A method for generating a seismic intensity attenuation model after an earthquake

Abstract

The application discloses a kind of post-earthquake influence field generation-oriented seismic intensity attenuation model updating method, it is related to seismic intensity attenuation model updating technical field, comprising: obtaining the station waveform data of target earthquake, station position data and source parameter, calculate peak type instrument intensity and spectral intensity;According to the difference of two, determine peak-spectral misalignment, combined with waveform energy distribution extraction pulse concentration;According to station position data and surrounding station peak-spectral misalignment extraction spatial isolated misalignment, construct structure boundary trigger pseudo uplift intensity by peak-spectral misalignment, pulse concentration and spatial isolated misalignment;According to the intensity correction peak type instrument intensity obtain reliable intensity, determine contribution weight, update seismic intensity attenuation model parameters;Using the model after updating, calculate target area grid point prediction intensity, generate post-earthquake influence field.The application can reduce the interference of non-destructive high peak value pulse caused by local structure boundary to model updating, improve the accuracy of post-earthquake influence field generation.

Description

Technical Field

[0001] This invention relates to the field of earthquake intensity attenuation model updating technology, and in particular to a method for updating earthquake intensity attenuation models oriented towards the generation of post-earthquake influence fields. Background Technology

[0002] The post-earthquake impact field is a crucial data product for quickly determining the affected area, intensity distribution, and key areas for emergency response after an earthquake. It typically relies on seismic intensity attenuation models, combined with magnitude, focal depth, epicentral distance, azimuth, historical isoseismal data, and station records. With the increased density of strong-motion stations and intensity meters, early post-earthquake waveform data can be more quickly fed into the data center and used to correct intensity attenuation models and generate the post-earthquake impact field. However, in densely populated urban areas, some stations may be installed near building walls, equipment foundations, underground structures, bridge piers, or hard pavement boundaries. When ground motion reaches these areas of abrupt changes in local stiffness, reflections, scattering, short-term superposition, or transient responses of the installed structures can easily occur, resulting in short-duration, high-peak components in the station records. This component inflates the instrumental intensity based on peak acceleration and peak velocity, but its contribution to the response within the relevant period range of common building damage is limited, leading to inconsistencies between instrumental intensity and macroscopic intensity, thus affecting the accuracy of the generated post-earthquake impact field.

[0003] Existing methods for generating post-earthquake impact fields typically use peak ground acceleration (PGA), peak velocity (PV), or instrumental intensity derived from these parameters as the basis for model updates, or perform overall optimization of the intensity attenuation model based on historical isoseismal samples. While these methods can improve the fitting ability of regional intensity attenuation relationships, they are insufficient in identifying short-duration high-peak responses caused by local station installation environments and shallow engineering structures, easily misinterpreting local non-destructive peak increases as regional seismic intensification. Especially when early post-earthquake macroscopic survey data is insufficient, if such station data directly participates in updating the long-axis, short-axis, and distance attenuation parameters, the seismic intensity attenuation model is prone to absorbing local anomalous responses, resulting in artificially high local heights in the post-earthquake impact field, abnormally outward expansion of the impact range, or axial distribution deviations, reducing the reliability of post-earthquake emergency assessment results. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies, such as the difficulty in identifying short, sustained high-peak pulses caused by local structural boundaries of stations, which leads to the misuse of peak-type instrument intensity for updating seismic intensity attenuation models, resulting in local artificial heights, abnormal outward expansion, or axial distribution deviations in the post-earthquake influence field. Therefore, this invention proposes a seismic intensity attenuation model update method oriented towards the generation of the post-earthquake influence field.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution:

[0006] A method for updating a seismic intensity attenuation model oriented towards the generation of post-earthquake influence fields includes:

[0007] S1. Obtain the station waveform data, station location data and source parameters of the target earthquake, calculate the horizontal composite acceleration time history based on the station waveform data, and extract the peak acceleration and peak velocity to convert them into peak instrument intensity.

[0008] S2. Extract the horizontal relative velocity response spectrum value based on the horizontal acceleration time history in the station waveform data, integrate the horizontal relative velocity response spectrum value within the determined integration interval to obtain the spectral intensity, and convert the spectral intensity into spectral intensity.

[0009] S3. Calculate the peak spectral misalignment based on the difference between the peak instrument intensity and the spectral intensity, and extract the pulse concentration based on the energy distribution of the horizontal synthetic acceleration time history.

[0010] S4. Calculate the spatial reference weight based on the station location data, extract the spatial isolated misalignment based on the peak spectrum misalignment of surrounding stations, and construct the structural boundary trigger pseudo-lifting intensity based on the peak spectrum misalignment, pulse concentration and spatial isolated misalignment.

[0011] S5. Determine the credible intensity based on the structural boundary triggered pseudo-uplift intensity, peak instrument intensity, and spectral intensity; determine the contribution weight based on the structural boundary triggered pseudo-uplift intensity; and update the parameters of the seismic intensity attenuation model by combining the credible intensity and the contribution weight.

[0012] S6. Using the updated seismic intensity attenuation model, calculate the predicted intensity of each grid point in the target area to generate the updated post-earthquake impact field.

[0013] Preferably, the process involves acquiring station waveform data, station location data, and source parameters of the target earthquake; calculating the horizontal composite acceleration time history based on the station waveform data; and extracting peak ground acceleration and peak velocity to convert them into peak-type instrumental intensity, including:

[0014] Obtain the east-west and north-south acceleration time histories of the target station;

[0015] The horizontal composite acceleration time history of the target station is constructed by summing the squares of the east-west acceleration time history and the squares of the north-south acceleration time history.

[0016] The maximum absolute value of the horizontal composite acceleration time history is extracted as the peak acceleration; the east-west acceleration time history and the north-south acceleration time history are integrated over time to obtain the east-west velocity time history and the north-south velocity time history.

[0017] The horizontal composite velocity time history is obtained by summing the squares of the east-west velocity time history and the squares of the north-south velocity time history and taking the square root.

[0018] The maximum value of the horizontal composite velocity time history is extracted as the peak velocity;

[0019] Based on the conversion relationship obtained from historical earthquake data or instrument intensity specifications, peak acceleration and peak velocity are converted into peak-type instrument intensity.

[0020] Preferably, the horizontal relative velocity response spectrum value is extracted based on the horizontal acceleration time history in the station waveform data, the horizontal relative velocity response spectrum value is integrated within a determined integration interval to obtain the spectral intensity, and the spectral intensity is converted into spectral intensity, including:

[0021] Based on the east-west acceleration time history, north-south acceleration time history, damping ratio determined by historical earthquake examples or according to the earthquake response spectrum calculation rules, and structural vibration period from the station waveform data, the east-west single-degree-of-freedom system response and the north-south single-degree-of-freedom system response are calculated respectively, and the east-west relative velocity response and the north-south relative velocity response are obtained.

[0022] The maximum values ​​of the absolute values ​​of the east-west relative velocity response and the north-south relative velocity response were extracted at the sampling time to obtain the east-west relative velocity response spectrum and the north-south relative velocity response spectrum.

[0023] The horizontal relative velocity response spectrum value is obtained by summing the squares of the east-west relative velocity response spectrum value and the squares of the north-south relative velocity response spectrum value and taking the square root.

[0024] Obtain the macroscopic intensity from the historical earthquake sample set, and determine the lower limit and upper limit of the candidate period that maximize the correlation between the spectral intensity and the macroscopic intensity within the candidate period range. These are used as the lower limit and upper limit of the spectral intensity integral period, respectively.

[0025] Within the range defined by the lower limit and the upper limit of the spectral intensity integration period, the horizontal relative velocity response spectrum value is integrated, and the integration result is divided by the difference between the upper limit and the lower limit of the spectral intensity integration period to obtain the spectral intensity.

[0026] Based on the fitting function between spectral intensity and macroscopic intensity in historical earthquake cases, spectral intensity is converted into spectral intensity.

[0027] Preferably, the peak spectral misalignment is calculated based on the difference between the peak instrument intensity and the spectral intensity, and the pulse concentration is extracted based on the energy distribution of the horizontal composite acceleration time history, including:

[0028] Subtracting the spectral intensity from the peak-type instrument intensity yields the peak spectral misalignment.

[0029] The full-time energy is obtained by integrating the square of the horizontal composite acceleration time history over the entire sampling time domain.

[0030] Calculate the time integral of the product of the sampling time and the square of the horizontal composite acceleration time history, and divide the time integral by the total time history energy to obtain the energy center time;

[0031] The time integral of the product of the square of the difference between the sampling time and the energy center time and the square of the time history of the horizontal composite acceleration is calculated. The square root of this time integral is then taken after dividing it by the total time history energy to obtain the energy spread time.

[0032] Dividing the peak acceleration by the square root of the ratio of the full-time energy to the energy spread time yields the pulse concentration.

[0033] Preferably, spatial reference weights are calculated based on station location data, and spatial isolated misalignment is extracted by combining peak spectral misalignment with that of surrounding stations, including:

[0034] Identify valid stations that are not in the target station but are spatially different from them as surrounding stations;

[0035] The spatial distance between the target station and each of the surrounding stations is calculated by taking the square root of the sum of the squares of the differences in the horizontal coordinates and the squares of the differences in the vertical coordinates between the target station and each of the surrounding stations.

[0036] The spatial reference weight is obtained by dividing the reciprocal of the spatial distance between the target station and a single surrounding station by the sum of the reciprocals of the spatial distances between the target station and all surrounding stations.

[0037] The peak spectrum misalignment of each surrounding station is weighted and summed using spatial reference weights to obtain the weighted peak spectrum misalignment reference value.

[0038] Subtracting the weighted peak misalignment reference value from the peak misalignment value of the target station yields the spatial isolated misalignment value.

[0039] Preferably, the structural boundary-triggered pseudo-lifting intensity is constructed based on peak spectral misalignment, pulse concentration, and spatially isolated misalignment, including:

[0040] The median and interquartile range of peak spectral misalignment, pulse concentration, and spatial isolated misalignment of all valid stations in this earthquake were obtained respectively.

[0041] When the interquartile range of any indicator is zero, the mean of the non-zero absolute deviation of that indicator is used to replace the interquartile range. The mean of the non-zero absolute deviation is the average of the non-zero absolute differences between the corresponding indicators of all valid stations in this earthquake and the median of that indicator.

[0042] When there is no non-zero absolute difference for the corresponding indicator, the robust standardization result corresponding to the indicator is set to zero.

[0043] After subtracting the corresponding median from each indicator of the target station, divide by the corresponding interquartile range or the mean of non-zero absolute deviation to obtain robust standardized results of peak spectral misalignment, pulse concentration, and spatial isolated misalignment, respectively.

[0044] Extract the portion of each robust standardized result that is greater than zero, and use it as the positive part of each indicator.

[0045] The structural boundary-triggered pseudo-uplift intensity is obtained by multiplying the positive parts of each indicator and dividing by the sum of the numerical value and the product.

[0046] Preferably, the credible intensity is determined based on the structural boundary-triggered pseudo-uplift intensity, peak intensity, and spectral intensity; the contribution weight is determined based on the structural boundary-triggered pseudo-uplift intensity; and the parameters of the seismic intensity attenuation model are updated by combining the credible intensity and the contribution weight, including:

[0047] Subtract the spectral intensity from the peak-type instrument intensity to obtain the peak-spectral intensity difference;

[0048] Extract the portion of the peak intensity difference that is greater than zero as the positive peak difference value;

[0049] Multiply the positive difference in the peak spectrum by the intensity of the pseudo-uplift triggered by the structural boundary to obtain the intensity deduction;

[0050] Subtract the intensity deduction from the peak intensity to obtain the reliable intensity;

[0051] The contribution weight is obtained by subtracting the pseudo-uplift intensity triggered by the structural boundary from the numerical value;

[0052] Obtain the epicentral distance or fault distance of the station, the azimuth of the station relative to the source, the magnitude of the earthquake, the source depth of the earthquake, as well as the major axis direction angle, major axis direction scale parameter and minor axis direction scale parameter in the earthquake intensity attenuation model.

[0053] The azimuth difference angle is obtained by subtracting the major axis direction angle from the azimuth angle. The major axis component is calculated based on the quotient of the cosine of the azimuth difference angle multiplied by the epicentral distance or fault distance and the major axis direction scale parameter. The minor axis component is calculated based on the quotient of the sine of the azimuth difference angle multiplied by the epicentral distance or fault distance and the minor axis direction scale parameter.

[0054] Calculate the square root of the sum of the squares of the major axis components and the squares of the minor axis components to obtain the directional equivalent distance;

[0055] An earthquake intensity attenuation model is constructed with directional equivalent distance, magnitude of the earthquake, and focal depth of the earthquake as independent variables. The parameters to be updated in the earthquake intensity attenuation model include constant terms, magnitude influence coefficient, focal depth influence coefficient, distance attenuation coefficient, major axis direction angle, major axis direction scale parameter, and minor axis direction scale parameter, among which the major axis direction scale parameter and the minor axis direction scale parameter are positive.

[0056] Calculate the square of the difference between the predicted intensity and the credible intensity output by the earthquake intensity attenuation model, and multiply it by the contribution weight to form a weighted error;

[0057] The updated seismic intensity attenuation model parameters are solved by minimizing the sum of weighted errors from all stations.

[0058] Preferably, the predicted intensity of each grid point within the target area is calculated using the parameter-updated seismic intensity attenuation model to generate an updated post-earthquake influence field, including:

[0059] Obtain the epicentral distance or fault distance, azimuth relative to the hypocenter, magnitude of the earthquake, and focal depth of each grid point within the target area;

[0060] Substitute the above parameters of each grid point into the updated seismic intensity attenuation model to calculate the predicted intensity of each grid point.

[0061] The spatial distribution set of the locations of each grid point in the target area and the corresponding predicted intensity is used as the updated post-earthquake impact field.

[0062] Compared with the prior art, the beneficial effects of the present invention are:

[0063] 1. This invention calculates peak-type instrumental intensity and spectral intensity separately, and further constructs peak-spectral misalignment, pulse concentration, and spatial isolation misalignment. It can identify peak-spectral misalignment-type false uplift of intensity triggered by structural boundaries from three aspects: intensity deviation, waveform energy concentration, and consistency with surrounding stations. For short-duration high-peak pulses caused by building walls, equipment foundations, underground engineering structures, or hard boundaries, this invention does not directly delete the corresponding station record, but quantifies the representativeness of the station record to regional ground motion by the false uplift intensity triggered by structural boundaries. This allows non-destructive peak rise caused by local installation environment or shallow engineering structures of the station to be effectively identified, thereby reducing the interference of falsely high peak-type instrumental intensity on the generation of post-earthquake impact fields.

[0064] 2. This invention adaptively subtracts peak-type instrument intensity based on the pseudo-uplift intensity triggered by structural boundaries to obtain a reliable intensity for model updates, and simultaneously generates contribution weights. The reliable intensity and contribution weights are jointly introduced into the parameter update process of the seismic intensity attenuation model. Stations suspected of being affected by local structural boundaries can still retain effective seismic information, but their abnormal peak values ​​will not excessively pull the attenuation parameters in the long axis direction, short axis direction, and distance direction. This allows the updated seismic intensity attenuation model to generate the post-earthquake impact field more stably, reducing local artificial height, abnormal expansion of the impact range, and axial distribution deviation, and improving the consistency between the rapid assessment results of post-earthquake disasters and the actual distribution of earthquake impacts. Attached Figure Description

[0065] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0066] Figure 1 This is a flowchart illustrating a method for updating a seismic intensity attenuation model for generating post-earthquake influence fields, as provided in an embodiment of the present invention. Detailed Implementation

[0067] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0068] Example: This example provides a method for updating a seismic intensity attenuation model for the generation of post-earthquake influence fields. See [link to example]. Figure 1 Specifically, including:

[0069] S1. Obtain the station waveform data, station location data and source parameters of the target earthquake, calculate the horizontal composite acceleration time history based on the station waveform data, and extract the peak acceleration and peak velocity to convert them into peak instrument intensity.

[0070] S2. Extract the horizontal relative velocity response spectrum value based on the horizontal acceleration time history in the station waveform data, integrate the horizontal relative velocity response spectrum value within the determined integration interval to obtain the spectral intensity, and convert the spectral intensity into spectral intensity.

[0071] S3. Calculate the peak spectral misalignment based on the difference between the peak instrument intensity and the spectral intensity, and extract the pulse concentration based on the energy distribution of the horizontal synthetic acceleration time history.

[0072] S4. Calculate the spatial reference weight based on the station location data, extract the spatial isolated misalignment based on the peak spectrum misalignment of surrounding stations, and construct the structural boundary trigger pseudo-lifting intensity based on the peak spectrum misalignment, pulse concentration and spatial isolated misalignment.

[0073] S5. Determine the credible intensity based on the structural boundary triggered pseudo-uplift intensity, peak instrument intensity, and spectral intensity; determine the contribution weight based on the structural boundary triggered pseudo-uplift intensity; and update the parameters of the seismic intensity attenuation model by combining the credible intensity and the contribution weight.

[0074] S6. Using the updated seismic intensity attenuation model, calculate the predicted intensity of each grid point in the target area to generate the updated post-earthquake impact field.

[0075] In one embodiment of the present invention, the process includes acquiring station waveform data, station location data, and source parameters of the target earthquake; calculating the horizontal composite acceleration time history based on the station waveform data; and extracting peak acceleration and peak velocity to convert them into peak-type instrumental intensity.

[0076] This embodiment provides a method for updating a seismic intensity attenuation model for generating post-earthquake impact fields. This method is executed collaboratively by a seismic data center, a strong-motion station data access unit, a source parameter access unit, a station spatial data management unit, a waveform processing unit, a spectral intensity calculation unit, a pseudo-uplift intensity calculation unit, a model parameter update unit, and a post-earthquake impact field generation unit. This embodiment focuses on peak-spectral misalignment-type pseudo-uplift of intensity triggered by structural boundaries, and specifically defines the station waveform processing, spectral intensity calculation, peak-spectral misalignment amount, pulse concentration amount, spatially isolated misalignment amount, structural boundary-triggered pseudo-uplift intensity, credible intensity, contribution weight, model parameter update, and post-earthquake impact field generation process.

[0077] It should be noted that peak-spectral misalignment-type intensity pseudo-uplift triggered by structural boundaries refers to the phenomenon where, when a station is installed near building walls, equipment foundations, underground engineering structures, bridge piers, hard pavement boundaries, or other areas of sudden changes in local stiffness, seismic motion is reflected, scattered, superimposed for short periods, or the installation structure responds transiently at these local structural boundaries. This results in short-duration high-peak components appearing in the station waveform, leading to a higher peak-type instrument intensity based on peak acceleration or peak velocity than its actual contribution to the macroscopic damage-related periodic response. This phenomenon manifests in the data as a higher peak-type instrument intensity relative to the spectral intensity, an abnormally concentrated peak value relative to the full-time energy distribution, and spatial isolation of this higher intensity relative to surrounding stations.

[0078] After a target earthquake occurs, the seismic data center receives the earthquake's source parameters, station waveform data from the target area and its surrounding regions, station location data, and current earthquake intensity attenuation model parameters. Source parameters include the earthquake's magnitude, focal depth, epicenter location, and fault geometry information when fault parameters are available. Station location data includes station number, horizontal coordinates, vertical coordinates, elevation information, and station installation location description. Station waveform data includes the east-west and north-south acceleration time histories of the target station. Waveform sampling times are uniformly converted to the same time base by the data center. If different stations have inconsistent sampling frequencies, they are unified to the sampling frequency required for model processing through resampling. If a station has significant missing data, saturation cutoff, missing timestamps, or incomplete horizontal components, the station is marked as invalid and will not participate in subsequent structural boundary-triggered pseudo-uplift intensity calculations and model parameter updates; however, its original records are still stored in the data center for manual verification.

[0079] It should be noted that a valid station refers to a station whose waveform data, station location data, and time information meet the calculation requirements of this embodiment; specifically, it includes stations with complete east-west acceleration time histories and north-south acceleration time histories, whose sampling times can be unified to the same time base, whose waveforms do not affect the peak value calculation and the saturation cutoff of the response spectrum calculation, and whose station location can be used for spatial distance calculation; stations that do not meet the above conditions are considered invalid stations and will not participate in the calculation of the pseudo-uplift intensity triggered by the structural boundary and the updating of model parameters.

[0080] For valid stations, the waveform processing unit reads the east-west acceleration time history of the i-th station. North-South Acceleration Time History Where i represents the station number, t represents the sampling time, E represents the east-west direction, and N represents the north-south direction. This represents the east-west acceleration value of the i-th station at sampling time t. This represents the north-south acceleration value of the i-th station at sampling time t. Before calculation, the waveform processing unit can perform baseline correction and frequency band constraint processing on the two horizontal acceleration time histories. For anomalous cusps, only non-seismic cusps formed by communication errors, single sampling point jumps, sensor saturation recovery glitches, or non-seismic triggering noise are deleted; ground motion pulse segments that exist continuously in multiple adjacent sampling points and can be included in the calculation of peak acceleration, response spectrum, or energy distribution are not deleted. Frequency band constraint processing is used to remove sensor drift, long-period trend terms, and obvious non-seismic interference, but is not used to delete short, continuous high-peak pulses that need to be identified in this invention, so that subsequent peak acceleration, peak velocity, and response spectrum calculations are based on the same type of effective seismic records.

[0081] The waveform processing unit constructs the horizontal composite acceleration time history of the i-th station based on the east-west and north-south acceleration time histories. The calculation method is as follows:

[0082]

[0083] in, This represents the horizontal composite acceleration value of the i-th station at sampling time t. This represents the east-west acceleration value of the i-th station at sampling time t. This represents the north-south acceleration value of the i-th station at sampling time t. The reason for using this calculation method is that the seismic intensity attenuation model update needs to reflect the comprehensive intensity of the impact of horizontal ground motion on buildings and the ground surface. A single horizontal component may be affected by the azimuth layout of the stations. Taking the square root after summing the squares can transform the two orthogonal horizontal components into the horizontal composite response at the same time, thus providing a unified input for subsequent calculations of peak acceleration, pulse concentration, and energy distribution.

[0084] After obtaining the horizontal composite acceleration time history, the waveform processing unit extracts the maximum absolute value of the horizontal composite acceleration time history as the peak acceleration of the i-th station. The calculation method is as follows:

[0085]

[0086] in, This represents the peak acceleration of the i-th station. This represents the horizontal composite acceleration value of the i-th station. This means taking the maximum value within the entire sampling time domain. The reason for using this calculation method is that peak acceleration is an important input for the intensity conversion of peak-type instruments, which can characterize the strongest instantaneous horizontal acceleration in the station record. However, it is also easily affected by short, continuous sharp pulses. Therefore, it is necessary to combine spectral intensity and pulse concentration to identify the possible false rise caused by it.

[0087] The waveform processing unit performs time integration on the east-west acceleration time history and the north-south acceleration time history respectively to obtain the east-west velocity time history. and north-south speed time history The calculation method is as follows:

[0088]

[0089]

[0090] in, This represents the east-west velocity value of the i-th station at sampling time t. This represents the north-south velocity value of the i-th station at sampling time t. This represents the east-west acceleration value of the i-th station at sampling time t. This represents the north-south acceleration value of the i-th station at sampling time t. Time integration can be achieved using trapezoidal integration or other numerical integration methods. The baseline drift before and after integration is corrected by the data center according to the same waveform processing rules to avoid the velocity time history being contaminated by low-frequency drift.

[0091] The horizontal composite velocity time history of the i-th station is obtained by summing the squares of the east-west velocity time history and the squares of the north-south velocity time history, and then taking the square root. The calculation method is as follows:

[0092]

[0093] in, This represents the horizontal composite velocity value of the i-th station at sampling time t. This represents the east-west velocity value of the i-th station at sampling time t. Let represent the north-south velocity value of the i-th station at sampling time t. The maximum value of the horizontal composite velocity time history is further extracted as the peak velocity, calculated as follows:

[0094]

[0095] in, This represents the peak speed of the i-th station. This represents the horizontal composite velocity value of the i-th station. Peak velocity is used because it reflects another type of peak effect of ground motion on structural response and surface movement. When used together with peak acceleration as input for peak-type instrument intensity conversion, it can take into account the influence of both peak acceleration and peak velocity on instrument intensity.

[0096] In obtaining and Then, the waveform processing unit calls the conversion relationship obtained from historical earthquake data or instrument intensity specifications to convert the peak acceleration and peak velocity into the peak-type instrument intensity of the i-th station. The calculation method is as follows:

[0097]

[0098] in, This represents the peak intensity of the instrument at station i. This represents the conversion function from peak acceleration and peak velocity to peak-type instrument intensity. This represents the peak acceleration of the i-th station. This represents the peak velocity of the i-th station; It can be determined by the existing instrument intensity calculation rules, historical earthquake case calibration relationships, or business specification conversion relationships in the data platform, and can be replaced by an instrument intensity conversion function that satisfies the same input-output relationship in different regional business systems.

[0099] It should be noted that peak-type instrumental intensity refers to the intensity characterization value obtained from the peak acceleration and peak velocity recorded by the station through the instrumental intensity conversion relationship. This intensity characterization value can reflect the maximum instantaneous ground motion level in the station records, but it is more sensitive to short-duration high-peak pulses. Therefore, in this embodiment, it is not directly used as the sole intensity input for model updates, but rather participates in the determination of reliable intensity along with spectral intensity, peak spectral misalignment, and structural boundary triggered pseudo-uplift intensity. The horizontal synthetic acceleration time history obtained in this step... The peak acceleration is used for subsequent calculations of full-time energy, energy center time, energy spread time, and pulse concentration. Used for subsequent pulse concentration calculations to obtain peak-type instrument intensity The data is used for subsequent calculations of peak spectral misalignment, confidence intensity, and model weighted update. Station location data is used for subsequent calculations of spatial reference weights and spatial isolation misalignment.

[0100] In one embodiment of the present invention, the horizontal relative velocity response spectrum value is extracted based on the horizontal acceleration time history in the station waveform data, the horizontal relative velocity response spectrum value is integrated within a determined integration interval to obtain the spectral intensity, and the spectral intensity is converted into spectral intensity, including:

[0101] After obtaining the horizontal acceleration time history of the effective stations, the spectral intensity calculation unit extracts the horizontal relative velocity response spectrum value based on the horizontal acceleration time history in the station waveform data, integrates the horizontal relative velocity response spectrum value within the determined integration interval to obtain the spectral intensity, and converts the spectral intensity into spectral intensity.

[0102] It should be noted that spectral intensity refers to the intensity characterization value obtained by converting the relative velocity response spectrum integral of horizontal ground motion at a station within the macroscopic damage-related period range, through the mapping relationship between spectral intensity and macroscopic intensity in historical earthquake cases. This spectral intensity is used to characterize the contribution of ground motion to the damage-related period response of common buildings. Compared with peak-type instrumental intensity, which relies solely on peak acceleration or peak velocity, it has lower direct sensitivity to short-duration, non-destructive sharp pulses.

[0103] Specifically, the spectral intensity calculation unit calculates the east-west acceleration time history for the i-th station. North-South Acceleration Time History The responses of a single-degree-of-freedom system in the east-west and north-south directions were calculated according to the seismic response calculation rules for single-degree-of-freedom systems. For a single-degree-of-freedom system with period T, its east-west relative displacement response was determined. and north-south relative displacement response They respectively satisfy the following dynamic equations:

[0104]

[0105]

[0106] in, Indicates the period of the i-th station The east-west relative displacement response is as follows. Indicates the period of the i-th station The east-west relative velocity response is as follows. Indicates the period of the i-th station The east-west relative acceleration response, Indicates the period of the i-th station The north-south relative displacement response is as follows. Indicates the period of the i-th station North-south relative velocity response This represents the north-south relative acceleration response of the i-th station under period T. Indicates the damping ratio. Indicates the period of structural vibration. This represents the east-west acceleration value of the i-th station at sampling time t. This represents the north-south acceleration value of the i-th station at sampling time t. The damping ratio... The structural vibration period can be determined by historical earthquake examples, earthquake response spectrum calculation rules, or commonly used structural response calculation rules for the target area. Values ​​are taken point by point according to the candidate period sequence. The interval of the candidate period sequence can be determined based on the station sampling frequency, the accuracy requirements of the response spectrum calculation, and the computing power of the data station.

[0107] The candidate period range satisfies the condition that the upper limit of the candidate period is greater than the lower limit, and that both the lower and upper limits of the candidate period fall within a period range where the station's sampling frequency can stably calculate the response spectrum. For candidate period ranges where the response spectrum numerical calculation is unstable, the number of effective period points within the candidate period range is insufficient, or the number of historical samples is insufficient for correlation evaluation, the data platform will not consider them as candidates for the spectral intensity integration period range. Therefore, the spectral intensity integration period range is not a pre-fixed single parameter, but rather adaptively determined based on the consistency between spectral intensity and macroscopic intensity within a candidate range that satisfies the computability of the response spectrum and the evaluability of historical samples. The reason for using a single-degree-of-freedom system response to calculate the relative velocity response spectrum is that macroscopic damage is not solely determined by a single sharp peak, but is related to the structural response energy within a certain period range. The relative velocity response spectrum can provide spectral response information that is more consistent with the damage of common buildings.

[0108] The spectral intensity calculation unit extracts the maximum absolute values ​​of the east-west relative velocity response and the north-south relative velocity response within the entire sampling time domain, respectively, to obtain the east-west relative velocity response spectrum value and the north-south relative velocity response spectrum value. The calculation method is as follows:

[0109]

[0110]

[0111] in, This represents the east-west relative velocity response spectrum value of the i-th station under period T. This represents the north-south relative velocity response spectrum value of the i-th station under period T. This represents the east-west relative velocity response of the i-th station under period T. This represents the north-south relative velocity response of the i-th station under period T.

[0112] Further, the square root of the sum of the squares of the east-west relative velocity response spectrum and the squares of the north-south relative velocity response spectrum is taken to obtain the horizontal relative velocity response spectrum. The calculation method is as follows:

[0113]

[0114] in, This represents the horizontal relative velocity response spectrum value of the i-th station under period T. This represents the east-west relative velocity response spectrum value. This represents the north-south relative velocity response spectrum value. The reason for using this calculation method is that the subsequent spectral intensity needs to reflect the combined effect of horizontal ground motion on the structural response. Combining the two horizontal response spectra can reduce the bias of single-direction records in the spectral intensity calculation.

[0115] When determining the spectral intensity integration period range, the data platform calls upon a historical earthquake sample set, which includes historical station waveforms, station locations, historical source parameters, and macroscopic intensity data. and upper limit of candidate period The candidate spectral intensities are calculated within multiple candidate period ranges, and the correlation between the candidate spectral intensities and historical macroscopic intensity is calculated. The lower limit and upper limit of the candidate period with the highest correlation are determined as the lower limit of the spectral intensity integral period. Upper limit of the integral period of spectral intensity The calculation method is as follows:

[0116]

[0117] in, This indicates the lower limit of the integral period of the spectral intensity. This indicates the upper limit of the spectral intensity integral period. This indicates the lower limit of the candidate period. The upper limit of the candidate period is represented by Corr, which represents the correlation calculation function. Indicates the i-th historical station within the candidate period range The spectral intensity obtained from internal calculation, This represents the macroscopic intensity at the location corresponding to the i-th historical station. The correlation calculation function can be Pearson correlation, Spearman correlation, or a correlation function that has been operationally verified to measure the consistency between spectral intensity and macroscopic intensity. The candidate period range does not use a fixed and unique value, but is adaptively determined by historical earthquake cases and macroscopic intensity data to adapt to differences in building types, site conditions, and earthquake case samples in different regions.

[0118] In determining and Then, the spectral intensity calculation unit integrates the horizontal relative velocity response spectrum value within the range determined by the lower limit and upper limit of the spectral intensity integration period, and divides the integration result by the difference between the upper and lower limits of the spectral intensity integration period to obtain the spectral intensity of the i-th station. The calculation method is as follows:

[0119]

[0120] in, Let represent the spectral intensity of the i-th station. This represents the horizontal relative velocity response spectrum value of the i-th station under period T. This indicates the lower limit of the integral period of the spectral intensity. This represents the upper limit of the spectral intensity integral period. The reason for using interval average integral is that this calculation method can reflect the overall spectral response level within the period range related to macroscopic damage, rather than relying solely on the response peak at a certain period point, thereby reducing the direct amplification of the intensity characterization by short, sustained high-peak sharp pulses.

[0121] The spectral intensity calculation unit converts spectral intensity into spectral intensity based on the fitting function between spectral intensity and macroscopic intensity from historical earthquake examples. The calculation method is as follows:

[0122]

[0123] in, This represents the spectral intensity of the i-th station. This represents a function that converts spectral intensity into spectral strength. This represents the spectral intensity of the i-th station; The intensity is obtained by fitting the station spectral intensity and macroscopic intensity from historical earthquake cases. It can be obtained by linear fitting, piecewise fitting, or a regression function that satisfies a monotonic mapping relationship verified in operation. The input is spectral intensity, and the output is spectral intensity. It is also updated synchronously with the historical earthquake sample of the target area in the data platform.

[0124] Specifically, the data platform uses the spectral intensity of each historical seismic station in historical earthquake cases as input samples and the macroscopic intensity at the location of that historical station or the intensity of a macroscopic survey point spatially adjacent to that historical station as output samples to fit a mapping function between spectral intensity and spectral intensity. When using piecewise fitting or regression functions, a monotonic constraint that does not decrease with increasing spectral intensity is applied to the mapping function to ensure that larger spectral intensities are not converted into lower spectral intensities. When the historical macroscopic intensity samples for the target area are insufficient to refit the mapping function, the data platform uses the spectral intensity to intensity conversion function already calibrated in the current operational system and refits this conversion function after subsequent macroscopic intensity survey data are supplemented. The training input, training output, and update conditions of the mapping function are all recorded by the data platform, ensuring that the same batch of station data can obtain spectral intensity according to the same conversion relationship. The spectral intensity obtained in this step... Used for subsequent peak-type instrument intensity Calculate the peak misalignment amount and use it for subsequent confidence intensity correction.

[0125] In one embodiment of the present invention, the peak-spectral misalignment is calculated based on the difference between the peak-type instrument intensity and the spectral intensity, and the pulse concentration is extracted based on the energy distribution of the horizontal synthetic acceleration time history, including:

[0126] After obtaining the peak instrument intensity at the i-th station Spectral Intensity Subsequently, the pseudo-lift intensity calculation unit calculates the peak spectral misalignment and extracts the pulse concentration based on the energy distribution of the horizontal composite acceleration time history. Specifically, the pseudo-lift intensity calculation unit subtracts the spectral intensity from the peak-type instrument intensity to obtain the peak spectral misalignment, calculated as follows:

[0127]

[0128] in, This represents the peak spectral misalignment at the i-th station. This represents the peak intensity of the instrument at station i. This represents the spectral intensity of the i-th station.

[0129] It should be noted that peak-spectral misalignment refers to the difference between peak-type instrument intensity and spectral intensity at the same station. It characterizes the degree of deviation between the intensity level reflected by peak-type indicators and the intensity level reflected by the response spectrum integral. When the peak-spectral misalignment is positive and relatively large, it indicates that the peak-type instrument intensity at that station is higher than the spectral intensity, suggesting that the station's waveform may contain short-duration high-peak components that contribute significantly to peak acceleration or peak velocity but have limited contribution to the macroscopic damage-related periodic response. The reason for using this calculation method is that when the peak-type instrument intensity is significantly higher than the spectral intensity, it indicates that the station's record may contain short-duration high-peak components that contribute significantly to peak indicators but have limited contribution to the macroscopic damage-related periodic range. The peak-spectral misalignment can quantify the degree of deviation between peak intensity and spectral intensity.

[0130] To determine whether the deviation has a sharp pulse concentration, the pseudo-lift intensity calculation unit integrates the square of the horizontal composite acceleration time history over the entire sampling time domain to obtain the full-time energy. The calculation method is as follows:

[0131]

[0132] in, This represents the full-time energy of the horizontal composite acceleration time history at the i-th station. This represents the horizontal composite acceleration value of the i-th station at sampling time t. The reason for using full-time energy is that the enhancement of ground motion in the real area usually results in a continuous increase in the effective vibration energy over the entire area, while sharp pulses caused by reflections from local structural boundaries, scattering from shallow engineering structures, or transient responses of installed structures may only raise the maximum peak value for a very short time. Therefore, it is necessary to observe the relationship between the peak value and the full-time energy simultaneously.

[0133] The pseudo-lift intensity calculation unit calculates the time integral of the product of the sampling time and the square of the horizontal composite acceleration time history, and divides this time integral by the total time history energy to obtain the energy center time. The calculation method is as follows:

[0134]

[0135] in, Let t represent the center time of the acceleration energy distribution at the i-th station, and t represent the sampling time. This represents the horizontal composite acceleration value of the i-th station at sampling time t. This represents the full-time energy of the horizontal composite acceleration time history at the i-th station. This calculation method is used because the energy center time indicates the concentrated location of the seismic energy at that station on the time axis, providing a benchmark for subsequent calculations of the energy spread time.

[0136] The pseudo-lift intensity calculation unit calculates the time integral of the product of the square of the difference between the sampling time and the energy center time and the square of the horizontal composite acceleration time history. This time integral is divided by the total time history energy, and the square root is taken to obtain the energy spread time. The calculation method is as follows:

[0137]

[0138] in, The time interval of the acceleration energy at station i on the time axis represents the expansion time of the acceleration energy at station i, and t represents the sampling time. This represents the center moment of the acceleration energy distribution at the i-th station. This represents the horizontal composite acceleration value of the i-th station at sampling time t. This represents the full-time energy of the horizontal composite acceleration time history at the i-th station. The energy spread time is used because it can characterize the temporal distribution width of energy around the energy center. The energy spread time corresponding to short, sharp pulses is usually small, while the energy spread time of continuously increasing ground motions is usually large.

[0139] After obtaining the full-time energy and energy spread time, the pseudo-lift intensity calculation unit divides the peak acceleration by the square root of the ratio of the full-time energy to the energy spread time to obtain the pulse concentration. The calculation method is as follows:

[0140]

[0141] in, This represents the pulse concentration at the i-th station. This represents the peak acceleration of the i-th station. This represents the full-time energy of the horizontal composite acceleration time history at the i-th station. This represents the expansion time of the acceleration energy of the i-th station on the time axis.

[0142] It should be noted that pulse concentration refers to a quantity used to characterize whether the maximum peak value of the station waveform is abrupt relative to the full-time energy distribution. It is determined jointly by peak acceleration, full-time energy, and energy spread time. A larger pulse concentration indicates that the maximum peak value in the station record is more concentrated relative to the overall ground motion energy distribution, more consistent with the waveform characteristics of a short, sustained, high-amplitude pulse. A smaller pulse concentration indicates that the peak value is more consistent with the full-time energy distribution, more likely originating from overall ground motion enhancement. The reason for using this calculation method is that... This represents the equivalent sustained energy level formed by the full-time energy and the energy spread time. The significantly higher value relative to this equivalent sustained energy level indicates that the maximum peak is more likely to originate from a short, sustained high-amplitude pulse, rather than from a sustained increase in overall seismic intensity. zero or If the waveform cannot be effectively calculated, it indicates that the station's waveform does not meet the conditions for effective vibration recording. The data center marks this station as an invalid station and will not participate in subsequent model updates. The peak misalignment obtained in this step... and pulse concentration Used for subsequent structural boundary trigger pseudo-lift strength calculation.

[0143] In one embodiment of the present invention, spatial reference weights are calculated based on station location data, spatial isolated misalignment is extracted by combining peak spectral misalignment with that of surrounding stations, and structural boundary-triggered pseudo-uplift intensity is constructed based on peak spectral misalignment, pulse concentration, and spatial isolated misalignment, including:

[0144] To further determine whether peak spectrum misalignment has local structural boundary triggering properties, the pseudo-lift intensity calculation unit calculates spatial reference weights based on station location data, extracts spatially isolated misalignment quantities by combining peak spectrum misalignment quantities from surrounding stations, and constructs a structural boundary triggering pseudo-lift intensity based on peak spectrum misalignment quantities, pulse concentration quantities, and spatially isolated misalignment quantities. Specifically, for the i-th target station, the data center determines valid stations outside the target station and with different spatial locations as surrounding stations. If multiple equipment records exist at the same location, valid records can be selected first according to equipment calibration reliability, waveform integrity, or the data center's station priority rules to avoid zero spatial distance leading to abnormal weight calculation. For the i-th station and the j-th surrounding station, the pseudo-lift intensity calculation unit calculates the spatial distance based on the square root of the sum of the squares of the differences in the horizontal and vertical coordinates of the two stations. The calculation method is as follows:

[0145]

[0146] in, This represents the spatial distance between the i-th station and the j-th station. This represents the horizontal coordinate of the i-th station. This represents the vertical coordinate of the i-th station. This represents the horizontal coordinate of the j-th station. This represents the longitudinal coordinate of the j-th station. The reason for using this calculation method is that stations that are spatially closer can better reflect the common changes in ground motion in the surrounding area of ​​the target station. Therefore, when determining whether the peak spectral misalignment of the target station is isolated, it is necessary to give priority to referring to effective stations that are spatially closer.

[0147] The pseudo-lift intensity calculation unit divides the reciprocal of the spatial distance between the target station and a single surrounding station by the sum of the reciprocals of the spatial distances between the target station and all surrounding stations to obtain the spatial reference weight. The calculation method is as follows:

[0148]

[0149] in, This represents the spatial reference weight of the j-th surrounding station relative to the i-th target station. This represents the spatial distance between the i-th station and the j-th station. This represents the sum of the reciprocals of the spatial distances between the i-th target station and all surrounding stations. The reason for using this calculation method is that it allows surrounding stations closer to the target station to have a greater reference value, while normalization ensures that the sum of the weights of all surrounding stations is one.

[0150] When the number of available surrounding stations is insufficient, the data platform expands the search range of surrounding stations according to the spatial retrieval rules set by the business system. If there are still no available surrounding stations after expansion, the spatial isolation misalignment of the station is set to zero, and the station cannot form a high structural boundary to trigger pseudo-uplift intensity based solely on peak spectrum misalignment and pulse concentration, so as to avoid the impact of a single station record on model updates that has not been spatially referenced and verified.

[0151] The pseudo-uplift intensity calculation unit uses spatial reference weights to perform a weighted summation of the peak spectral misalignment of each surrounding station, obtaining the weighted peak spectral misalignment reference value. The calculation method is as follows:

[0152]

[0153] in, This represents the weighted peak spectrum misalignment reference value of the stations surrounding the i-th station. This represents the spatial reference weight of the j-th surrounding station relative to the i-th target station. This represents the peak spectral misalignment of the j-th surrounding station. The reason for using this calculation method is that if multiple stations around the target station simultaneously exhibit peak instrument intensity higher than spectral intensity, the misalignment is more likely to reflect regional ground motion differences rather than an isolated response triggered by the local installation environment or structural boundary of the target station.

[0154] The pseudo-lift intensity calculation unit subtracts the weighted peak spectrum misalignment reference value from the peak spectrum misalignment value of the target station to obtain the spatially isolated misalignment value. The calculation method is as follows:

[0155]

[0156] in, This represents the spatial isolation misalignment of the i-th station relative to its surrounding stations. This represents the peak spectral misalignment at the i-th target station. This represents the weighted peak spectrum misalignment reference value of stations surrounding the i-th target station.

[0157] It should be noted that spatially isolated misalignment refers to the deviation of the peak spectrum misalignment of the target station relative to the weighted peak spectrum misalignment reference value of surrounding effective stations. This value is used to distinguish between regional and local peak spectrum misalignment. When the target station and surrounding stations have similar peak spectrum misalignments, the misalignment is more likely to reflect regional seismic motion characteristics. When the peak spectrum misalignment of the target station is significantly higher than the reference level of surrounding stations, the misalignment is more likely to be related to the local structural boundary, installation environment, or shallow engineering structure response near the target station. The reason for using this calculation method is that when... When the value is positive and relatively large, it indicates that the peak spectral misalignment of the target station is significantly higher than the reference level of surrounding stations. This misalignment exhibits spatial isolation, which is more consistent with the pseudo-uplift of intensity caused by peak spectral misalignment triggered by building walls, equipment foundations, underground engineering structures, bridge piers, or other localized hard boundaries. The spatially isolated misalignment amount obtained in this step... Peak spectral misalignment and pulse concentration Together they are used to construct the pseudo-lift strength triggered by the subsequent structural boundary.

[0158] After obtaining the peak spectral misalignment values ​​of all valid stations in this earthquake... Pulse Concentration and spatially isolated misalignment Subsequently, the pseudo-lift intensity calculation unit obtained the median peak spectral misalignment, median pulse concentration, and median spatial isolated misalignment for all valid stations in this earthquake, and also obtained the interquartile range of peak spectral misalignment, interquartile range of pulse concentration, and interquartile range of spatial isolated misalignment for all valid stations in this earthquake, denoted as... , , , , and .in, This represents the median of the peak spectral misalignment at all valid stations during this earthquake. This represents the median of the pulse concentration at all valid stations during this earthquake. This represents the median of the spatially isolated misalignment of all valid stations during this earthquake. The interquartile range represents the peak spectral misalignment of all valid stations in this earthquake. The interquartile range represents the pulse concentration of all effective stations in this earthquake. The interquartile range represents the spatially isolated misalignment of all valid stations in this earthquake. The median and interquartile range are used because early station data for this earthquake may include local anomalous stations, and using robust statistics can reduce the impact of a few anomalous high values ​​on the batch reference scale.

[0159] When the interquartile range of any indicator is zero, the pseudo-lift intensity calculation unit replaces the interquartile range with the mean of the non-zero absolute deviation of that indicator. The mean of the non-zero absolute deviation is the average of the non-zero absolute differences between the corresponding indicators of all valid stations in this earthquake and the median of that indicator. When there is no non-zero absolute difference for a corresponding indicator, the robust standardization result corresponding to that indicator is set to zero, thereby avoiding the scale term being zero and maintaining the adaptability of the same batch of station data. To uniformly represent the interquartile range and its replacement scale value, the effective scale values ​​corresponding to the peak spectral misalignment, impulse concentration, and spatially isolated misalignment are denoted as follows: , and When the corresponding interquartile range is not zero, the effective scale value is taken as the interquartile range; when the corresponding interquartile range is zero and there is a non-zero absolute difference, the effective scale value is taken as the mean of the non-zero absolute deviation; when the corresponding indicator does not have a non-zero absolute difference, the robust standardization result corresponding to the indicator is directly set to zero.

[0160] After subtracting the median from each indicator of the target station and dividing by the corresponding effective scale value, robust normalization results for peak spectral misalignment, pulse concentration, and spatially isolated misalignment are obtained. The calculation method is as follows:

[0161]

[0162]

[0163]

[0164] in, This represents the robust standardized result of the peak spectral misalignment at the i-th station. This represents the robust normalization result of the pulse concentration at station i. This represents the robust standardization result of the spatial isolation misalignment of the i-th station. This represents the peak spectral misalignment at the i-th station. This represents the pulse concentration at the i-th station. This represents the spatial isolation misalignment of the i-th station. , and These represent the medians of the corresponding indicators. , and These represent the effective scale values ​​of the corresponding indicators. The reason for using this calculation method is that different earthquake cases have different magnitudes, station densities, station installation environments, and site conditions. Directly using the original indicators can easily lead to inconsistencies in scale across earthquake cases. Robust standardization can convert the performance of the target station in this earthquake into the degree of deviation relative to the effective stations in the same batch.

[0165] The pseudo-uplift intensity calculation unit extracts the portion of each robust standardized result that is greater than zero, which is then used as the positive part of each indicator. The calculation method is as follows:

[0166]

[0167]

[0168]

[0169] in, This represents the positive part of the robust normalization result of the peak spectral misalignment at the i-th station. This represents the positive part of the robust normalization result of the pulse concentration at station i. This represents the positive portion of the robust standardization result for the spatial isolation misalignment of the i-th station. The reason for using the positive portion is that this invention needs to suppress the spurious uplift effect where the peak instrument intensity is higher than the spectral intensity, the peak value is abnormally concentrated relative to the energy distribution, and the misalignment is isolated relative to the surrounding stations. For indicators that are lower than the batch reference level or in the opposite direction, the intensity of the spurious uplift should not be increased.

[0170] Furthermore, the pseudo-uplift strength calculation unit divides the product of the positive components of each index by the sum of the dimensionless constant 1 and that product to obtain the pseudo-uplift strength triggered by the structural boundary. The calculation method is as follows:

[0171]

[0172] in, This indicates the pseudo-uplift intensity triggered by the structural boundary of the i-th station. This represents the positive part of the robust normalization result of the peak spectral misalignment at the i-th station. This represents the positive part of the robust normalization result of the pulse concentration at station i. This represents the positive part of the robust standardization result of the spatial isolated misalignment of the i-th station. The dimensionless constant 1 is used to ensure that the pseudo-uplift intensity triggered by the structural boundary falls into the bounded interval that can be used for weight adjustment.

[0173] It should be noted that the structural boundary-triggered pseudo-uplift intensity refers to a dimensionless intensity value used to characterize the possibility of a peak-spectrum misalignment-type pseudo-uplift in intensity at a target station due to the influence of a local structural boundary. This intensity is jointly determined by the positive deviations of the peak-spectrum misalignment, the pulse concentration, and the spatial isolation misalignment. A larger value indicates that the peak intensity of the target station is more likely to originate from reflection, scattering, or transient response of the installed structure from the local structural boundary, rather than from a true synchronous enhancement of regional ground motion. The reason for using this calculation method is that only when all three terms—peak-spectrum misalignment, pulse concentration, and spatial isolation misalignment—are positively deviated will their product increase, and the structural boundary-triggered pseudo-uplift intensity will increase accordingly. When any one of these terms is not significant, the product will be suppressed, thus avoiding misjudgment of station records based solely on peak values, spectral intensity differences, or spatial isolation phenomena. The structural boundary-triggered pseudo-uplift intensity obtained in this step... Used for determining contribution weights in subsequent credibility intensity correction and model update.

[0174] In one embodiment of the present invention, a reliable intensity is determined based on the structural boundary-triggered pseudo-uplift intensity, peak intensity, and spectral intensity; a contribution weight is determined based on the structural boundary-triggered pseudo-uplift intensity; and the parameters of the seismic intensity attenuation model are updated by combining the reliable intensity and the contribution weight, including:

[0175] The pseudo-uplift intensity triggered by the structural boundary is obtained. Subsequently, the model parameter update unit determines the credible intensity based on the structural boundary-triggered pseudo-uplift intensity, peak intensity, and spectral intensity. It then determines the contribution weight based on the structural boundary-triggered pseudo-uplift intensity and updates the parameters of the seismic intensity attenuation model by combining the credible intensity and the contribution weight. Specifically, the model parameter update unit first subtracts the spectral intensity from the peak intensity to obtain the peak-spectral intensity difference, calculated as follows:

[0176]

[0177] in, This represents the peak intensity difference at the i-th station. This represents the peak intensity of the instrument at station i. This represents the spectral intensity of the i-th station.

[0178] The model parameter update unit extracts the portion of the peak intensity difference that is greater than zero as the positive peak difference value. The calculation method is as follows:

[0179]

[0180] in, This represents the positive difference in the peak spectrum of the i-th station. This represents the peak-spectral intensity difference at the i-th station. The reason for using a positive difference is that this invention needs to deduct the pseudo-increase in peak-type instrument intensity relative to spectral intensity. When the peak-type instrument intensity is not higher than the spectral intensity, the spectral intensity difference should not be used as the deduction object to avoid incorrectly correcting normal low-peak records to even lower intensities.

[0181] The model parameter update unit multiplies the positive difference in the peak spectrum with the intensity of the pseudo-uplift triggered by the structural boundary to obtain the intensity deduction. The calculation method is as follows:

[0182]

[0183] in, This represents the intensity deduction for the i-th station. This indicates the pseudo-uplift intensity triggered by the structural boundary of the i-th station. This represents the positive peak-spectral difference of the i-th station. The reason for using this calculation method is that the intensity deduction depends not only on the magnitude by which the peak-type instrument intensity exceeds the spectral intensity, but also on whether the difference simultaneously possesses the characteristics of short, sustained, sharp pulse concentration and spatial isolation, thereby avoiding the loss of effective station information caused by simply substituting spectral intensity for peak-type instrument intensity.

[0184] The model parameter update unit subtracts the intensity deduction from the peak instrument intensity to obtain the confidence intensity, calculated as follows:

[0185]

[0186] in, This represents the credibility intensity of the i-th station used for model updates. This represents the peak intensity of the instrument at station i. This represents the intensity deduction for the i-th station.

[0187] It should be noted that the credible intensity refers to the station intensity value used in updating the parameters of the seismic intensity attenuation model. This credible intensity is not simply derived from peak intensity or directly from spectral intensity. Instead, it is obtained by adaptively subtracting the positively inflated portion of the peak intensity relative to the spectral intensity based on the structural boundary triggered pseudo-uplift intensity. When the structural boundary triggered pseudo-uplift intensity is small, the credible intensity is close to the peak intensity; when the structural boundary triggered pseudo-uplift intensity is large, the credible intensity falls back towards the spectral intensity. The reason for using this calculation method is that when the structural boundary triggered pseudo-uplift intensity is small, the intensity subtraction is close to zero, and the credible intensity is close to the peak intensity. When the structural boundary triggered pseudo-uplift intensity is large, the credible intensity falls back from the peak intensity towards the spectral intensity, allowing suspected stations to retain some effective seismic information while reducing the impact of short-duration, non-destructive sharp pulses on model updates.

[0188] The model parameter update unit uses the dimensionless constant 1 to subtract the pseudo-uplift intensity triggered by the structural boundary to obtain the contribution weight, which is calculated as follows:

[0189]

[0190] in, This represents the contribution weight of the i-th station to the update of the seismic intensity attenuation model. The denot represents the pseudo-uplift intensity triggered by the structural boundary of the i-th station, and the dimensionless constant 1 represents the maximum baseline value of the contribution weight.

[0191] It should be noted that the contribution weight refers to the error weight of the target station in the update of the seismic intensity attenuation model parameters. This contribution weight is obtained by subtracting the structural boundary-triggered pseudo-uplift intensity from the dimensionless constant 1. The larger the structural boundary-triggered pseudo-uplift intensity, the smaller the contribution weight, indicating that the station is less representative of the regional seismic intensity attenuation law; conversely, the smaller the structural boundary-triggered pseudo-uplift intensity, the larger the contribution weight, indicating that the station can participate more fully in the model parameter update. The reason for using this calculation method is that the larger the structural boundary-triggered pseudo-uplift intensity, the more likely the station is to be affected by the transient response of local hard boundaries, equipment foundations, underground engineering structures, or installed structures, and the weaker its representativeness of the regional seismic intensity attenuation law. Therefore, its contribution in the model parameter update should be reduced accordingly.

[0192] The model parameter update unit obtains the epicentral distance or fault distance of the i-th station. The azimuth angle of the i-th station relative to the earthquake source The magnitude of this earthquake The focal depth of this earthquake And the major axis orientation angle in the earthquake intensity attenuation model The major axis dimension parameter A and the minor axis dimension parameter B. The epicentral distance or fault distance can be selected based on the way the earthquake source is represented. This represents the azimuth angle from the epicenter to the i-th station. This indicates the magnitude of the earthquake. This indicates the focal depth of the earthquake. The major axis direction angle represents the intensity attenuation, A represents the major axis scale parameter, and B represents the minor axis scale parameter; A and B are positive values. The epicenter location, focal depth, and fault geometry information of this earthquake are used in conjunction with station location data to determine... and The magnitude of this earthquake and focal depth Used as input for the source term in the earthquake intensity attenuation model.

[0193] The model parameter update unit subtracts the major axis direction angle from the azimuth angle to obtain the azimuth difference angle. It then calculates the major axis component based on the quotient of the cosine of the azimuth difference angle multiplied by the epicentral or fault distance and the major axis direction scale parameter. Similarly, it calculates the minor axis component based on the quotient of the sine of the azimuth difference angle multiplied by the epicentral or fault distance and the minor axis direction scale parameter. Finally, it calculates the square root of the sum of the squares of the major and minor axis components to obtain the equivalent directional distance. The calculation method is as follows:

[0194]

[0195] in, This represents the directional equivalent distance of the i-th station. This represents the epicentral distance or fault distance of the i-th station. Let represent the azimuth angle of the i-th station relative to the earthquake source. The major axis direction angle represents the intensity attenuation, A represents the major axis direction scale parameter, and B represents the minor axis direction scale parameter.

[0196] It should be noted that the directional equivalent distance refers to the equivalent distance obtained by projecting the distance and azimuth of the station relative to the seismic source onto the major and minor axes of intensity attenuation. This quantity is used to reflect the directional differences in seismic intensity attenuation within the same distance expression, enabling the model to distinguish between attenuation patterns with a wider impact range along the major axis and a shorter impact range along the minor axis, thereby supporting the generation of the axial distribution of the post-earthquake impact field. The reason for using the directional equivalent distance is that seismic intensity attenuation usually differs along the major and minor axes. If a single distance term is used directly, the model will find it difficult to express the axial expansion pattern of high-intensity areas. The directional equivalent distance allows the azimuth, major axis scale, and minor axis scale to be uniformly incorporated into the attenuation model.

[0197] The model parameter update unit constructs an earthquake intensity attenuation model with directional equivalent distance, earthquake magnitude, and earthquake focal depth as independent variables. The calculation method is as follows:

[0198]

[0199] in, This represents the predicted intensity at the i-th station output by the earthquake intensity attenuation model. This represents the set of model parameters to be updated. Represents a constant term. Indicates the magnitude impact coefficient. This represents the influence coefficient of focal depth. Indicates the distance attenuation coefficient. This indicates the magnitude of the earthquake. This indicates the focal depth of the earthquake. This represents the directional equivalent distance of the i-th station. (Using...) The reason is that the distance decay relationship usually changes non-linearly with the increase of distance. Adding a dimensionless constant 1 can avoid the logarithmic term being incalculable when the directional equivalent distance is zero, while ensuring that the near-field region can still participate in the model calculation.

[0200] In this embodiment, the set of parameters to be updated for the seismic intensity attenuation model is as follows:

[0201]

[0202] in, This represents the set of model parameters to be updated. Represents a constant term. Indicates the magnitude impact coefficient. This represents the influence coefficient of focal depth. Indicates the distance attenuation coefficient. A represents the major axis direction angle, B represents the major axis direction scale parameter, and A represents the minor axis direction scale parameter.

[0203] It should be noted that when model parameter updates involve magnitude influence coefficients and focal depth influence coefficients, the data center will combine historical earthquake case samples with valid station samples from the current earthquake to form a joint update sample. The historical earthquake case samples include station reliability intensity, spectral intensity, peak intensity, or macroscopic intensity data for different magnitudes and focal depths to ensure that the magnitude influence coefficients and focal depth influence coefficients can be identified by samples with differentiated focal parameters. When the update is performed only based on valid station data from a single target earthquake, since the magnitude and focal depth of this earthquake are the same for all stations, the data center will use the existing values ​​of the magnitude influence coefficients and focal depth influence coefficients from the current earthquake intensity attenuation model, updating only the constant term, distance attenuation coefficient, major axis direction angle, major axis direction scale parameter, and minor axis direction scale parameter. This avoids the problem of parameter unidentification caused by unchanged magnitude and focal depth in a single earthquake sample. When using a machine learning framework to perform model updates, the scope of the magnitude influence coefficient and focal depth influence coefficient in the update depends on whether the joint sample includes historical earthquakes of different magnitudes and focal depths.

[0204] The model parameter update unit calculates the square of the difference between the predicted intensity and the reliable intensity output by the seismic intensity attenuation model, multiplies it by the contribution weight to form a weighted error, and solves for the updated seismic intensity attenuation model parameters by minimizing the sum of the weighted errors of all valid stations. The calculation method is as follows:

[0205]

[0206] in, This represents the updated set of parameters for the earthquake intensity attenuation model. This represents the set of model parameters to be updated. This represents the contribution weight of the i-th station to the update of the seismic intensity attenuation model. This represents the predicted intensity at the i-th station output by the earthquake intensity attenuation model. This represents the credibility intensity of the i-th station used for model updates. This represents the epicentral distance or fault distance of the i-th station. Let represent the azimuth angle of the i-th station relative to the earthquake source. This indicates the magnitude of the earthquake. This indicates the depth of the earthquake's focal point.

[0207] The above minimization can be achieved using weighted least squares, gradient descent, quasi-Newton methods, or existing machine learning optimization frameworks in the data platform. During optimization, the current intensity attenuation model parameters are used as initial values, keeping the major axis scale parameter A and minor axis scale parameter B positive. The major axis angle can also be adjusted. Angle normalization is performed. When solving using numerical optimization, the data platform can ensure that A and B remain positive values ​​in each iteration by applying positive constraints, exponential parameter representation, or projection processing to the major and minor axis scale parameters; and by normalizing the major axis angle... When performing angle normalization, the angle is normalized to the same angle period to avoid the instability of model update results caused by the repeated expression of equivalent directions.

[0208] The reason for adopting this weighted update method is that the intensity of the spurious uplift triggered by structural boundaries simultaneously affects both the credible intensity and the contribution weight. This ensures that stations suspected of having local structural boundary-triggered peak-spectrum misalignment-type spurious uplifts are neither completely deleted nor are abnormally high peak-type instrumental intensities forcibly applied to the major axis azimuth angle, major axis scale parameters, minor axis scale parameters, and distance attenuation coefficient. This reduces the contamination of the regional seismic intensity attenuation pattern by local non-destructive sharp pulses. The updated seismic intensity attenuation model parameters obtained in this step... Used for generating subsequent post-earthquake impact fields.

[0209] In one embodiment of the present invention, the predicted intensity of each grid point within the target area is calculated using a parameter-updated seismic intensity attenuation model to generate an updated post-earthquake influence field, including:

[0210] After obtaining the updated earthquake intensity attenuation model parameters Subsequently, the post-earthquake impact field generation unit uses the updated seismic intensity attenuation model to calculate the predicted intensity of each grid point within the target area, thereby generating an updated post-earthquake impact field. Specifically, the data platform divides the target area into G grid points based on the target area boundary, emergency service range, epicenter location, and operational mapping accuracy, and obtains the location coordinates of each grid point. For the g-th grid point, the post-earthquake impact field generation unit calculates the epicentral distance or fault distance of that grid point relative to the epicenter. azimuth relative to the epicenter And obtain the magnitude of this earthquake. and the focal depth of this earthquake Where g represents the grid point number, This represents the epicentral distance or fault distance of the g-th grid point. This represents the azimuth angle of the g-th grid point relative to the earthquake source. This indicates the magnitude of the earthquake. This indicates the depth of the earthquake's focal point.

[0211] Substituting the parameters of each grid point into the updated seismic intensity attenuation model, the predicted intensity of each grid point is calculated as follows:

[0212]

[0213] in, This represents the predicted intensity at the g-th grid point within the target area. Indicates the use of updated parameters Earthquake intensity attenuation model, This represents the epicentral distance or fault distance of the g-th grid point. This represents the azimuth angle of the g-th grid point relative to the earthquake source. This indicates the magnitude of the earthquake. This indicates the focal depth of the earthquake. The reason for using this calculation method is that the updated earthquake intensity attenuation model has incorporated the effective station information of this earthquake using credible intensity and contribution weights, and suppressed the influence of peak-spectrum misalignment-type intensity pseudo-uplift triggered by structural boundaries. Therefore, inputting the distance, azimuth, magnitude, and focal depth of the grid points into the model can yield a predicted intensity that is more suitable for rapid post-earthquake assessment.

[0214] The post-earthquake impact field generation unit uses the locations of each grid point within the target area and their corresponding predicted intensities to form a spatial distribution set, which serves as the updated post-earthquake impact field. The calculation method is as follows:

[0215]

[0216] in, This indicates the updated post-earthquake impact field. This represents the horizontal coordinate of the g-th grid point. This represents the vertical coordinate of the g-th grid point. This represents the predicted intensity at the g-th grid point. This indicates the total number of grid points within the target area.

[0217] It should be noted that the post-earthquake impact field refers to the spatial distribution result composed of the locations of each grid point within the target area and their corresponding predicted intensities. This spatial distribution result can be represented as a raster layer with isopleths, an emergency impact range layer, or a result returned by a data interface, used to characterize the predicted intensity distribution at different spatial locations after the target earthquake. The data platform can output the updated post-earthquake impact field as a raster layer with isopleths, an emergency impact range layer, or a result returned by a data interface, and can continue to repeat the steps in this embodiment based on subsequently added station waveforms, macro-intensity survey data, and earthquake case samples to achieve continuous updating of the earthquake intensity attenuation model.

[0218] Through the above implementation process, the horizontal synthetic acceleration time history, peak acceleration, and peak instrument intensity are directly obtained from the station waveform data. The spectral intensity is obtained by integrating the horizontal relative velocity response spectrum and transforming it through a history fitting function. The peak spectral misalignment is used to characterize the deviation of the peak instrument intensity from the spectral intensity. The pulse concentration is used to characterize the abruptness of the maximum peak relative to the full-time energy distribution. The spatial isolation misalignment is used to characterize the local isolation deviation of the target station relative to the surrounding stations. The structural boundary triggered pseudo-uplift intensity is determined by the above three types of results and further enters the calculation of credible intensity and contribution weight. The credible intensity and contribution weight are used together to update the parameters of the seismic intensity attenuation model. The updated model parameters are then used to calculate the predicted intensity of each grid point and generate the post-earthquake influence field. Therefore, in this embodiment, each intermediate result is directly used by the subsequent steps, forming a complete technical closed loop from station waveform acquisition, spectral intensity calculation, pseudo-uplift identification, model parameter update to post-earthquake influence field generation.

[0219] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for updating a seismic intensity attenuation model oriented towards the generation of post-earthquake influence fields, characterized in that, Includes the following steps: S1. Obtain the station waveform data, station location data and source parameters of the target earthquake, calculate the horizontal composite acceleration time history based on the station waveform data, and extract the peak acceleration and peak velocity to convert them into peak instrument intensity. S2. Extract the horizontal relative velocity response spectrum value based on the horizontal acceleration time history in the station waveform data, integrate the horizontal relative velocity response spectrum value within the determined integration interval to obtain the spectral intensity, and convert the spectral intensity into spectral intensity. S3. Calculate the peak spectral misalignment based on the difference between the peak instrument intensity and the spectral intensity, and extract the pulse concentration based on the energy distribution of the horizontal synthetic acceleration time history. S4. Calculate the spatial reference weight based on the station location data, extract the spatial isolated misalignment based on the peak spectrum misalignment of surrounding stations, and construct the structural boundary trigger pseudo-lifting intensity based on the peak spectrum misalignment, pulse concentration and spatial isolated misalignment. S5. Determine the credible intensity based on the structural boundary triggered pseudo-uplift intensity, peak instrument intensity, and spectral intensity; determine the contribution weight based on the structural boundary triggered pseudo-uplift intensity; and update the parameters of the seismic intensity attenuation model by combining the credible intensity and the contribution weight. S6. Using the updated seismic intensity attenuation model, calculate the predicted intensity of each grid point in the target area to generate the updated post-earthquake impact field.

2. The method for updating a seismic intensity attenuation model for generating post-earthquake influence fields according to claim 1, characterized in that, Acquire station waveform data, station location data, and source parameters for the target earthquake. Calculate the horizontal composite acceleration time history based on the station waveform data, and extract peak ground acceleration and peak velocity to convert them into peak-type instrumental intensity, including: Obtain the east-west and north-south acceleration time histories of the target station; The horizontal composite acceleration time history of the target station is constructed by summing the squares of the east-west acceleration time history and the squares of the north-south acceleration time history. The maximum absolute value of the horizontal composite acceleration time history is extracted as the peak acceleration; the east-west acceleration time history and the north-south acceleration time history are integrated over time to obtain the east-west velocity time history and the north-south velocity time history. The horizontal composite velocity time history is obtained by summing the squares of the east-west velocity time history and the squares of the north-south velocity time history and taking the square root. The maximum value of the horizontal composite velocity time history is extracted as the peak velocity; Based on the conversion relationship obtained from historical earthquake data or instrument intensity specifications, peak acceleration and peak velocity are converted into peak-type instrument intensity.

3. The method for updating a seismic intensity attenuation model for generating post-earthquake influence fields according to claim 1, characterized in that, The horizontal relative velocity response spectrum is extracted from the horizontal acceleration time history in the station waveform data. The spectral intensity is obtained by integrating the horizontal relative velocity response spectrum within a defined integration interval, and then converted into spectral intensity, including: Based on the east-west acceleration time history, north-south acceleration time history, damping ratio determined by historical earthquake examples or according to the earthquake response spectrum calculation rules, and structural vibration period from the station waveform data, the east-west single-degree-of-freedom system response and the north-south single-degree-of-freedom system response are calculated respectively, and the east-west relative velocity response and the north-south relative velocity response are obtained. The maximum values ​​of the absolute values ​​of the east-west relative velocity response and the north-south relative velocity response were extracted at the sampling time to obtain the east-west relative velocity response spectrum and the north-south relative velocity response spectrum. The horizontal relative velocity response spectrum value is obtained by summing the squares of the east-west relative velocity response spectrum value and the squares of the north-south relative velocity response spectrum value and taking the square root. Obtain the macroscopic intensity from the historical earthquake sample set, and determine the lower limit and upper limit of the candidate period that maximize the correlation between the spectral intensity and the macroscopic intensity within the candidate period range. These are used as the lower limit and upper limit of the spectral intensity integral period, respectively. Within the range defined by the lower limit and the upper limit of the spectral intensity integration period, the horizontal relative velocity response spectrum value is integrated, and the integration result is divided by the difference between the upper limit and the lower limit of the spectral intensity integration period to obtain the spectral intensity. Based on the fitting function between spectral intensity and macroscopic intensity in historical earthquake cases, spectral intensity is converted into spectral intensity.

4. The method for updating a seismic intensity attenuation model for generating post-earthquake influence fields according to claim 1, characterized in that, The peak-spectral misalignment is calculated based on the difference between the peak instrument intensity and the spectral intensity, and the pulse concentration is extracted based on the energy distribution of the horizontal synthetic acceleration time history, including: Subtracting the spectral intensity from the peak-type instrument intensity yields the peak spectral misalignment. The full-time energy is obtained by integrating the square of the horizontal composite acceleration time history over the entire sampling time domain. Calculate the time integral of the product of the sampling time and the square of the horizontal composite acceleration time history, and divide the time integral by the total time history energy to obtain the energy center time; The time integral of the product of the square of the difference between the sampling time and the energy center time and the square of the time history of the horizontal composite acceleration is calculated. The square root of this time integral is then taken after dividing it by the total time history energy to obtain the energy spread time. Dividing the peak acceleration by the square root of the ratio of the full-time energy to the energy spread time yields the pulse concentration.

5. The method for updating a seismic intensity attenuation model for generating post-earthquake influence fields according to claim 1, characterized in that, Spatial reference weights are calculated based on station location data, and spatial isolated misalignment is extracted by combining peak spectral misalignment with that of surrounding stations, including: Identify valid stations that are not in the target station but are spatially different from them as surrounding stations; The spatial distance between the target station and each of the surrounding stations is calculated by taking the square root of the sum of the squares of the differences in the horizontal coordinates and the squares of the differences in the vertical coordinates between the target station and each of the surrounding stations. The spatial reference weight is obtained by dividing the reciprocal of the spatial distance between the target station and a single surrounding station by the sum of the reciprocals of the spatial distances between the target station and all surrounding stations. The peak spectrum misalignment of each surrounding station is weighted and summed using spatial reference weights to obtain the weighted peak spectrum misalignment reference value. Subtracting the weighted peak misalignment reference value from the peak misalignment value of the target station yields the spatial isolated misalignment value.

6. The method for updating a seismic intensity attenuation model for generating post-earthquake influence fields according to claim 1, characterized in that, The structural boundary-triggered pseudo-lifting intensity is constructed based on peak spectral misalignment, pulse concentration, and spatially isolated misalignment, including: The median and interquartile range of peak spectral misalignment, pulse concentration, and spatial isolated misalignment of all valid stations in this earthquake were obtained respectively. When the interquartile range of any indicator is zero, the mean of the non-zero absolute deviation of that indicator is used to replace the interquartile range. The mean of the non-zero absolute deviation is the average of the non-zero absolute differences between the corresponding indicators of all valid stations in this earthquake and the median of that indicator. When there is no non-zero absolute difference for the corresponding indicator, the robust standardization result corresponding to the indicator is set to zero. After subtracting the corresponding median from each indicator of the target station, divide by the corresponding interquartile range or the mean of non-zero absolute deviation to obtain robust standardized results of peak spectral misalignment, pulse concentration, and spatial isolated misalignment, respectively. Extract the portion of each robust standardized result that is greater than zero, and use it as the positive part of each indicator. The structural boundary-triggered pseudo-uplift intensity is obtained by multiplying the positive parts of each indicator and dividing by the sum of the numerical value and the product.

7. The method for updating a seismic intensity attenuation model for generating post-earthquake influence fields according to claim 1, characterized in that, The credible intensity is determined based on the structural boundary-triggered pseudo-uplift intensity, peak intensity, and spectral intensity. The contribution weight is determined based on the structural boundary-triggered pseudo-uplift intensity. The parameters of the seismic intensity attenuation model are then updated by combining the credible intensity and the contribution weight, including: Subtract the spectral intensity from the peak-type instrument intensity to obtain the peak-spectral intensity difference; Extract the portion of the peak intensity difference that is greater than zero as the positive peak difference value; Multiply the positive difference in the peak spectrum by the intensity of the pseudo-uplift triggered by the structural boundary to obtain the intensity deduction; Subtract the intensity deduction from the peak intensity to obtain the reliable intensity; The contribution weight is obtained by subtracting the pseudo-uplift intensity triggered by the structural boundary from the numerical value; Obtain the epicentral distance or fault distance of the station, the azimuth of the station relative to the source, the magnitude of the earthquake, the source depth of the earthquake, as well as the major axis direction angle, major axis direction scale parameter and minor axis direction scale parameter in the earthquake intensity attenuation model. The azimuth difference angle is obtained by subtracting the major axis direction angle from the azimuth angle. The major axis component is calculated based on the quotient of the cosine of the azimuth difference angle multiplied by the epicentral distance or fault distance and the major axis direction scale parameter. The minor axis component is calculated based on the quotient of the sine of the azimuth difference angle multiplied by the epicentral distance or fault distance and the minor axis direction scale parameter. Calculate the square root of the sum of the squares of the major axis components and the squares of the minor axis components to obtain the directional equivalent distance; An earthquake intensity attenuation model is constructed with directional equivalent distance, magnitude of the earthquake, and focal depth of the earthquake as independent variables. The parameters to be updated in the earthquake intensity attenuation model include constant terms, magnitude influence coefficient, focal depth influence coefficient, distance attenuation coefficient, major axis direction angle, major axis direction scale parameter, and minor axis direction scale parameter, among which the major axis direction scale parameter and the minor axis direction scale parameter are positive. Calculate the square of the difference between the predicted intensity and the credible intensity output by the earthquake intensity attenuation model, and multiply it by the contribution weight to form a weighted error; The updated seismic intensity attenuation model parameters are solved by minimizing the sum of weighted errors from all stations.

8. The method for updating a seismic intensity attenuation model for generating post-earthquake influence fields according to claim 1, characterized in that, Using the updated seismic intensity attenuation model, the predicted intensity at each grid point within the target area is calculated to generate the updated post-earthquake impact field, including: Obtain the epicentral distance or fault distance, azimuth relative to the hypocenter, magnitude of the earthquake, and focal depth of each grid point within the target area; Substitute the above parameters of each grid point into the updated seismic intensity attenuation model to calculate the predicted intensity of each grid point. The spatial distribution set of the locations of each grid point in the target area and the corresponding predicted intensity is used as the updated post-earthquake impact field.

Images