A method for generating a fault structure earthquake scenario rupture process

Abstract

The present application relates to a kind of fault structure earthquake scenario rupture process generation method, comprising: S1, obtain the parameter related to earthquake as input parameter;S2, according to input parameter, using the spatial random field hybrid model of combining concave-convex body generates final slip distribution field;S3, according to final slip distribution field and focus position, the rupture time of each subfault is calculated;S4, according to final slip distribution field, the rise time of each subfault is calculated;S5, according to rupture time, rise time, using Gaussian function simulates seismic moment release process, the mean square relative error between the characteristic value of generated earthquake scenario and the characteristic value of real earthquake data is calculated, and return S2 is iteratively optimized, until error is less than preset value, generate focus model file.The present application can generate a large number of earthquake scenarios with statistical characteristics.

Description

Technical Field

[0001] This invention relates to the field of earthquake engineering technology, and in particular to a method for generating fault structure earthquake scenario rupture processes. Background Technology

[0002] Determining the ground motion parameters of an engineering site is the cornerstone of seismic design and safety assessment. Whether it is a major project such as a nuclear power plant, a large bridge, or a super high-rise building, or an ordinary civil building, its seismic fortification standards and structural design are highly dependent on the accurate prediction of the intensity and characteristics of the ground motion of possible future earthquakes.

[0003] Currently, known methods for constructing earthquake source models typically define the spatial distribution of slip, rupture time, and rise time on the fault plane through a series of mathematical and physical assumptions. However, the probabilistic methods commonly used in seismic design cannot take into account the fault rupture process. Furthermore, ground motion at near-fault engineering sites is significantly affected by the fault rupture process. Therefore, this invention discloses a method for generating the rupture process of a fault structure earthquake scenario. Summary of the Invention

[0004] The purpose of this invention is to provide a method for generating fault-structured earthquake scenarios that can generate a large number of scenario earthquakes with statistically consistent characteristics based on key seismological parameters (such as magnitude and fault type).

[0005] To achieve the above objectives, the present invention provides the following solution: A method for generating fault structure earthquake scenario rupture process includes: S1. Obtain earthquake-related parameters as input parameters; S2. Based on the input parameters, the final slip distribution field is generated using a spatial random field hybrid model that combines concave and convex bodies; S3. Based on the final slip distribution field and the location of the seismic source, obtain the rupture time of each sub-fault; S4. Calculate the rise time of each sub-fault based on the final slip distribution field; S5. Based on the rupture time and the rise time, use a Gaussian function to simulate the seismic moment release process, calculate the mean square relative error between the generated seismic scenario feature values ​​and the actual seismic data feature values, and return to S2 for continuous iterative optimization until the error is less than the preset value, and generate the source model file.

[0006] Optionally, the input parameters include magnitude, fault size, seismic moment, average slip, fault type, dip angle, slip direction, source location, latitude and longitude, sampling interval, shear modulus, and autocorrelation function.

[0007] Optionally, the final slip distribution field can be generated using a spatial random field hybrid model incorporating concave and convex bodies, including: Based on the average slip amount in the input parameters, a uniform slip distribution is formed in the spatial domain; Based on the uniform slip distribution of the slip amount, a two-dimensional Fourier transform is used to generate a deterministic slip spectrum in the wavenumber domain, and an autocorrelation function is used in the wavenumber domain to generate a random slip spectrum. The deterministic and random slip spectra are combined using a low-pass filter function to generate a hybrid slip spectrum; A two-dimensional inverse Fourier transform is applied to the hybrid slip spectrum to generate the background slip distribution in the spatial domain; The parameters of the concave-convex body are determined by empirical relationships. The concave-convex body is superimposed on the background slip distribution. The boundary of the fault slip distribution is smoothed and matched with the target seismic moment to obtain the final slip distribution field.

[0008] Optionally, the expression for the hybrid slip spectrum is: ; In the formula, The spectrum is a mixed slip distribution. For deterministic slip spectrum, For random slip spectrum, , represents the wavenumbers of the fault strike and dip, respectively, and F is a wavenumber combination function used to select the low wavenumber components of the deterministic part and the high wavenumber components of the random part.

[0009] Optionally, determining the location of the earthquake source includes: Method 1: Set the location of the seismic source at the center of the fault; Method 2: Randomly distribute the earthquake source locations within a preset range of the fault width along the dip direction; Method 3: Based on the empirical relationship between focal depth and magnitude, the focal locations are determined by randomly distributing them along the strike direction; Method 4: Based on the statistical patterns of the distribution of historical earthquake source locations, calculate the joint probability of each sub-fault becoming the rupture initiation point on the final slip distribution field, and randomly select the sub-fault as the source location according to the probability weight.

[0010] Optionally, based on the final slip distribution field and the epicenter location, the rupture time of each sub-fault can be obtained, including: The initial rupture time is determined by the location of the seismic source and the background rupture velocity; The rupture time of each sub-fault is calculated by applying a random perturbation to the initial rupture time using the sub-fault slip and seismic moment.

[0011] Optionally, the calculation of the rise time of each sub-fault includes: Method 1: Determine the constant rise time of the sub-fault based on the self-similarity constraint relationship with the seismic moment; Method 2: Assuming that the rise time is completely correlated with the square root of the sub-fault slip, and doubling the rise time of the fault within a preset range from the surface to represent the weakening of the near-surface fault properties, the average rise time of the fault plane is subject to self-similarity constraints with dip factor and seismic moment. Method 3: Assuming that the rise time is completely correlated with the square root of the sub-fault slip, and doubling the rise time of the fault within a preset range from the surface, setting a deep weakening zone, applying uncorrelated random perturbations to the rise time of the sub-fault, and using the seismic moment self-similarity relationship to constrain the average rise time on the fault plane in combination with the slip direction factor and dip angle factor.

[0012] The beneficial effects of this invention are as follows: This invention employs a spatial random field hybrid model combining concave and convex bodies to simulate the large-scale low-wavenumber characteristics and small-scale high-wavenumber characteristics of earthquake rupture, realistically reflecting the concentrated energy release characteristics on the fault plane and significantly improving the physical realism of the slip distribution field. During the simulation process, the mechanical characteristics of near-surface faults are fully considered, and a deep weakening zone is introduced, achieving refined simulation of the shallow geological structure of complex faults. Simultaneously, a Gaussian function is used to simulate the seismic moment release process, and the mean square relative error between the generated seismic scenario characteristic values ​​(duration, dominant frequency, peak velocity) and the characteristic values ​​of real seismic data is calculated and iteratively optimized to ensure that the generated seismic scenario is highly consistent with the real earthquake in terms of statistical characteristics. This invention compensates for the lack of real near-fault seismic records in actual engineering projects, enabling the generation of seismic scenarios for specific engineering sites and providing a guarantee for the prediction of seismic motion parameters in major engineering projects. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of a method for generating a fault structure earthquake scenario rupture process according to an embodiment of the present invention; Figure 2 This is a distribution diagram of the amount of slip, rupture time, and slip direction on the fracture surface during the rupture process, according to an embodiment of the present invention. Figure 3 This is a comparison chart of the duration of characteristic values ​​calculated from a large number of earthquake scenarios generated by the fault structure earthquake scenario rupture process generation method in this embodiment of the invention, under different fault types, and the duration of characteristic values ​​calculated from real earthquake data. Figure 4 This is a comparison chart of the dominant frequencies of eigenvalues ​​calculated from a large number of earthquake scenarios generated by the fault structure earthquake scenario rupture process generation method in this embodiment of the invention, and the dominant frequencies of eigenvalues ​​calculated from real earthquake data under different fault types. Figure 5 This is a comparison chart showing the peak rates of eigenvalue seismic moment release calculated from a large number of seismic scenarios generated by the fault structure seismic scenario rupture process generation method of this invention, under different fault types, with the peak rates of eigenvalue seismic moment release calculated from real seismic data. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] This embodiment proposes a method for generating fault structure earthquake scenario rupture processes, including: S1. Obtain earthquake-related parameters as input parameters; S2. Based on the input parameters, the final slip distribution field is generated using a spatial random field hybrid model that combines concave and convex bodies; S3. Based on the final slip distribution field and the location of the seismic source, obtain the rupture time of each sub-fault; S4. Calculate the rise time of each sub-fault based on the final slip distribution field; S5. Based on the rupture time and the rise time, use a Gaussian function to simulate the seismic moment release process, calculate the mean square relative error between the generated seismic scenario feature values ​​and the actual seismic data feature values, and return to S2 for continuous iterative optimization until the error is less than the preset value, and generate the source model file.

[0018] Furthermore, the input parameters include magnitude, fault size, seismic moment, average slip, fault type, dip angle, slip direction, source location, latitude and longitude, sampling interval, shear modulus, and autocorrelation function.

[0019] Specifically, a custom number of earthquake scenarios can be generated with a specified magnitude. Fault types can include normal faults, reverse faults, strike-slip faults, dip-slip faults, and all types of faults, with all types of faults being the default if no fault type is specified. Fault size and mean slip can be determined using various empirical methods relating to magnitude for different fault types. Seismic moment can be determined based on empirical formulas relating to magnitude. Dip angle and slip direction can be randomly determined within empirical ranges based on the specified fault type. For the slip direction, it is optional to add a standard deviation as a disturbance to the preset value. The shear modulus can be calculated based on the seismic moment, fault size, and mean slip determined using empirical relationships, or the default value of the shear modulus can be used. Meanwhile, in order to meet the discretization requirements of grid division in numerical simulation, the fault size is adjusted to an integer multiple of the sampling interval (sub-fault size). The shear modulus is used as a proportional coefficient connecting the physical relationship between seismic moment, fault area and average slip, to ensure that the generated seismic scenario conforms to physical laws. For autocorrelation functions, Gaussian, exponential, von Kármán and fractal functions can be used, and the von Kármán autocorrelation function is used by default if no autocorrelation function is specified.

[0020] Furthermore, the final slip distribution field is generated using a spatial random field hybrid model combining concave and convex bodies, including: Based on the average slip amount in the input parameters, a uniform slip distribution is formed in the spatial domain; Based on the uniform slip distribution of the slip amount, a two-dimensional Fourier transform is used to generate a deterministic slip spectrum in the wavenumber domain. This deterministic slip spectrum characterizes the large-scale low wavenumber characteristics of fault rupture. In the wavenumber domain, an autocorrelation function is used to generate a random slip spectrum, which characterizes the small-scale high wavenumber characteristics of fault rupture. The deterministic and random slip spectra are combined using a low-pass filter function to generate a hybrid slip spectrum; A two-dimensional inverse Fourier transform is applied to the hybrid slip spectrum to generate the background slip distribution in the spatial domain; The parameters of the concave-convex body are determined by empirical relationships. The concave-convex body is superimposed on the background slip distribution. The boundary of the fault slip distribution is smoothed and matched with the target seismic moment to obtain the final slip distribution field.

[0021] Specifically, firstly, a deterministic low-wavenumber slip spectrum characterizing the average slip of the fault plane is determined. Then, a random high-wavenumber slip spectrum is generated based on four autocorrelation functions: Gaussian, exponential, fractal, and von Kármán. In the wavenumber domain, a preset filtering function is used to combine the deterministic low-wavenumber slip spectrum with the random high-wavenumber slip spectrum to generate a hybrid slip spectrum. A two-dimensional inverse Fourier transform is performed on the hybrid slip spectrum to obtain the slip distribution in the spatial domain. A statistically consistent concave-convex body is superimposed in the spatial domain to simulate the concentration of energy release in the high-slip region of the fault plane during a real earthquake. Next, the fault boundary is smoothed; a Hanning window can be used to prevent unrealistically large slip amounts at the fault boundary. Finally, the slip distribution is matched with the target seismic moment to obtain the final slip distribution field, ensuring that the slip distribution field of the generated earthquake scenario maintains physical consistency with the input magnitude parameters in terms of total energy.

[0022] Furthermore, determining the location of the earthquake source includes: Method 1: Set the epicenter at the fault center; Method 2: Randomly distribute earthquake source locations within a preset range of fault width along the dip direction; Method 3: Based on the empirical relationship between focal depth and magnitude, randomly distribute the earthquakes along the strike direction to determine the focal location; Method 4: Based on the statistical patterns of the distribution of historical earthquake source locations, calculate the joint probability of each sub-fault becoming the rupture initiation point on the final slip distribution field, and randomly select the sub-fault as the source location according to the probability weight.

[0023] Furthermore, based on the final slip distribution field and the epicenter location, the rupture time of each sub-fault is obtained, including: The initial rupture time is determined by the location of the seismic source and the background rupture velocity; The rupture time of each sub-fault is calculated by applying a random perturbation to the initial rupture time using the sub-fault slip and seismic moment.

[0024] Specifically, the background rupture velocity is generated by setting it as the average rupture velocity, which is randomly selected between 0.7 and 0.8 times the shear wave velocity at the epicenter. To simulate the mechanical properties of shallow, weak zones, the background rupture velocity is further attenuated by 0.7 times in areas within 5 kilometers of the fault depth. The crust1.0 model is used to obtain the average characteristics of shear wave velocities for different geological structures based on the input latitude and longitude coordinates. The shear wave velocities of each sub-fault are obtained using a depth-dependent linear interpolation method.

[0025] Furthermore, the calculation of the rise time of each sub-fault includes: Method 1: Determine the constant sub-fault rise time based on the self-similarity constraint relationship with the seismic moment; Method 2: Assuming that the rise time is completely correlated with the square root of the sub-fault slip, and doubling the rise time of the fault within a preset range from the surface to represent the weakening of the near-surface fault properties, the average rise time of the fault plane is subject to self-similarity constraints with dip factor and seismic moment. Method 3: Assuming that the rise time is completely correlated with the square root of the sub-fault slip, and doubling the rise time of the fault within a preset range from the surface, setting a deep weakening zone, applying uncorrelated random perturbations to the rise time of the sub-fault, and using the seismic moment self-similarity relationship to constrain the average rise time on the fault plane in combination with the slip direction factor and dip angle factor.

[0026] The method of this embodiment will be further described below with reference to the accompanying drawings: like Figure 1 As shown, a method for generating a fault structure earthquake scenario rupture process includes the following steps: S1. Determine the magnitude, earthquake moment, fault size, average slip, fault type, dip angle, slip direction, focal location, latitude and longitude, sampling interval, shear modulus, and autocorrelation function; S2. Regarding the slip distribution, a spatial random field hybrid model combining concave and convex bodies is used to generate the final slip distribution field. S3. Regarding rupture time, the rupture time of each sub-fault is calculated based on the correlation between the slip amount of the sub-fault and the rupture time, and perturbation is added to the rupture time parameter. S4. Regarding the rise time, the rise time of each sub-fault is calculated based on the correlation between the slip of the sub-fault and the rise time, and the physical rationality is ensured by using the self-similar constraint relationship. S5. The seismic moment release process is simulated using a Gaussian function based on the rupture time, rise time, and slip distribution. S6. Calculate the mean square relative error between the generated large number of earthquake scenario feature values ​​and the feature values ​​of real earthquake data, and continuously iterate and optimize until the error is less than 10%; S7. Integrate the slip distribution, rupture time distribution, and rise time distribution into a source model file according to a specific format.

[0027] The magnitude, fault size, seismic moment, average slip, fault type, dip angle, slip direction, focal location, latitude and longitude, shear modulus, fault type, and autocorrelation function are used as input parameters.

[0028] For the input parameters, fault size and mean slip can be determined using various empirical relationships under different fault types at a specified magnitude; the seismic moment can be determined based on empirical relationships with magnitude; all fault types are assumed by default; the dip angle and slip direction can be randomly determined within an empirical range based on the specified fault type; for the slip direction, it is optional to add a standard deviation as a disturbance based on preset values; a slip constraint probability model can be constructed based on the statistical laws of the distribution of historical earthquake source locations, and the source location can be randomly selected or assumed on the fault plane based on probability weights; the shear modulus can be calculated based on the seismic moment, fault size, and mean slip determined by empirical relationships, or the default value can be used. The autocorrelation function is assumed to be the von Kármán autocorrelation function.

[0029] Generate a hybrid slip distribution combining the concave and convex bodies based on the determined input parameters, including: First, a uniform slip distribution in the spatial domain is constructed based on the average slip of the input parameters. This distribution is then transformed into a deterministic slip spectrum D(ks,kd) in the wavenumber domain through a two-dimensional Fourier transform. This deterministic slip spectrum represents the large-scale low wavenumber characteristics of the fault.

[0030] Then, a random high wavenumber slip spectrum is constructed in the wavenumber domain. A two-dimensional wavenumber matrix is ​​constructed based on the fault size and sampling interval. The power spectral density is calculated using the autocorrelation function and the wavenumber matrix. After normalizing the power spectral density, the square root is taken to obtain the amplitude spectrum. Finally, a spectrum with the same dimensions as the wavenumber matrix and numerical values ​​within the range is generated. A random phase spectrum uniformly distributed within the interval is used to combine the amplitude spectrum and the random phase spectrum using Euler's formula to generate a complex form of random field spectrum. This spectrum is then subjected to conjugate symmetry processing to ensure that the spatial random field generated by the inverse Fourier transform is a pure real field. This yields a slip spectrum S(ks,kd) representing the randomness of the small-scale high wavenumber characteristics on the fault. Finally, a low-pass filter function is used to combine the deterministic slip spectrum and the random slip spectrum to generate a mixed slip spectrum. A two-dimensional inverse Fourier transform is then performed to obtain the background slip distribution in the spatial domain. The expression for the mixed slip spectrum is: ; In the formula The spectrum is a mixed slip distribution. For deterministic slip spectrum, For random slip spectrum, , Let F represent the wavenumbers of the fault strike and dip, respectively. F is the wavenumber combination function used to select the low wavenumber components of the deterministic part and the high wavenumber components of the random part. Its expression is: ; In the formula = L / 2, = W / 2, N=1, where L and W are the fault length and width, respectively.

[0031] Determining the concave-convex body parameters through empirical relationships and superimposing the concave-convex bodies onto the background slip distribution specifically includes: determining the number of concave-convex bodies on the fault plane based on fault size and type using empirical statistical relationships. If there is only one concave-convex body, its area is set to 22% of the fault plane area. If there are two concave-convex bodies, the area of ​​one is set to 16% of the fault plane area, and the area of ​​the other is set to 6% of the fault plane area. If there are more than two concave-convex bodies, the maximum concave-convex body area is set while ensuring that the total area of ​​all concave-convex bodies is 22% of the fault plane area. Assuming 16% of the fault plane area, the remaining approximately 6% is randomly allocated to the other concave and convex bodies. When determining the location of the concave and convex bodies, their geometric centers are randomly set within the range of 5% to 95% of the fault width along the dip direction. Based on the physical constraint that the average slip of the concave and convex bodies is approximately twice the average slip of the fault plane, the required slip increment at the location of the concave and convex bodies is calculated and superimposed on the concave and convex body region. The ratio of the total seismic moment after superimposing the concave and convex bodies to the target seismic moment is calculated. The slip distribution of the entire fault plane is scaled to make the slip distribution field of the superimposed concave and convex bodies match the target seismic moment.

[0032] Finally, the Hanning window function is used to smooth the boundary of the fault slip distribution and then matched with the target seismic moment to obtain the final slip distribution field.

[0033] The location of the earthquake source can be determined using the following four different methods: Method 1: Set at the center of the fault; Method 2: Randomly distributed along the strike direction, and distributed along the dip direction within 20% to 80% of the fault width; Method 3: Determine the location of the earthquake source along the fault dip based on the empirical relationship between focal depth and magnitude, and distribute them randomly along the strike direction; Method 4: Based on the statistical patterns of the distribution of historical earthquake source locations, calculate the joint probability of each sub-fault becoming the rupture initiation point on the generated slip distribution field, and randomly select the sub-fault as the source location according to the probability weight.

[0034] The rupture time is determined based on the generated final slip distribution field, and is generated in the following manner: Based on the final slip distribution field and the location of the seismic source, the rupture time of each sub-fault is obtained, including: The initial rupture time is determined by the location of the seismic source and the background rupture velocity; The rupture time of each sub-fault is calculated by applying a random perturbation to the initial rupture time using the sub-fault slip and seismic moment.

[0035] The rise time is determined based on the generated final slip distribution field, and the rise time distribution is generated in the following three different ways: Method 1: Determine the constant sub-fault rise time based on the self-similarity constraint relationship with the seismic moment; Method 2: Assume that the rise time is completely correlated with the square root of the sub-fault slip, and double the rise time of the fault within 5 kilometers of the surface to represent the weakening of the near-surface fault properties. Finally, self-similar constraints are imposed on the average rise time of the fault plane with dip factor and seismic moment to satisfy reasonable physical laws. Method 3: Assuming the rise time is perfectly correlated with the square root of the sub-fault slip, and doubling the rise time of the fault within 5 km of the surface, a deep weakening zone is established in deeper areas. Specifically, starting at the greater of a fault depth of 15 km or the focal depth, within a 3 km range extending downwards, the proportionality coefficient between the sub-fault rise time and the square root of the slip is linearly increased to 2. Secondly, uncorrelated random perturbations are applied to the sub-fault rise time. Finally, combining the slip direction factor and dip factor, the average rise time on the fault plane is constrained using the seismic moment self-similarity relationship to ensure the model conforms to physical laws.

[0036] Finally, the seismic moment release process is simulated using a Gaussian function. The seismic moment release rate is calculated based on the rupture time and rise time of each sub-event, and the contributions of all sub-events are superimposed to form the final source time function on the fault plane.

[0037] Figure 2 The diagram shows the distribution of slip amount, rupture time, and slip direction on the fault surface generated during the rupture process in this embodiment.

[0038] Using the aforementioned fault-structure earthquake scenario rupture process generation method, earthquake scenarios with magnitudes between 5.5 and 8.0 are generated at 0.01 intervals. These earthquake scenarios are then distinguished according to fault type (strike-slip, dip-slip, and all types) and grouped into intervals of 0.2. Based on this, source characteristic values ​​for different fault types are calculated, including duration, dominant frequency, and peak moment release rate. Duration is the time taken for the cumulative moment to reach 0.1% to 99.9% of the total moment; dominant frequency is the frequency corresponding to the maximum amplitude of the source time function amplitude spectrum; and peak moment release rate is the peak value of the source time function. The source characteristic values ​​of the earthquake scenario data under different fault types are compared with those of the actual earthquake data. Through iterative optimization, the mean square relative error of each source characteristic value for both the earthquake scenario data and the actual earthquake data under different fault types is kept within a reasonable range (10%). The expression for the mean square relative error is as follows: ; In the formula, N represents the total number of magnitude intervals. Let be the mean of the characteristic values ​​of all synthetic earthquake scenarios within the i-th magnitude interval. Let be the mean of the characteristic values ​​of all real earthquake data within the i-th magnitude interval.

[0039] Figure 3 The figure shows a comparison of the duration of source characteristic values ​​calculated by the two types of data under different fault types. It can be seen from the figure that the average error of the duration of strike-slip faults is about 5.0%, the average error of the duration of dip-slip faults is about 2.9%, and the average error of the duration of all types is about 4.5%.

[0040] Figure 4 The comparison of the dominant frequencies of the source characteristics calculated from the two types of data under different fault types is shown. It can be seen that the mean error of the dominant frequency of strike-slip faults is about 8.6%, the mean error of the dominant frequency of dip-slip faults is about 4.4%, and the mean error of the dominant frequency of all types is about 7.2%.

[0041] Figure 5 The comparison of the peak moment release rates of the source characteristic values ​​calculated by the two types of data under different fault types is shown. It can be seen that the average error of the peak moment release rate of strike-slip faults is about 3.8%, the average error of the peak moment release rate of dip-slip faults is about 1.7%, and the average error of the peak moment release rate of all types of earthquakes is about 3.0%.

[0042] Because there is limited real earthquake data with magnitudes greater than 7, the errors in the two types of data are relatively large, but remain within an acceptable range. Therefore, this invention also calculates the mean square relative error of each characteristic value between earthquake scenario data and real data with magnitudes between 5.5 and 7 under different fault types. Figure 3 The average error of the duration of the two types of data for earthquakes with magnitudes between 5.5 and 7 is about 0.7% for strike-slip faults, about 1.6% for dip-slip faults, and about 0.8% for all types of earthquakes. Figure 4 The magnitude data shown are between 5.5 and 7. The mean error of the dominant frequency for strike-slip faults is approximately 3.2%, for dip-slip faults approximately 3.8%, and for all types of faults approximately 2.8%. Figure 5 The average error of the peak velocity for the two types of data, which are between magnitude 5.5 and 7, is about 4.5% for strike-slip faults, about 0.4% for dip-slip faults, and about 2.3% for all types of peak velocities.

[0043] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for generating a fault structure earthquake scenario rupture process, characterized in that, include: S1. Obtain earthquake-related parameters as input parameters; S2. Based on the input parameters, generate the final slip distribution field using a spatial random field hybrid model combining concave and convex bodies, including: Based on the average slip in the input parameters, a uniform slip distribution is formed in the spatial domain; Based on the uniform slip distribution of the slip amount, a two-dimensional Fourier transform is used to generate a deterministic slip spectrum in the wavenumber domain, and an autocorrelation function is used in the wavenumber domain to generate a random slip spectrum. The deterministic and random slip spectra are combined using a low-pass filter function to generate a hybrid slip spectrum; A two-dimensional inverse Fourier transform is applied to the hybrid slip spectrum to generate the background slip distribution in the spatial domain; The concave-convex body parameters are determined by empirical relationships, and the concave-convex body is superimposed on the background slip distribution. The boundary of the fault slip distribution is smoothed and matched with the target seismic moment to obtain the final slip distribution field. S3. Based on the final slip distribution field and the location of the seismic source, obtain the rupture time of each sub-fault; S4. Calculate the rise time of each sub-fault based on the final slip distribution field; S5. Based on the rupture time and the rise time, use a Gaussian function to simulate the seismic moment release process, calculate the mean square relative error between the generated seismic scenario feature values ​​and the actual seismic data feature values, and return to S2 for continuous iterative optimization until the error is less than the preset value, and generate the source model file.

2. The method for generating fault structure earthquake scenario rupture process according to claim 1, characterized in that, The input parameters include magnitude, fault size, seismic moment, average slip, fault type, dip angle, slip direction, source location, latitude and longitude, sampling interval, shear modulus, and autocorrelation function.

3. The method for generating fault structure earthquake scenario rupture process according to claim 1, characterized in that, The expression for the hybrid slip spectrum is: ； In the formula, The spectrum is a mixed slip distribution. For deterministic slip spectrum, For random slip spectrum, , represents the wavenumbers of the fault strike and dip, respectively, and F is a wavenumber combination function used to select the low wavenumber components of the deterministic part and the high wavenumber components of the random part.

4. The method for generating fault structure earthquake scenario rupture process according to claim 1, characterized in that, The determination of the earthquake source location includes: Method 1: Set the location of the seismic source at the center of the fault; Method 2: Randomly distribute the earthquake source locations within a preset range of the fault width along the dip direction; Method 3: Based on the empirical relationship between focal depth and magnitude, the focal locations are determined by randomly distributing them along the strike direction; Method 4: Based on the statistical patterns of the distribution of historical earthquake source locations, calculate the joint probability of each sub-fault becoming the rupture initiation point on the final slip distribution field, and randomly select the sub-fault as the source location according to the probability weight.

5. The method for generating fault structure earthquake scenario rupture process according to claim 4, characterized in that, Based on the final slip distribution field and the location of the seismic source, the rupture time of each sub-fault is obtained, including: The initial rupture time is determined by the location of the seismic source and the background rupture velocity; The rupture time of each sub-fault is calculated by applying a random perturbation to the initial rupture time using the sub-fault slip and seismic moment.

6. The method for generating fault structure earthquake scenario rupture process according to claim 1, characterized in that, The calculation of the rise time of each sub-fault includes: Method 1: Determine the constant rise time of the sub-fault based on the self-similarity constraint relationship with the seismic moment; Method 2: Assuming that the rise time is completely correlated with the square root of the sub-fault slip, and doubling the rise time of the fault within a preset range from the surface to represent the weakening of the near-surface fault properties, the average rise time of the fault plane is subject to self-similarity constraints with dip factor and seismic moment. Method 3: Assuming that the rise time is completely correlated with the square root of the sub-fault slip, and doubling the rise time of the fault within a preset range from the surface, setting a deep weakening zone, applying uncorrelated random perturbations to the rise time of the sub-fault, and using the seismic moment self-similarity relationship to constrain the average rise time on the fault plane in combination with the slip direction factor and dip angle factor.

Images