Near-fault broadband strong vibration simulation method

Abstract

The invention discloses a near-fault broadband strong vibration simulation method, which comprises the following steps: S1, data collection and preprocessing: collecting a small vibration record in a research area as a source of an empirical Green function, and simultaneously obtaining information such as a geologic structure and stratum parameters of the area for model construction of a discrete wave number method; s2, correcting an empirical Green function, and selecting records similar to a target earthquake focus mechanism from the collected small earthquake records; and S3, calculating based on a geological structure and stratum parameters of a research area by using a discrete wavenumber method. According to the near-fault broadband strong vibration simulation method, the advantages of the two methods are fused, the description of the discrete wave number method on the medium structure at the low frequency band is accurate, the corrected empirical Green function can consider more actual factors at the high frequency band, accurate simulation of the broadband is achieved, and small vibration records are used as the empirical Green function, so that the simulation accuracy of the near-fault broadband strong vibration simulation method is improved. And information such as a real field effect is included, so that a simulation result is closer to an actual seismic oscillation condition.

Description

Technical Field

[0001] The present invention relates to the technical field of geotechnical earthquake engineering, and specifically to a method for simulating near-fault broadband strong ground motion. Background Art

[0002] Earthquakes can trigger strong ground motions, resulting in serious disasters. There are many factors affecting the characteristics and intensity of ground motions, including source rupture, the propagation of seismic waves in the crust, and local topography and near-surface soil geological conditions (i.e., site effects). To better understand the formation mechanism of earthquake disasters, predict the impact of earthquakes on specific regions, and provide a basis for the seismic design of engineering structures, it is necessary to accurately present the ground motion during an earthquake through simulation methods. Near-fault ground motions have unique characteristics, such as the forward directivity effect and the slip-and-thrust effect of fault rupture, which can cause large velocity pulses and permanent ground displacements in near-fault ground motions. These characteristics, combined with the coupled effect of the temporal and spatial variations of ground motions themselves, are extremely likely to cause serious damage to long-span structures such as bridge structures or pipelines. The near-fault strong ground motion is significantly destructive to engineering structures due to its broadband characteristics (the coexistence of high-frequency pulse effects and low-frequency long-period vibrations). The empirical Green's function method uses small earthquake records to synthesize strong ground motions of large earthquakes and is reliable in high-frequency simulations (>1 Hz), but the low-frequency part (<1 Hz) is limited by the simplification of the source model and path effects and has insufficient accuracy. The discrete wavenumber method performs excellently in low-frequency long-period simulations but has a high computational complexity.

[0003] Traditional ground motion simulation methods often simplify the site effects. For example, the site is regarded as a uniform horizontally layered medium, ignoring the influence of factors such as local topographic changes and groundwater level fluctuations on the propagation and amplification of seismic waves, resulting in a large deviation between the simulation results and the actual situation under some special site conditions. For complex geological structures, such as fractured zones and soft interlayers near faults, traditional simulation methods may not be able to accurately describe their mechanical properties and the scattering and absorption effects on seismic waves, which may make the simulated ground motion inconsistent with the actual situation in terms of frequency components and amplitudes. Especially in the near-fault area, the influence of these geological structures is more significant, and the simplification process will reduce the simulation accuracy. Therefore, we propose a method for simulating near-fault broadband strong ground motion to solve the problems raised above. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for simulating near-fault broadband strong ground motion to solve the problem that traditional ground motion simulation methods often simplify the site effects, for example, regarding the site as a uniform horizontally layered medium, ignoring the influence of factors such as local topographic changes and groundwater level fluctuations on the propagation and amplification of seismic waves, resulting in a large deviation between the simulation results and the actual situation under some special site conditions as proposed in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions: A near-fault broadband strong ground motion simulation method, comprising the following steps: S1. Data collection and preprocessing, collecting small earthquake records in the study area as the source of empirical Green's functions, and at the same time obtaining information such as the geological structure and formation parameters of the area for the construction of the discrete wavenumber method model; S2. Modifying the empirical Green's function, selecting records similar to the target earthquake source mechanism from the collected small earthquake records; S3. Discrete wavenumber method calculation, establishing a layered medium model based on the geological structure and formation parameters of the study area; S4. High and low frequency fusion technology, taking 1 Hz as the frequency boundary, realizing seamless connection of high and low frequency simulation results through a smooth transition function, introducing a spectrum matching algorithm to correct the spectral characteristics of the synthesized time history to ensure broadband physical consistency; S5. Filtering and superposition, respectively performing filtering processing on the low frequency and high frequency ground motion components; S6. Parametric characterization of site effects, constructing a site transfer function based on soil layer dynamic parameters to quantify the modulation effect of nonlinear effects on high frequency pulses and low frequency vibrations.

[0006] Preferably, in S1, preprocessing the small earthquake records includes operations such as baseline correction and filtering to improve data quality. And according to the research purpose and fault distribution, select a suitable monitoring area, and arrange seismic monitoring stations in this area. The distribution of the stations should cover the area around the fault as much as possible to obtain comprehensive ground motion data. Then, according to the required frequency range and accuracy requirements of the simulation, select a suitable seismic sensor to ensure that broadband strong ground motion signals can be accurately collected.

[0007] Preferably, in S1, removing noise and outliers in the data to improve data quality. Common methods include limit filtering, median filtering, mean filtering, etc. For example, limit filtering limits the change range of the signal by setting a threshold, and the signal exceeding the threshold is weakened or replaced with a neighboring value, which can effectively suppress the influence of abnormal waveforms. And due to factors such as amplifier zero drift, unstable low-frequency performance of sensors, and environmental interference, the collected signal often deviates from the baseline and even changes with time, which is called the trend term of the signal and needs to be removed. According to the simulation frequency range, select a suitable filter to filter the data, such as low-pass filtering, high-pass filtering, band-pass filtering, etc. The low-pass filter can remove high-frequency noise in the signal, the high-pass filter can remove the low-frequency trend term, and the band-pass filter can select to retain the signal within a specific frequency range.

[0008] Preferably, in step S2, small earthquake records similar to the target earthquake source mechanism are selected from the earthquake database or actual monitoring data. The epicenter location of the small earthquake should be as close as possible to the fault area of the target earthquake, and parameters such as the magnitude and focal depth of the small earthquake also need to have a certain similarity to the target earthquake. Usually, a time history waveform of a period of time after the P-wave onset is intercepted as the Green's function for synthesizing the large earthquake, and the intercepted duration can be determined according to specific circumstances, generally several tens of seconds, such as about 40 s. The intercepted data segment is subjected to baseline correction and filtering. Baseline correction can remove the DC component in the signal and make the baseline of the waveform return to zero.

[0009] Preferably, in step S2, the source parameters of the small earthquake are adjusted and corrected to make them closer to the situation of the target large earthquake. This includes considering factors such as rupture scale, stress drop, asperity-related parameters, and rupture velocity. For example, after determining the set magnitude, the seismic moment is obtained according to the empirical relationship between magnitude and seismic moment, and then the fault rupture area is determined with reference to the empirical relationship between rupture area and seismic moment. For the asperities on the fault plane, parameters such as their shape, quantity, and size can be set. At the same time, according to the ratio of stress drops between large and small earthquakes and the relationship between the number of sub-faults divided, parameters such as the stress drop of the small earthquake are adjusted. In addition, for the rupture velocity, the correlation coefficient with the shear velocity needs to be considered, and the value range of the correlation coefficient is determined according to the type of the target earthquake (sub-shear rupture or super-shear rupture). Since there may be differences in the propagation paths and site conditions between small earthquakes and the target large earthquake, the small earthquake records need to be adjusted accordingly. The propagation medium model can be established by analyzing information such as the geological structure and topography of the study area.

[0010] Preferably, in step S3, the discrete wavenumber method is used to calculate the theoretical Green's function. This method can effectively calculate the wave propagation problem in an elastic half-space and obtain the wave field response at different frequencies. Through discrete wavenumber integration, the seismic wave propagation function under given source and medium conditions is solved. However, the discrete wavenumber method may have certain limitations in dealing with complex geological structures and near-field effects, so it needs to be combined with the empirical Green's function to make up for these deficiencies.

[0011] Preferably, in step S4, the corrected empirical Green's function and the theoretical Green's function calculated by the discrete wavenumber method are fused. A common approach is to give play to the advantages of the two methods in different frequency bands. For example, in the low-frequency band (such as 0.05 - 1.0 Hz), the result calculated by the discrete wavenumber method is adopted because the discrete wavenumber method can better consider the overall structure of the medium and the wave propagation characteristics at low frequencies; in the high-frequency band (such as 1.0 - 25.0 Hz), the corrected empirical Green's function is adopted because the empirical Green's function contains the complexity information of the real source rupture process, propagation medium, and shallow surface site response, and can more accurately simulate the details of ground motion at high frequencies.

[0012] Preferably, in S5, the two parts of acceleration waveforms are superimposed at an appropriate cross - frequency (such as around 1.0 Hz) in the frequency domain to obtain a broadband ground motion acceleration waveform.

[0013] Preferably, in S6, an empirical relationship between the site effect parameters, the site geological conditions, and the ground motion parameters is established based on the simulation results and actual earthquake records.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This near - fault broadband strong ground motion simulation method adopts a novel structural design, and its specific content is as follows: (1) The empirical Green's function method uses small - earthquake record data as the Green's function, which contains information on the true source rupture process, the propagation medium, and the complexity of the shallow - site response. The discrete wavenumber method can accurately handle the propagation of seismic waves in layered media. The combination of the two can integrate the advantages of the two methods and more accurately simulate the complex characteristics of near - fault broadband ground motion, including the propagation, reflection, refraction of seismic waves, and site effects, thereby improving the simulation accuracy.

[0015] (2) The discrete wavenumber method has strong capabilities in dealing with different geological structures and soil layer characteristics. The empirical Green's function method can adapt to different geological conditions and source mechanisms by selecting different small - earthquake records as the Green's function. The combined method can better handle various complex geological situations. Whether it is a hard site or a soft site, as well as different stratigraphic distributions and geological structures, it can more accurately simulate ground motion.

[0016] (3) Near - fault ground motion usually contains rich frequency components and has a greater destructive effect on engineering structures. This combined method can perform efficient simulations in a wide frequency band. It can not only accurately simulate the low - frequency components for studying the seismic response of long - period structures but also precisely handle the high - frequency components to provide reliable ground motion input for studying the local response of short - period structures and geotechnical bodies.

[0017] (4) The discrete wavenumber method has a high computational efficiency and can quickly solve the problem of seismic wave propagation in layered media. The empirical Green's function method can also reduce the amount of calculation by using the information of small - earthquake records when synthesizing large - earthquake vibrations. After the combination of the two, the computational efficiency can be improved while ensuring the simulation accuracy, saving calculation time and cost, making it more suitable for large - scale ground motion simulations and seismic hazard assessments in practical engineering.

[0018] (5) can not only give the acceleration, velocity and displacement time histories of ground motions, but also provide detailed information such as the spectral characteristics and spatial distribution of ground motions. These information are of great value for the seismic design of engineering structures, earthquake disaster assessment and seismological research, and can help engineers and researchers better understand the characteristics and action mechanisms of ground motions, so as to adopt more effective seismic measures and research methods. Description of the Drawings

[0019] Figure 1 Schematic diagram of the working steps of the strong ground motion simulation method of the present invention; Figure 2 Schematic diagram of the time-domain waveform of the present invention; Figure 3 Schematic diagram of the acceleration waveform of near-fault broadband ground motion of the present invention. Detailed Embodiments

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Please refer to Figures 1-3, the present invention provides a technical solution: a near-fault broadband strong ground motion simulation method, comprising the following steps: S1. Data collection and preprocessing, collecting small earthquake records in the study area as the source of empirical Green's functions, and at the same time obtaining information such as the geological structure and formation parameters of the area for the model construction of the discrete wavenumber method. In S1, preprocessing of the small earthquake records includes operations such as baseline correction and filtering to improve data quality. And according to the research purpose and fault distribution, select a suitable monitoring area, and arrange seismic monitoring stations in this area. The distribution of the stations should cover the area around the fault as much as possible to obtain comprehensive ground motion data. Then, according to the required frequency range and accuracy requirements of the simulation, select a suitable seismic sensor to ensure that broadband strong ground motion signals can be accurately collected; In S1, noise and outliers in the data are removed to improve data quality. Common methods include clipping filtering, median filtering, mean filtering, etc. For example, clipping filtering limits the change range of the signal by setting a threshold, and the signal exceeding the threshold is weakened or replaced with a neighboring value, which can effectively suppress the influence of abnormal waveforms. And due to factors such as amplifier zero drift, unstable low-frequency performance of the sensor, and environmental interference, the collected signal often deviates from the baseline and even changes with time, which is called the trend term of the signal and needs to be removed. According to the simulation frequency range, select a suitable filter to filter the data, such as low-pass filtering, high-pass filtering, band-pass filtering, etc. The low-pass filter can remove high-frequency noise in the signal, the high-pass filter can remove low-frequency trend terms, and the band-pass filter can select to retain signals within a specific frequency range; S2. Modify the empirical Green's function, select records similar to the target earthquake source mechanism from the collected small earthquake records. In S2, select small earthquake records similar to the target earthquake source mechanism from the earthquake database or actual monitoring data. The epicenter position of the small earthquake should be as close as possible to the fault area of the target earthquake, and parameters such as the magnitude and focal depth of the small earthquake also need to have a certain similarity to the target earthquake. Usually, a time history waveform of a period of time after the P-wave onset is intercepted as the Green's function for synthesizing the large earthquake, and the intercepted duration can be determined according to specific circumstances, generally dozens of seconds, such as about 40 s. Perform baseline correction and filtering on the intercepted data segment. Baseline correction can remove the DC component in the signal and make the baseline of the waveform return to zero;In S2, the source parameters of small earthquakes are adjusted and corrected to make them closer to the situation of the target large earthquake. This includes considering factors such as rupture scale, stress drop, asperity-related parameters, rupture velocity, etc. For example, after determining the magnitude setting, the seismic moment is obtained according to the empirical relationship between magnitude and seismic moment, and then the fault rupture area is determined with reference to the empirical relationship between rupture area and seismic moment. For the asperities on the fault plane, parameters such as their shape, quantity, and size can be set. At the same time, according to the ratio of stress drops between large and small earthquakes and the relationship of the number of sub-faults divided, parameters such as the stress drop of small earthquakes are adjusted. In addition, for the rupture velocity, the correlation coefficient with the shear velocity needs to be considered, and the value range of the correlation coefficient is determined according to the type of target earthquake (sub-shear rupture or super-shear rupture). Since there may be differences in the propagation paths and site conditions between small earthquakes and the target large earthquake, the small earthquake records need to be adjusted accordingly. The propagation medium model can be established by analyzing information such as the geological structure and topography of the study area; In S3, the discrete wavenumber method is used for calculation. Based on the geological structure and formation parameters of the study area, a layered medium model is established. In S3, the discrete wavenumber method is used to calculate the theoretical Green's function, which can effectively calculate the wave propagation problem in an elastic half-space and obtain the wave field response at different frequencies; Through discrete wavenumber integration, the seismic wave propagation function under given source and medium conditions is solved; However, the discrete wavenumber method may have certain limitations in dealing with complex geological structures and near-field effects, so it needs to be combined with the empirical Green's function to make up for these deficiencies; In S4, the high-low frequency fusion technology is used. Taking 1 Hz as the frequency boundary, the seamless connection of high-frequency and low-frequency simulation results is achieved through a smooth transition function, and a spectrum matching algorithm is introduced to correct the spectrum characteristics of the synthesized time history to ensure broadband physical consistency. In S4, the corrected empirical Green's function is fused with the theoretical Green's function calculated by the discrete wavenumber method. A common approach is to give play to the advantages of the two methods in different frequency bands. For example, in the low-frequency band (such as 0.05 - 1.0 Hz), the results calculated by the discrete wavenumber method are adopted because the discrete wavenumber method can better consider the overall structure of the medium and the wave propagation characteristics at low frequencies; In the high-frequency band (such as 1.0 - 25.0 Hz), the corrected empirical Green's function is adopted because the empirical Green's function contains information on the complexity of the real source rupture process, propagation medium, and shallow surface site response, and can more accurately simulate the details of ground motion at high frequencies; In S5, filtering and superposition are carried out. The ground motion components of low frequency and high frequency are filtered respectively. In S5, the two parts of acceleration waveforms are superposed at an appropriate cross frequency (such as around 1.0 Hz) in the frequency domain to obtain the broadband ground motion acceleration waveform;S6. Parametric characterization of site effects, construction of a site transfer function based on soil layer dynamic parameters, quantification of the modulation effect of nonlinear effects on high-frequency pulses and low-frequency vibrations. In S6, based on the simulation results and actual seismic records, an empirical relationship is established between site effect parameters, site geological conditions, and ground motion parameters. For example, a regression equation is established for parameters such as the site amplification factor, soil shear wave velocity, overburden thickness, magnitude, and epicentral distance, and key parameters with a greater impact on site effects are selected from numerous parameters as the main indicators for parametric characterization of site effects.

[0022] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A near-fault broadband strong vibration simulation method, characterized in that: The following steps are involved: S1. Data collection and preprocessing: collect small earthquake records in the study area as the source of empirical Green's function, and obtain information such as the geological structure and formation parameters of the area for model construction of discrete wavenumber method; S2. Modify the empirical Green's function, and select records with similar focal mechanisms to the target earthquake from the collected small earthquake records; S3. Discrete wavenumber method calculation: establish a layered medium model based on the geological structure and formation parameters of the study area; S4, high-low frequency fusion technology, with 1Hz as the frequency boundary, realizes seamless connection of high-frequency and low-frequency simulation results through smooth transition function, introduces spectrum matching algorithm, corrects the spectrum characteristics of the synthesis time course, and ensures wide-band physical consistency; S5, filtering and superposition, filtering the low-frequency and high-frequency seismic motion components separately; S6, parametric characterization of site effects, constructing a site transfer function based on soil layer dynamic parameters, and quantifying the modulation effect of nonlinear effects on high-frequency pulses and low-frequency vibrations.

2. A near-fault broadband strong vibration simulation method according to claim 1, characterized in that: In the S1, the small earthquake records are preprocessed, including baseline correction, filtering and other operations to improve the data quality. According to the research purpose and the distribution of faults, a suitable monitoring area is selected, and seismic monitoring stations are arranged in the area. The distribution of the stations should cover the area around the fault as much as possible to obtain comprehensive seismic motion data. Then, according to the frequency range and accuracy requirements required for the simulation, suitable seismic sensors are selected to ensure that wide-band strong vibration signals can be accurately collected.

3. A near-fault broadband strong vibration simulation method according to claim 1, characterized in that: In the S1, noise and outliers in the data are removed to improve data quality. Common methods include limiting filtering, median filtering, mean filtering, etc. For example, limiting filtering limits the range of signal changes by setting a threshold. Signals exceeding the threshold are weakened or replaced by adjacent values, which can effectively suppress the influence of abnormal waveforms. Due to factors such as amplifier zero drift, unstable low-frequency performance of sensors, and environmental interference, the collected signals often deviate from the baseline and even change over time. This is called the trend item of the signal and needs to be removed. According to the simulated frequency range, a suitable filter is selected to filter the data, such as low-pass filtering, high-pass filtering, band-pass filtering, etc. Low-pass filters can remove high-frequency noise in the signal, high-pass filters can remove low-frequency trend items, and band-pass filters can choose to retain signals within a specific frequency range.

4. A near-fault broadband strong vibration simulation method according to claim 1, characterized in that: In the S2, small earthquake records with similar focal mechanisms to the target earthquake are selected from the earthquake database or actual monitoring data. The epicenter of the small earthquake should be as close as possible to the fault area of ​​the target earthquake, and the parameters such as the magnitude and focal depth of the small earthquake should also have a certain similarity with the target earthquake. Usually, a time waveform of a period of time after the initial movement of the P wave is intercepted as the Green's function of the synthetic large earthquake, and the intercepted time can be determined according to the specific situation, generally tens of seconds, such as about 40 seconds. The intercepted data segment is baseline corrected and filtered. The baseline correction can remove the DC component in the signal and return the baseline of the waveform to zero.

5. The method for simulating near-fault broadband strong vibrations according to claim 1, characterized in that: In the S2, the source parameters of the small earthquake are adjusted and corrected to make them closer to the target large earthquake, which includes considering factors such as rupture scale, stress drop, asperity-related parameters, rupture velocity, etc. For example, after the magnitude is set, the seismic moment is obtained according to the empirical relationship between the magnitude and the seismic moment, and the fault rupture area is determined by referring to the empirical relationship between the rupture area and the seismic moment. For the asperities on the fault plane, their shape, number, size and other parameters can be set. At the same time, according to the ratio of the stress drop of large and small earthquakes and the number of divided sub-faults, etc., the stress drop and other parameters of the small earthquake are adjusted. In addition, for the rupture velocity, its correlation coefficient with the shear velocity should be considered, and the value range of the correlation coefficient is determined according to the type of target earthquake (subshear rupture or supershear rupture). Since the propagation paths and site conditions of small earthquakes and target large earthquakes may be different, the small earthquake records need to be adjusted accordingly. The propagation medium model can be established by analyzing the geological structure, topography and other information of the study area.

6. A near-fault broadband strong vibration simulation method according to claim 1, characterized in that: In S3, the discrete wavenumber method is used to calculate the theoretical Green's function. This method can effectively calculate the wave propagation problem in the elastic half-space, obtain the wave field response at different frequencies, and solve the seismic wave propagation function under given source and medium conditions through discrete wavenumber integration. However, the discrete wavenumber method may have certain limitations when dealing with complex geological structures and near-field effects, so it needs to be combined with the empirical Green's function to make up for these shortcomings.

7. A near-fault broadband strong vibration simulation method according to claim 1, characterized in that: In S4, the modified empirical Green's function is integrated with the theoretical Green's function calculated by the discrete wavenumber method. A common practice is to take advantage of the two methods in different frequency bands. For example, the result calculated by the discrete wavenumber method is used in the low-frequency band (such as 0.05-1.0 Hz), because the discrete wavenumber method can better consider the overall structure of the medium and the propagation characteristics of the wave at low frequencies; the modified empirical Green's function is used in the high-frequency band (such as 1.0-25.0 Hz), because the empirical Green's function contains the complex information of the real source rupture process, the propagation medium and the response of the shallow site, and can more accurately simulate the details of the earthquake motion in the high-frequency band.

8. The method for simulating near-fault broadband strong vibrations according to claim 1, characterized in that: In S5, the two acceleration waveforms are superimposed at a suitable crossover frequency (eg, about 1.0 Hz) in the frequency domain to obtain a broadband earthquake acceleration waveform.

9. The method for simulating near-fault broadband strong vibrations according to claim 1, characterized in that: In S6, based on the simulation results and actual earthquake records, an empirical relationship between the site effect parameters and the site geological conditions and seismic motion parameters is established. For example, a regression equation is established between the site amplification factor and parameters such as soil shear wave velocity, cover layer thickness, magnitude, and epicentral distance. Key parameters that have a greater impact on site effects are selected from a large number of parameters as the main indicators for parametric characterization of site effects.

Images