal parameters, the layered plane dislocation model is used to calculate the Green function based on the multi-source seismic data to obtain the seismic rupture fault model of the preset area, including:

[0108] The relevant crustal parameters are input into the layered plane dislocation model, and the internal layered model of the Earth in the preset area is set according to the relevant crustal parameters;

[0109] Discretizing the Earth's interior layered model to obtain multiple sublayers;

[0110] The propagation data of seismic waves in the sub-layers are calculated in sequence by using the Thomson-Haskell propagation algorithm, and the Green's function value of each sub-layer is obtained by using the inverse Hankel transform;

[0111] Determining the total deformation caused by the seismic waves in the Earth's crust by linear superposition;

[0112] The earthquake rupture fault model is obtained according to the total deformation and the Green's function value.

[0113] Specifically, first, the fused multi-source data is calculated using the Green's function based on the layered plane dislocation model. A layered model of the earth's interior is set, including information such as source depth, source location, receiver depth, etc., and the complex layered model is discretized into a series of uniform sub-layers for easy numerical calculation. The matrix of seismic wave propagation is calculated to describe the propagation of waves between different medium layers. Seismic wave propagation analysis is performed to calculate the response functions of different seismic waves near the seismic source. Wavenumber integration is performed to calculate the kernel function of the seismic wave, which describes the response of a specific point in the crust under the action of a given seismic source. The superposition of basic solutions is performed to calculate the total deformation caused by the seismic source in the crust, and the geometric model product of the earthquake rupture fault is output. Combined with Figure 5 The process of the customized Green algorithm shown in the figure solves the problem of seismic wave propagation in complex crustal structures by converting the partial differential equations of motion into conventional equations in the wavenumber domain. It uses the Hankel transform and the Thomson-Haskell propagation algorithm to improve the accuracy and stability of the calculation.

[0114] In a preferred embodiment of the present invention, Figure 8 As shown in the figure, when a layered model is used, the data summary and process required for a customized Green's algorithm are as follows: The customized Green's function method based on a layered model requires: layer model information, including the number of layers in the layered model, the thickness of each layer, and physical properties (such as shear wave velocity, compression wave velocity, and density). The physical parameters between layers, i.e., the interaction and transmission characteristics between different layers; earthquake source information, i.e., fault plane information and fault geometry, such as length, width, strike, dip, and slip angle; slip, i.e., the amount of slip occurring on the fault; and the coordinates of observation points, i.e., the locations of points to be observed within the specified earthquake simulation area.

[0115] Specifically, when the layered model is layered, the layered model information is needed, including the number of layers of the layered model and the thickness, physical properties (such as shear wave velocity, compression wave velocity, density, etc.) of each layer, and the physical parameters between layers, that is, the interaction and transmission characteristics between different layers. At the same time, the input seismic source information is needed, including the geometric characteristics (such as length, width, strike, dip, and slip angle) of the fault plane and the slip amount. The coordinates of the observation points are also crucial for specifying the positions of the points to be observed in the seismic simulation area. Next, these data are integrated by customizing the Green algorithm to integrate the layered model, seismic source and observation point information, and the seismic wave propagation is simulated by using the customized Green function to calculate the coseismic displacement field of the ground surface and underground structure caused by the earthquake. Finally, the regional coseismic displacement field is output, reflecting the displacement changes of each observation point in the region; the whole process combines the layered model information, seismic source information and observation point coordinates, and uses the customized Green function method to realize the accurate simulation of the earthquake rupture process and the coseismic displacement field, providing important theoretical support and data basis for earthquake research and disaster prevention.

[0116] In another preferred embodiment of the present application, it is specifically divided into seven steps, which are:

[0117] The geometric model of the earthquake rupture fault and the crustal velocity model parameters are input, specifically: the input of the algorithm includes the geometric model of the earthquake rupture fault and the crustal velocity model parameters. These parameters provide the location of the seismic source, the geometric shape of the fault, and the physical properties of the crust, such as thickness, density, Poisson's ratio, P-wave velocity and S-wave velocity, etc. Combined with the Figure 6 The principle of solving the response of the seismic source in the medium by the customized Green algorithm is shown in the figure, Figure 6 The position and characteristics of the seismic source and the Cartesian coordinates of the reference point are marked in the figure. These parameters define the basic characteristics of the seismic source and are the basis for calculating the propagation of seismic waves and the response of the ground surface. Through iteration or recursion, the propagation of seismic waves in a multi-layer elastic crust is calculated, starting from the layer where the seismic source is located, propagating upwards or downwards layer by layer, and the wave field is updated by applying the corresponding propagation matrix until the target layer or the calculated depth is reached. Among them, the three-dimensional coordinate system is marked with X, Y and Z axes, and the origin is marked as "0". There are two points on the Z axis: , , ) and ( , , ), and a point M(x, y, z) at a certain position in space. ( , , ) represents the position of the seismic source, and the coordinates of the seismic source are ,​ , ) gives its specific location in the Earth's crust. ( , , ) represents the mirror point of the seismic source on the surface, used to model reflected waves. M(x,y,z) is a point in space where the seismic wave arrives, with coordinates (x,y,z) indicating its position relative to the seismic source. The dashed lines connect these points, representing the propagation path of the seismic wave, including direct and reflected waves. The "±" marks in the figure indicate the phase or direction of propagation of the wave.

[0118] Set up the internal layered model, specifically: include setting up the layered model of the Earth's interior. Include information such as source depth, source location, receiver depth, etc. Then calculate the thickness and depth of the layers. And discretize the complex layered model into a series of uniform sub-layers to facilitate numerical calculation. Make the subsequent seismic wave propagation calculation more feasible.

[0119] Calculate and store the table of Bessel functions, specifically: calculate and store the table of Bessel functions. Bessel functions can be more effectively calculated through the discretization of complex medium models.

[0120] Set the parameters of the seismic source, specifically: these parameters describe the type and intensity of the seismic source. The subroutine determines the type of seismic source according to the value of the stress function array, and sets the stress function value in the stress function derivative array accordingly. These stress functions are used to describe the moment of force of the seismic source in each direction.

[0121] Calculate the matrix of seismic wave propagation, specifically: use two different model operators to calculate the response function of different seismic waves near the source, calculate the matrix of seismic wave propagation, which describes the propagation of waves between different medium layers. Use two different model operators to calculate the response function of different seismic waves near the source.

[0122] Perform wave number integration to convert seismic waves from frequency domain to spatial domain, specifically: use the Thomson-Haskell propagation algorithm to calculate the propagation of seismic waves in a multi-layered elastic crust. This algorithm simulates the process of seismic wave propagation from the source point to the far field through iteration or recursion, taking into account the physical properties of each layer of medium. This step ensures the accuracy and stability of the calculation, especially when dealing with complex crustal structures. Convert the solution in the wave number domain back to the spatial domain. This step is achieved through inverse Hankel transformation, which converts the calculation results from the wave number domain back to the actual spatial domain, generating displacement vectors, strain tensors, stress tensors, and vertical tilt results. Store the calculated Green's function values in an array, which are subsequently used to simulate the propagation of seismic waves inside the Earth.

[0123] The kernel function calculation is performed, specifically, the response of a specific point in the crust under the action of a given seismic source is described, and a linear equation set is solved to calculate the total deformation caused by the seismic source in the crust. The total deformation caused by the seismic source in the crust is calculated through linear superposition (convolution integral). This step superimposes multiple basic solutions to generate the final deformation field, provides detailed information of the surface displacement, strain and stress distribution caused by the earthquake, and outputs the customized Green function of the corresponding model, three models respectively being a compensated linear vector doublet model, a dip-slip model and a strike-slip model.

[0124] In an optional embodiment, by inputting relevant crust parameters into a layered plane dislocation model and performing discretization processing, the non-homogeneity of the crust can be more accurately reflected, and the accuracy of the model is improved. The use of the Thomson-Haskell propagation algorithm and the inverse Hankel transform enables the propagation characteristics of seismic waves in complex media to be efficiently calculated, providing key support for the simulation of the earthquake rupture process. The total deformation obtained through linear superposition and the earthquake rupture fault model clearly show the dislocation distribution and rupture propagation path in the earthquake rupture process.

[0125] Optionally, the inversion of the earthquake rupture fault model by the gradient algorithm to obtain the earthquake rupture surface of the preset region comprises:

[0126] Model data is obtained according to the earthquake rupture fault model;

[0127] The model data is geometrically corrected, and the fault structure of the preset region is obtained by grid division of the fault of the preset region.

[0128] Model calculation is performed according to the fault structure, and the earthquake rupture surface is obtained according to the calculation result of the model calculation.

[0129] Specifically, the process of inverting the seismic rupture fault model to obtain the seismic rupture surface through the gradient algorithm is a complex step that combines numerical simulation and optimization algorithms. First, the model data is obtained based on the constructed seismic rupture fault model, which includes the geometric characteristics of the fault, the dislocation distribution, and the physical parameters related to seismic wave propagation. For example, in seismic research, the model data may involve the strike, dip, and slip angle of the fault, as well as the dislocation amount at different depths. Second, the model data is geometrically corrected to ensure that the geometric characteristics of the fault are consistent with the actual geological conditions. At the same time, the fault in the preset area is meshed, which discretizes the complex fault structure into multiple small units for more detailed numerical calculation. The accuracy of meshing directly affects the accuracy of the inversion result. Finally, the model calculation is performed based on the meshed fault structure, and the gradient algorithm is used to invert the calculation results. The gradient algorithm optimizes the model parameters through iteration, gradually reduces the difference between the model prediction and the actual observation data, and finally obtains the seismic rupture surface. This process can clearly reveal the dislocation distribution and rupture propagation path on the fault surface during the seismic rupture process, providing key information for seismic mechanism research.

[0130] In the preferred embodiment of the present application, the gradient algorithm is used to invert the seismic rupture surface, and the gradient algorithm is used to optimize the model parameters through iteration, gradually reducing the difference between the model prediction and the actual observation data, and finally obtaining the seismic rupture surface. This process can clearly reveal the dislocation distribution and rupture propagation path on the fault surface during the seismic rupture process, providing key information for seismic mechanism research. Figure 7 As shown in the workflow diagram of the inversion algorithm, the working process of the seismic rupture surface inversion algorithm is exemplarily given:

[0131] Obtain the seismic rupture fault model data; perform data geometric correction and mesh the fault; data geometric correction to ensure data accuracy and consistency, fault meshing helps to more accurately describe the seismic fault structure; calculate the seismic fault dip, calculate the slip gradient loss function and the slip gradient loss function; read the Green function, calculate the fault slip compensation point source, calculate the maximum gradient of the loss function, specifically, read the Green function to provide the basic model parameters of the seismic model; calculate the fault slip compensation point source to handle the complexity of the fault; calculate the maximum gradient of the loss function; calculate the slip model displacement field, calculate the sub-region slip gradient, and obtain the model fitting result, specifically, calculate the slip model displacement field to provide the fitting result of the model; calculate the sub-region slip gradient; perform seismic displacement processing to obtain the calculation model, specifically, performing seismic displacement processing can ensure that the seismic displacement is properly processed. Calculate the slip gradient of the loss function to optimize the convergence of the process; perform inversion calculation based on the obtained model, specifically, the program is responsible for performing inversion calculation based on the given data and model parameters, using the inversion main program to coordinate and call other subprograms to complete the entire inversion process, and output the fitting result and residual error of the inversion.

[0132] In optional embodiments, by performing geometric correction and meshing on the seismic rupture fault model, the geometric characteristics of the model can be ensured to be highly consistent with the actual geological conditions, while the accuracy and efficiency of numerical calculation are improved. The use of gradient algorithm further optimizes the inversion process, and by iteratively adjusting the model parameters, the actual observation data can be more accurately fitted, so as to obtain a high-precision seismic rupture surface.

[0133] Optionally, the model calculation according to the fault structure, and the inversion according to the calculation result of the model calculation to obtain the seismic rupture surface, comprises:

[0134] The model calculation according to the fault structure obtains the seismic fault dip angle, the slip gradient loss function and the sliding gradient loss function of the preset area, and calculates the fault slip compensation point source, the loss function maximum gradient, the sliding model displacement field and the sub-area sliding gradient.

[0135] The seismic fault dip angle, the slip gradient loss function, the sliding gradient loss function, the sliding compensation point source, the loss function maximum gradient, the sliding model displacement field and the sub-area sliding gradient are taken as the calculation result, and the calculation result is inverted by an analytical algorithm of surface deformation caused by internal shear and tensile faults to obtain the seismic rupture surface.

[0136] Specifically, a series of key parameters need to be obtained through model calculation first, including the seismic fault dip angle, the slip gradient loss function, the sliding gradient loss function, the fault slip compensation point source, the loss function maximum gradient, the sliding model displacement field and the sub-area sliding gradient. These parameters reflect the geometric characteristics of the fault and the physical properties in the seismic rupture process. Among them, the seismic fault dip angle describes the inclination of the fault, the slip gradient loss function and the sliding gradient loss function are used to quantify the inhomogeneity of fault slip, and the fault slip compensation point source is used to describe the distribution characteristics of fault slip. Through the calculation of these parameters, the dynamic changes of the fault in the seismic rupture process can be comprehensively understood. Subsequently, these calculation results are taken as input, and an analytical algorithm of surface deformation caused by internal shear and tensile faults is used for inversion. This analytical algorithm can analyze the sliding characteristics of the fault in the seismic rupture process and its influence on the ground deformation through mathematical model, so as to obtain the seismic rupture surface. This process not only considers the physical process inside the fault, but also combines the ground observation data, so that the inversion result can more accurately reflect the actual seismic rupture situation.

[0137] In optional embodiments, by calculating key parameters such as seismic fault dip angle, landslide gradient loss function, etc., the complex physical phenomena in the process of seismic rupture can be comprehensively captured, providing rich information for inversion. The use of analytical algorithms further improves the accuracy of inversion, enabling the characteristics of the seismic rupture surface to be clearly revealed.

[0138] Optionally, the analyzing the seismic rupture surface to obtain a coseismic displacement field comprises:

[0139] Discretizing the seismic rupture surface into a plurality of discrete point double-couple sources;

[0140] Analyzing according to the discrete point double-couple sources to determine the coseismic displacement field.

[0141] Specifically, the seismic rupture surface is discretized into a plurality of discrete point double-couple sources. This process is to decompose the complex seismic rupture surface into a series of simple point sources, each representing a small area on the rupture surface and assuming that it releases energy in the form of a double-couple source. The double-couple source model, as a commonly used method for describing seismic rupture, can effectively simulate the shear and tensile rupture characteristics on the fault plane during the seismic rupture process. Through discretization, the complex rupture surface problem can be converted into a superposition problem of multiple point sources, facilitating analytical calculation. Subsequently, according to the discrete point double-couple sources, the coseismic displacement field is determined through analytical calculation. Analytical calculation is based on seismic wave propagation theory and Green's function, by calculating the displacement response caused by each discrete point double-couple source on the ground, and linearly superimposing these responses, the coseismic displacement field caused by the entire seismic rupture surface is finally obtained. This process can describe the displacement changes on the ground during an earthquake in detail, including the distribution of horizontal and vertical displacement. For example, in earthquake research, this method can clearly show the deformation characteristics of the ground during the seismic rupture process, providing key data support for earthquake disaster assessment and earthquake engineering.

[0142] In optional embodiments, by discretizing the seismic rupture surface into a plurality of discrete point double-couple sources, the complex rupture surface problem can be effectively simplified, making it easier to analyze and calculate. The use of the double-couple source model further improves the accuracy of the description of the characteristics of the seismic rupture, making the calculation results of the coseismic displacement field more accurate.

[0143] Optionally, the analyzing according to the discrete point double-couple sources to determine the coseismic displacement field comprises:

[0144] Obtaining a general equation in the wave number domain according to the distribution and parameters of the discrete point double-couple sources through Hankel custom transformation;

[0145] Determining a special solution that satisfies the source and boundary conditions according to the general equation through the Thomson-Haskell propagator algorithm;

[0146] convert the special solution into a solution in the spatial domain by inverse Hankel;

[0147] perform linear superposition operation according to the solution in the spatial domain to obtain seismic variation data;

[0148] obtain the coseismic displacement field according to the seismic variation data.

[0149] Specifically, by Hankel customization transformation, the problem is converted from the spatial domain to the wave number domain according to the distribution and parameters of the discrete point double couple source, and a conventional equation is obtained. This transformation takes advantage of the Hankel transformation in dealing with radial symmetric problems, and can effectively simplify the mathematical description of seismic wave propagation. Subsequently, the Thomson-Haskell propagator algorithm is used to solve the special solution that satisfies the source and boundary conditions according to the conventional equation in the wave number domain. The Thomson-Haskell algorithm is a classic numerical method for seismic wave propagation, which is particularly suitable for wave propagation problems in layered media, and can efficiently calculate the propagation path and response of seismic waves in complex media. Then, the special solution in the wave number domain is converted into a solution in the spatial domain by inverse Hankel transformation, which restores the abstract wave number domain results to actual physical quantities and provides a basis for subsequent analysis. Finally, linear superposition operation is performed according to the solution in the spatial domain to obtain seismic variation data, and the coseismic displacement field is obtained accordingly. The application of linear superposition principle here is based on the linear characteristics of seismic wave propagation. By superimposing the responses of multiple discrete point double couple sources, the influence of the seismic rupture process on the ground surface displacement can be completely described. For example, in seismic research, this method can accurately calculate the displacement changes of each point on the ground surface when an earthquake occurs, providing key data support for earthquake disaster assessment and seismic engineering.

[0150] In the preferred embodiment of the present application, a customized Green algorithm or an analytical algorithm for surface deformation caused by internal shear and tensile faults in space is used to simulate the interaction between seismic waves and underground structures. The finite rectangular fault is discretized into multiple discrete point double couple sources, the motion partial differential equation is converted into a conventional equation in the wave number domain by Hankel customization transformation, and the special solution that satisfies the source and boundary conditions is obtained by Thomson-Haskell propagator algorithm. The entire process is handled by analytical solution, and numerical calculation is required when inverse Hankel customization transformation is performed. The deformation field caused by the earthquake is calculated by linear superposition (convolution integral), and the displacement vector, strain tensor, stress tensor, and vertical tilt are output, and the coseismic displacement field on the ground surface is obtained.

[0151] In optional embodiments, the propagation of seismic waves in complex media can be efficiently handled by the combined use of Hankel transform and Thomson-Haskell algorithm, ensuring the accuracy and reliability of the calculation results. The use of inverse Hankel transform further converts the solution in the wave number domain into actual physical quantities in the spatial domain, making the results more practically meaningful. The final co-seismic displacement field obtained by linear superposition can clearly show the deformation distribution of the ground surface at the time of the earthquake, providing an important theoretical basis for earthquake disaster assessment, seismic engineering design, and earthquake emergency response. This method not only improves the scientific nature of earthquake research, but also enhances the precision and effectiveness of earthquake disaster response, providing strong support for earthquake science and engineering applications.

[0152] In combination Figure 9 As shown in the embodiments of the present application, a seismic rupture fault model and a co-seismic displacement field construction system are provided, which include:

[0153] A data acquisition unit is configured to acquire GNSS data, strong motion instrument data, and accelerometer data of a preset area.

[0154] A data fusion unit is configured to perform fusion processing on the GNSS data, the strong motion instrument data, and the accelerometer data to obtain multi-source seismic data of the preset area.

[0155] A first inversion unit is configured to perform seismic waveform inversion on the multi-source seismic data to obtain relevant crust parameters of the preset area.

[0156] A fault model construction unit is configured to combine the relevant crust parameters and calculate Green's function using a layered plane dislocation model according to the multi-source seismic data to obtain a seismic rupture fault model of the preset area.

[0157] A second inversion unit is configured to perform inversion on the seismic rupture fault model by a gradient algorithm to obtain a seismic rupture surface of the preset area.

[0158] A displacement field construction unit is configured to analyze the seismic rupture surface to obtain a co-seismic displacement field.

[0159] The earthquake rupture fault model and the construction system of the coseismic displacement field of the present application eliminate the limitations of a single data source through the fusion processing of multi-source data. For example, the high-precision displacement information of GNSS data is combined with the high-frequency seismic waveform information of strong motion instrument and accelerometer data, which can more comprehensively reflect the dynamic changes of the ground surface and underground when the earthquake occurs, fully exert the advantages of each data source, make up for the shortcomings of a single data source, and more comprehensively reflect the dynamic changes of the ground surface and underground during the earthquake process. Secondly, the application of the seismic waveform inversion technology further optimizes the estimation of the crustal parameters, providing a more accurate physical basis for subsequent model calculation. Thirdly, the Green function calculation based on the layered plane dislocation model considers the complex layered structure of the crust, which can more realistically simulate the propagation and deformation effect of seismic waves in the multi-layer medium, thereby improving the accuracy of the earthquake rupture fault model. Finally, the gradient algorithm is used to invert the earthquake rupture fault model, and the analytical algorithm is used to calculate the coseismic displacement field, which further improves the fitting accuracy and reliability of the model. The method of combining multi-source data fusion with advanced algorithms in the present application not only can effectively reduce the influence of noise interference and data error, but also can more accurately reveal the earthquake mechanism and ground surface deformation characteristics, providing more accurate and reliable data support for seismology research, earthquake disaster assessment and earthquake prevention and disaster reduction.

[0160] Although the present application discloses as above, the protection scope of the present application is not limited to this. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application, and these changes and modifications will fall within the protection scope of the present application.

Claims

1. A method for constructing an earthquake rupture fault model and a co-seismic displacement field, characterized in that: include: Obtain GNSS data, strong motion data and accelerometer data for a preset area; fusing the GNSS data, the strong motion detector data, and the accelerometer data to obtain multi-source seismic data for the preset area; Performing seismic waveform inversion on the multi-source seismic data to obtain relevant crustal parameters of the preset area; In combination with the relevant crustal parameters, Green's function is calculated using a layered plane dislocation model according to the multi-source seismic data to obtain an earthquake rupture fault model of the preset area; Inverting the earthquake rupture fault model using a gradient algorithm to obtain an earthquake rupture surface in the preset area; Analyzing the earthquake rupture surface to obtain a co-seismic displacement field includes: discretizing the earthquake rupture surface into a plurality of discrete point double-dual sources; Through the Hankel custom transformation, according to the distribution and parameters of the discrete point double-even source, the conventional equation in the wave number domain is obtained; Determining a special solution satisfying source and boundary conditions according to the general equations using the Thomson–Haskell propagator algorithm; Convert the special solution into a solution in the spatial domain by inverse Hankel; Performing a linear superposition operation based on the solution in the spatial domain to obtain seismic change data; The co-seismic displacement field is obtained according to the seismic change data.

2. The method for constructing an earthquake rupture fault model and a coseismic displacement field according to claim 1, characterized in that: The obtaining of GNSS data, strong motion detector data, and accelerometer data of a preset area includes: Determining the surface position and displacement information of the preset area by satellite positioning technology, and using the surface position and the displacement information as the GNSS data; Determining the ground acceleration, ground velocity, and ground displacement of the preset area by using data recorded by a strong motion detector, and using the ground acceleration, ground velocity, and ground displacement as the strong motion detector data; The acceleration change values ​​of the ground and buildings in the preset area are determined by an acceleration sensor, and the acceleration change values ​​are used as the accelerometer data.

3. The method for constructing an earthquake rupture fault model and a coseismic displacement field according to claim 1, characterized in that: The fusing the GNSS data, the strong motion detector data, and the accelerometer data to obtain multi-source seismic data of the preset area includes: performing time alignment processing and spatial coordinate conversion processing on the GNSS data, the strong motion motion instrument data, and the accelerometer data according to the same time reference and space reference, respectively, to obtain the GNSS data, the strong motion motion instrument data, and the accelerometer data with time and space differences eliminated; assigning weights to the GNSS data, the strong motion sensor data, and the accelerometer data respectively to eliminate temporal and spatial differences through a weighted data fusion filtering algorithm; The multi-source seismic data is obtained by fusing the GNSS data, the strong motion detector data and the accelerometer data according to the weights corresponding to the data that eliminate the temporal and spatial differences.

4. The method for constructing an earthquake rupture fault model and a coseismic displacement field according to claim 1, characterized in that: The performing seismic waveform inversion on the multi-source seismic data to obtain relevant crustal parameters of the preset area includes: Perform forward simulation based on the multi-source seismic data and a preset underground medium model to generate a theoretical seismic waveform for the preset area; comparing the theoretical seismic waveform with the observed waveform of the preset area to determine an error between the theoretical seismic waveform and the observed waveform; Iterative optimization is performed based on the error to obtain the relevant crustal parameters.

5. The method for constructing an earthquake rupture fault model and a co-seismic displacement field according to claim 1, characterized in that: The method of calculating the Green's function using a layered plane dislocation model based on the multi-source seismic data in combination with the relevant crustal parameters to obtain an earthquake rupture fault model of the preset area includes: Inputting the relevant crustal parameters into the layered plane dislocation model, and setting the Earth interior layered model of the preset area according to the relevant crustal parameters; Discretizing the Earth's interior layered model to obtain multiple sublayers; The propagation data of seismic waves in the sub-layers are calculated in sequence by using the Thomson-Haskell propagation algorithm, and the Green's function value of each sub-layer is obtained by using the inverse Hankel transform; Determining the total deformation caused by the seismic waves in the Earth's crust by linear superposition; The earthquake rupture fault model is obtained according to the total deformation and the Green's function value.

6. The method for constructing an earthquake rupture fault model and a coseismic displacement field according to claim 1, characterized in that: The inversion of the earthquake rupture fault model by a gradient algorithm to obtain the earthquake rupture surface of the preset area includes: acquiring model data according to the earthquake rupture fault model; Performing geometric correction on the model data and meshing the faults in the preset area to obtain the fault structure of the preset area; Model calculation is performed according to the fault structure, and then inversion is performed according to the calculation results of the model calculation to obtain the earthquake rupture surface.

7. The method for constructing an earthquake rupture fault model and a coseismic displacement field according to claim 6, characterized in that: The performing of model calculation according to the fault structure and then performing inversion according to the calculation results of the model calculation to obtain the earthquake rupture surface includes: Performing model calculation based on the fault structure to obtain the earthquake fault dip, landslide gradient loss function, and sliding gradient loss function of the preset area, and calculating the fault sliding compensation point source, the maximum gradient of the loss function, the sliding model displacement field, and the sub-area sliding gradient; The earthquake fault dip, the landslide gradient loss function, the sliding gradient loss function, the sliding compensation point source, the maximum gradient of the loss function, the sliding model displacement field, and the sub-region sliding gradient are used as the calculation results. The calculation results are inverted through an analytical algorithm of surface deformation caused by internal shear and tensile faults in space to obtain the earthquake rupture surface.

8. A system for constructing an earthquake rupture fault model and a co-seismic displacement field, characterized in that: include: A data acquisition unit, used to acquire GNSS data, strong motion data and accelerometer data of a preset area; a data fusion unit, configured to fuse the GNSS data, the strong motion detector data, and the accelerometer data to obtain multi-source seismic data of the preset area; a first inversion unit, configured to perform seismic waveform inversion on the multi-source seismic data to obtain relevant crustal parameters of the preset area; a fault model construction unit, configured to calculate a Green's function based on the multi-source seismic data using a layered plane dislocation model in combination with the relevant crustal parameters, to obtain an earthquake rupture fault model of the preset area; a second inversion unit, configured to invert the earthquake rupture fault model using a gradient algorithm to obtain an earthquake rupture surface in the preset area; The displacement field construction unit is used to analyze the earthquake rupture surface to obtain a co-seismic displacement field, including: discretizing the earthquake rupture surface into a plurality of discrete point double-dual sources; Through the Hankel custom transformation, according to the distribution and parameters of the discrete point double-even source, the conventional equation in the wave number domain is obtained; Determining a special solution satisfying source and boundary conditions according to the general equations using the Thomson–Haskell propagator algorithm; Convert the special solution into a solution in the spatial domain by inverse Hankel; Performing a linear superposition operation based on the solution in the spatial domain to obtain seismic change data; The co-seismic displacement field is obtained according to the seismic change data.

Images