Event-driven real-time earthquake information processing and intelligent distribution system

By implementing an event-driven real-time earthquake information processing and intelligent distribution system, accurate monitoring and intelligent distribution of earthquake events have been achieved, solving the problems of information redundancy and low emergency response efficiency in existing technologies, and improving the reliability and continuity of earthquake early warning.

CN122860504BActive Publication Date: 2026-11-17SHANDONG SEISMOLOGICAL BUREAU
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611373329.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-09-07
Publication Date
2026-11-17
Estimated Expiration
2046-09-07

AI Technical Summary

Technical Problem

Existing real-time earthquake information processing and intelligent distribution systems cannot quantify and classify the urgency of earthquake events from multiple dimensions, resulting in information redundancy and low emergency response efficiency. They are unable to effectively manage the large number of earthquake events that emerge in a short period of time, affecting the reliability and continuity of earthquake early warning.

Method used

An event-driven real-time earthquake information processing and intelligent distribution system is adopted. The earthquake event monitoring module extracts data fluctuation characteristics and magnitude standard values. Combined with the earthquake type determination module and the intelligent distribution judgment module, the system can accurately monitor and intelligently distribute earthquake events. The system uses epicenter level and regional affiliation for quantitative classification to ensure the targetedness and stability of the distribution.

Benefits of technology

It improves the accuracy and targeting of earthquake event monitoring and distribution, reduces information redundancy, ensures stable and continuous distribution by the earthquake network center during high-frequency events, and guarantees the efficiency and reliability of earthquake information processing and distribution.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122860504B_ABST
    Figure CN122860504B_ABST
Patent Text Reader

Abstract

This invention relates to the field of earthquake monitoring and early warning technology. It discloses an event-driven real-time earthquake information processing and intelligent distribution system, comprising: an earthquake event monitoring module for determining whether an earthquake event has occurred; an earthquake type determination module for determining the earthquake type of the earthquake event; an intelligent distribution determination module for determining whether to execute an intelligent distribution operation; an earthquake event distribution module for intelligently distributing the event; and an event record aggregation module for constructing an event record of the earthquake event. This invention can accurately distinguish between different types of earthquake events of varying urgency, improving the targeting of earthquake information distribution and emergency response efficiency. Furthermore, by using a heartbeat detection mechanism to judge the health status of the distribution center in real time, it ensures that the earthquake network center can maintain continuous distribution even when facing a concentrated outbreak of high-frequency earthquake events, guaranteeing the efficiency and reliability of real-time earthquake information processing and intelligent distribution.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of earthquake monitoring and early warning technology, and more specifically, to an event-driven real-time earthquake information processing and intelligent distribution system. Background Technology

[0002] As a sudden natural disaster, the extent of damage caused by earthquakes is closely related to the timeliness and coverage of early warning information. After an earthquake occurs, the ability to detect, analyze, and distribute earthquake events in the shortest possible time is of great significance for mitigating earthquake disaster losses and protecting people's lives and property. With the continuous improvement of earthquake monitoring network construction and the rapid development of communication technology, real-time processing and intelligent distribution of earthquake information have become important means in the field of earthquake early warning.

[0003] Reference patent application CN121838425A discloses a graded automatic push system and method for earthquake early warning information based on regional division. This system dynamically divides risk areas by combining geographical and demographic information; assesses risk and generates early warning level sequences using decision tree and random forest algorithms; generates personalized early warning information sets based on a level matching template library; pushes information in a graded manner through a broadcast network and backup satellite channels to ensure coverage; and generates optimization instructions based on feedback data comparison benchmarks, dynamically updating the division and evaluation model to form a closed loop. This achieves accurate grading and reliable delivery of earthquake early warning information. Existing real-time earthquake information processing and intelligent distribution systems typically employ a one-size-fits-all approach to push all earthquake information. This approach fails to quantify and classify the urgency of earthquake events across multiple dimensions, resulting in a flood of low-correlation earthquake information in the push channels. This leads to severe information redundancy, reducing the targeted nature of earthquake early warnings and the efficiency of emergency response. Furthermore, it cannot systematically arrange and manage the large number of earthquake events that emerge in a short period of time, easily causing problems such as disordered earthquake event distribution order and duplicate distribution of earthquake events. This undermines the reliability and continuity of real-time earthquake information processing and intelligent distribution, and fails to guarantee the long-term stable operation of earthquake network centers in emergency situations.

[0004] In view of this, the present invention proposes an event-driven real-time earthquake information processing and intelligent distribution system to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: an event-driven real-time earthquake information processing and intelligent distribution system, applied to an earthquake network center, comprising: The earthquake event monitoring module is used to extract data fluctuation characteristics from multi-source raw data of the target area. The data fluctuation characteristics include fluctuation coefficients and standard magnitude values. Based on the data fluctuation characteristics, it determines whether an earthquake event has occurred in the target area. The earthquake type determination module is used to parse type features from the phase files of earthquake events and input the type features into the earthquake classification model to determine the earthquake type of the earthquake event. The intelligent distribution determination module is used to extract the magnitude regional characteristics of earthquake events, including the epicenter level and epicenter region affiliation, and to determine the distribution nature of earthquake events based on the magnitude regional characteristics, and to determine whether to perform intelligent distribution operation. The earthquake event distribution module is used to construct earthquake information based on the source parameters, fill the earthquake information into the event template to generate distribution events, and arrange the distribution events into distribution queues based on the event distribution criteria when the distribution center is in a healthy state, and intelligently distribute the distribution events in the distribution queues. The event log summary module is used to build event logs for earthquake events and import the event logs into a historical database for data storage and updates.

[0006] Furthermore, the fluctuation coefficient and the standard value of magnitude are extracted as follows: Using the current time as the first sampling point, and with a preset sampling step size as the interval, A sampling points with interval distribution are deduced by going backward along the time line; Collect multi-source raw data from A sampling points one by one, mark the fields of the multi-source raw data, and arrange the raw values ​​of the multi-source raw data with the same fields into a data sequence; Starting from the second sampling point, the original values ​​of two adjacent sampling points are used as comparison objects, and the fluctuation coefficient of the data sequence is calculated by the fluctuation calculation function; The multi-source raw data from C monitoring stations were analyzed sequentially, and the raw values ​​of magnitude from the same sampling point were extracted to obtain C magnitude values. The average magnitude is calculated by summing the C magnitude values. The average magnitude and the C magnitude values ​​are then substituted into the standard deviation formula to calculate the standard magnitude.

[0007] Furthermore, the method for determining whether an earthquake event has occurred is as follows: The number of data sequences whose volatility coefficients exceed a preset volatility threshold is counted to obtain the volatility exceedance amount; When the fluctuation threshold is greater than or equal to 1 and the magnitude standard value is greater than the preset magnitude standard threshold, it is determined that an earthquake event has occurred in the target area. When the fluctuation threshold is less than 1, or the magnitude standard value is less than or equal to the preset magnitude standard threshold, it is determined that no earthquake event has occurred in the target area.

[0008] Furthermore, the type characteristics include waveform arrival time difference, P-wave amplitude ratio, P-wave main period, S-wave main period, P-wave low-frequency energy ratio, and S-wave low-frequency energy ratio.

[0009] Furthermore, the analytical method for the proportion of P-wave low-frequency energy is as follows: The analysis time is marked after extending the P-wave arrival time by a preset interception period. The time period between the start time and the analysis time is recorded as the analysis window, and the waveform data of the acceleration value within the analysis window is recorded as the acceleration window signal. A short-time Fourier transform is performed on the acceleration window signal, using the Hanning window as the sliding window function; Set the window length and overlap step, slide along the time axis frame by frame, calculate the local spectrum of all frames, and stack the local spectra of all frames in chronological order to generate a time-spectrum graph; The effective response frequency band and sampling frequency of the accelerometer are queried from the technical parameter table. The lower limit of the analysis frequency is set with one-quarter of the effective response frequency band as the first standard, and the upper limit of the analysis frequency is set with half of the sampling frequency as the second standard. The intermediate frequency analysis frequency is also customized. Using the lower limit to the upper limit of the analysis frequency as the first constraint and the analysis window as the second constraint, the time spectrum is double-integrated to calculate the total energy of the entire frequency band. Using the lower limit of the analysis frequency to the mid-frequency analysis frequency as the low-frequency range, the total energy of the low-frequency band is calculated, and the total energy of the low-frequency band is divided by the total energy of the entire frequency band to calculate the energy ratio of the P-wave low-frequency band.

[0010] Furthermore, earthquake types include natural earthquakes and non-natural earthquakes; Label natural earthquakes as 00 and non-natural earthquakes as 11; When the output earthquake type label is 00, the earthquake event is a natural earthquake; when the output earthquake type label is 11, the earthquake event is a non-natural earthquake.

[0011] Furthermore, the nature of distribution includes necessary distribution and unnecessary distribution; The method for determining whether to perform intelligent distribution is as follows: When the epicenter area belongs to the province and the epicenter level is greater than 0, both the epicenter area and the epicenter level are recorded as necessary features. When the epicenter area belongs to a neighboring province and the epicenter level is greater than or equal to the preset first level threshold, both the epicenter area affiliation and the epicenter level are recorded as necessary features. When the epicenter area is classified as domestic and the epicenter level is greater than or equal to the preset second-level threshold, both the epicenter area classification and the epicenter level are recorded as necessary features. The number of necessary features in an earthquake event is counted. If the number of necessary features is 2, the necessary property is determined to be necessary distribution, and the intelligent distribution operation is executed. If the number of necessary features is less than 2, the necessary properties are deemed unnecessary to distribute, and the intelligent distribution operation is not performed.

[0012] Furthermore, the source parameters include the time of occurrence, epicenter location, epicenter coordinates, source depth, and epicenter distance; earthquake information is a summary of source parameters, earthquake type, epicenter magnitude, and epicenter regional affiliation.

[0013] Furthermore, the method for determining the health status of a distribution center is as follows: Based on a preset detection cycle, a heartbeat detection signal is periodically sent to the distribution center, and the timestamp of the distribution center's response to the heartbeat detection signal is recorded. Get the timestamp of the distribution center at the current moment, and record the duration between the current timestamp and the timestamp of the most recent heartbeat detection signal response as the heartbeat interval value; When the heartbeat interval value is less than or equal to the preset heartbeat timeout threshold, the distribution center is determined to be in a healthy state; When the heartbeat interval value is greater than the preset heartbeat timeout threshold, the distribution center is determined to be in an unhealthy state.

[0014] Furthermore, the event distribution criteria are as follows: mandatory distribution is the primary criterion, and epicenter level is the secondary criterion. The method for intelligent event distribution is as follows: Query the mandatory status of all dispatch events one by one, and arrange the dispatch events with the mandatory status of mandatory dispatch before the dispatch events with the mandatory status of non-mandatory dispatch to obtain the basic queue; The distribution events in the basic queue are arranged in descending order of epicenter level to generate a distribution queue. Each distribution event is marked with a unique event ID. The first distribution event in the distribution queue is recorded as the current event. If the current earthquake is a natural earthquake and the epicenter is located within the province, the earthquake information of the current event will be sent to all user groups simultaneously. If the current earthquake type is a natural earthquake, the epicenter area belongs to a neighboring province, and the epicenter level is greater than or equal to the preset first level threshold, the earthquake information of the current event will be sent to the first user group and the second user group. If the current earthquake is a natural earthquake, the epicenter is located within China, and the epicenter level is greater than or equal to the preset second-level threshold, the earthquake information of the current event will be sent to the second user group and the third user group. If the earthquake type of the current event is a non-natural earthquake, and the epicenter level is greater than or equal to the preset second level threshold, the earthquake information of the current event will be sent to the second user group. Iterate through all dispatch events in the dispatch queue until all dispatch events have been intelligently dispatched.

[0015] The technical advantages of this invention, based on an event-driven real-time earthquake information processing and intelligent distribution system, are as follows: (1): This invention uses a dual verification mechanism of fluctuation amplitude and magnitude consistency to accurately monitor earthquake events from the dimensions of data time sequence anomalies and spatial measurement consistency, thereby eliminating the comprehensive interference caused by the event and spatial dimensions, reducing the probability of false triggering of earthquake events due to station noise, electromagnetic interference or data packet loss, and avoiding the problem of missed earthquake events due to insufficient sensitivity of individual stations, thus improving the accuracy of earthquake event monitoring results.

[0016] (2): This invention adopts a combination of quantitative classification of epicenter level and attribution analysis of epicenter area, which can accurately distinguish different types of earthquake events with different urgency levels. It avoids the problem of information redundancy and key information submersion caused by the one-size-fits-all full push of earthquake information, improves the pertinence of earthquake information distribution and emergency response efficiency, reduces interference from invalid information, and ensures that the earthquake network center can maintain a stable, orderly and non-repetitive continuous distribution state when facing the concentrated outbreak of high-frequency earthquake events by judging the health status of the distribution center in real time through the heartbeat detection mechanism. This ensures the efficiency and reliability of real-time processing and intelligent distribution of earthquake information. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the event-driven real-time earthquake information processing and intelligent distribution system provided in Embodiment 1 of the present invention. Figure 2 This is a schematic diagram of the architecture of an event-driven real-time earthquake information processing and intelligent distribution system provided in Embodiment 1 of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Please refer to Figures 1-2 As shown in this embodiment, the event-driven real-time earthquake information processing and intelligent distribution system is applied to an earthquake network center and includes: The earthquake event monitoring module extracts data fluctuation characteristics from multi-source raw data of the target area and determines whether an earthquake event has occurred in the target area based on the data fluctuation characteristics. The target area is the corresponding provincial region where the earthquake network center performs real-time processing and intelligent distribution of earthquake information, and serves as the source and analysis location for subsequent relevant data and information.

[0020] Real-time monitoring of earthquake events is the trigger point for the operation of the entire system. During the earthquake monitoring process, the earthquake network center continuously generates massive amounts of monitoring data. These data come from different monitoring stations and different sensor types, constituting multi-source raw data. Multi-source raw data includes, but is not limited to, acceleration values, velocity values, displacement values, GNSS ground displacement values, and magnitude records from seismic network centers.

[0021] To ensure the accuracy and reliability of earthquake event monitoring and avoid misjudgments caused by anomalies in a single data source, it is necessary to conduct comprehensive data compliance analysis on raw data from multiple sources. This can effectively overcome the limitations of a single data source in a specific scenario and improve the robustness and accuracy of earthquake event monitoring. Specifically, data compliance analysis includes data volatility analysis and magnitude consistency analysis.

[0022] Data volatility analysis is used to determine whether multi-source raw data has exhibited abnormal fluctuations beyond the normal range. Under normal conditions, multi-source raw data should remain relatively stable, but when an earthquake event occurs, multi-source raw data will exhibit drastic fluctuations.

[0023] Magnitude consistency analysis is used to determine whether the magnitudes calculated by different monitoring stations for the same earthquake event are consistent. Since the attenuation characteristics of seismic waves are different on different propagation paths, the magnitudes calculated by different monitoring stations may differ to some extent. If the difference is too large, it indicates that there may be data anomalies or false triggering.

[0024] Data fluctuation characteristics are used to specifically represent the results of data compliance analysis and serve as a direct basis for subsequent judgment on whether an earthquake event has occurred in the target area. The data fluctuation characteristics include the fluctuation coefficient and the standard value of magnitude; a dual analysis is conducted from two dimensions: the fluctuation range of the data at different time lines and the fluctuation range of the data at different monitoring stations.

[0025] The wave coefficient and the standard value of magnitude are extracted as follows: Using the current moment as the first sampling point, and with a preset sampling step size as the interval, A sampling points are deduced backward along the time line at intervals. The preset sampling step size refers to the time interval between two adjacent sampling points, which ensures that any two adjacent sampling points maintain an equal time interval. In this embodiment, the preset sampling step size is the reciprocal of the sampling frequency. For example, when the sampling frequency is 100Hz, the preset sampling step size is 0.01s. Collect multi-source raw data from A sampling points one by one, mark the fields of the multi-source raw data, and arrange the original values ​​of the multi-source raw data with the same field in sequence to generate B data sequences. Fields are used to concisely and accurately represent the meaning of specific data in multi-source raw data, thereby providing a basis for the accurate identification of multi-source raw data. Starting from the second sampling point, the original values ​​of two adjacent sampling points are used as comparison objects, and the fluctuation coefficient of the data sequence is calculated by the fluctuation calculation function; The expression for the fluctuation calculation function is: ; In the formula, For fluctuation coefficient, For the first data sequence The original values ​​of each sampling point For the first data sequence The original values ​​of each sampling point; The multi-source raw data from C monitoring stations were analyzed sequentially, and the raw values ​​of magnitude from the same sampling point were extracted to obtain C magnitude values. The average of the C magnitude values ​​is calculated by summing them up. The average magnitude and the C magnitude values ​​are then substituted into the standard deviation formula to calculate the standard magnitude. The expression for the standard deviation formula is: ; In the formula, This is the standard value for magnitude. For the first Each earthquake magnitude value, This represents the average magnitude.

[0026] Once the data fluctuation characteristics are obtained, they can be used as a direct basis for determining whether an earthquake event has occurred in the target area. By determining whether an earthquake event has occurred in the target area, a foundation is provided for the real-time processing and intelligent distribution of earthquake information. The method for determining whether an earthquake event has occurred is as follows: The fluctuation coefficients of each of the B data sequences are compared with a preset fluctuation threshold, and the number of data sequences with fluctuation coefficients greater than the preset fluctuation threshold is counted to obtain the fluctuation threshold exceedance. The preset fluctuation threshold is the minimum fluctuation coefficient value required to determine the occurrence of an earthquake event, and serves as the benchmark threshold for fluctuation threshold judgment. Specifically, the preset fluctuation threshold is obtained by collecting a large number of historical fluctuation coefficients that conform to earthquake events and calculating their average value. When the fluctuation threshold is greater than or equal to 1, and the magnitude standard value is greater than the preset magnitude standard threshold, it indicates that anomalies have occurred in both the data fluctuation level and the magnitude fluctuation level of the multi-source raw data of the target area. At this time, the target area meets the preconditions for the occurrence of an earthquake event, and it is determined that an earthquake event has occurred in the target area. The preset magnitude standard threshold refers to the minimum magnitude standard value that will be determined as an earthquake event, which provides a basic threshold for the magnitude standard value of an earthquake event. Specifically, the preset magnitude standard threshold is obtained by statistical analysis of the upper limit of the distribution of magnitude standard values ​​calculated from multi-source raw data of historical earthquake-free periods or the lower limit of the statistical distribution of magnitude standard values ​​of a large number of historical earthquake events. When the fluctuation threshold is less than 1, or the magnitude standard value is less than or equal to the preset magnitude standard threshold, it indicates that the multi-source raw data of the target area has not shown any abnormalities at both the data fluctuation level and the magnitude fluctuation level. At this time, the target area does not meet the preconditions for the occurrence of an earthquake event, and it is determined that no earthquake event has occurred in the target area.

[0027] It should be noted that by comprehensively analyzing and judging the two dimensions of data fluctuation amplitude and magnitude fluctuation amplitude, the phenomenon of false alarms and missed alarms of earthquake events caused by noise interference from a single monitoring station, abnormal single data source, or poor data quality is effectively avoided, and the accuracy and reliability of earthquake event monitoring are significantly improved.

[0028] The earthquake type determination module parses type features from the phase files of earthquake events and inputs these features into the earthquake classification model to determine the earthquake type of the earthquake event. Because different types of earthquake events have different focal mechanisms and waveform characteristics, it is necessary to further determine the specific type of earthquake after an earthquake event is detected. Natural earthquakes mostly occur deep in the Earth's crust, and their seismic phase characteristics are characterized by complex P-wave initial motion directions and fluctuations in the P-wave / S-wave amplitude ratio within a certain range. Non-natural earthquakes (such as those caused by artificial blasting or mine collapses) are characterized by relatively simple P-wave initial motion directions and distinct P-wave / S-wave amplitude ratios. In order to effectively distinguish the specific type of earthquake, it is necessary to collect seismic phase files of the earthquake event. In this embodiment, the phase file is a seismic waveform data file recorded by the seismic network center. It is obtained by querying the waveform database and is used to record complete waveform data of P-waves, S-waves and surface waves. The phase file includes, but is not limited to, P-wave arrival time, S-wave arrival time, maximum P-wave amplitude, maximum S-wave amplitude, P-wave vibration period, and S-wave vibration period.

[0029] Type features are various type features that can be parsed from seismic phase files to represent specific earthquake types, and provide reliable data support for the subsequent identification and determination of specific earthquake types; Specifically, the type characteristics include waveform arrival time difference, P-wave amplitude ratio, P-wave main period, S-wave main period, P-wave low-frequency energy ratio, and S-wave low-frequency energy ratio.

[0030] Waveform arrival time difference refers to the difference between the arrival time of the P-wave and the arrival time of the S-wave in a seismic event; the arrival time difference between the P-wave and the S-wave reflects the distance between the seismic source and the monitoring station, and is one of the important parameters for distinguishing different types of earthquakes. In this embodiment, when extracting the waveform arrival time difference, the arrival time of the P wave and the arrival time of the S wave are respectively parsed from the phase file, and the absolute value of the difference between the two is taken to calculate the waveform arrival time difference.

[0031] The P-wave amplitude ratio is the ratio between the maximum amplitude of the P-wave and the maximum amplitude of the S-wave. The P-wave amplitude ratio is one of the effective criteria for distinguishing between natural and non-natural earthquakes. The P-wave amplitude ratio of natural earthquakes usually shows a statistical regularity within a certain range, while the P-wave amplitude ratio of artificial blasting often shows obvious deviations. In this embodiment, when extracting the P-wave and S-wave amplitude ratio, the maximum amplitude of the P-wave and the maximum amplitude of the S-wave are analyzed from the phase file, and the P-wave and S-wave amplitude ratio is calculated by dividing the maximum amplitude of the P-wave by the maximum amplitude of the S-wave.

[0032] The P-wave principal period refers to the vibration period corresponding to the wave with the strongest energy and largest amplitude in the P-wave oscillation; the S-wave principal period refers to the vibration period corresponding to the wave with the strongest energy and largest amplitude in the S-wave oscillation. The periodic characteristics of seismic waves reflect the characteristics of the source rupture process and the propagation medium. Different types of earthquake events have significant differences in their periodic characteristics. Specifically, in natural earthquakes, the rupture process of crustal rocks is slow and large in scale, the rupture propagation speed is slow, and the low-frequency components excited are abundant. Therefore, the main period of P waves is usually long. In non-natural earthquakes, energy is released instantaneously in a very short time and a very small space. The high-frequency components excited by this point source explosion are extremely abundant. Therefore, the main period of P waves is usually short.

[0033] The P-wave low-frequency energy ratio refers to the ratio of the energy carried by the P-wave low-frequency vibration in the total energy of the entire seismic wave in a seismic event. Different P-wave low-frequency energy ratios correspond to different earthquake types. The analytical method for the proportion of P-wave low-frequency energy is as follows: Starting with the arrival time of the P-wave, the analysis time is marked after extending forward by a preset interception duration. The period between the start time and the analysis time is recorded as the analysis window, and the waveform data of the acceleration values ​​within the analysis window are captured and recorded as the acceleration window signal. The preset interception duration is a value used to limit the duration of the analysis window to ensure that the duration of the analysis window remains within a reasonable range. Specifically, the preset interception duration is usually obtained by averaging historical data. For example, the preset interception duration is 4 seconds. A short-time Fourier transform is performed on the acceleration window signal, and a Hanning window is used as the sliding window function to suppress spectral leakage. Set the window length (e.g., 256 sampling points) and the overlap step size (e.g., 50% of the window length), slide along the time axis frame by frame, calculate the local spectrum of all frames, and stack the local spectra of all frames in chronological order to generate a time-frequency spectrum; where the horizontal axis of the time-frequency spectrum is time, the vertical axis is frequency, and the square of the complex modulus represents the energy density at that time-frequency point. The effective response frequency band and sampling frequency of the accelerometer are queried from the technical parameter table. The lower limit of the analysis frequency is set with one-quarter of the effective response frequency band as the first standard, and the upper limit of the analysis frequency is set with half of the sampling frequency as the second standard. The intermediate frequency analysis frequency is also customized. Using the lower limit to the upper limit of the analysis frequency as the first constraint and the analysis window as the second constraint, the time spectrum is double-integrated to calculate the total energy of the entire frequency band. The formula for calculating the total energy across the entire frequency band is: ; In the formula, Total energy across the entire frequency band. To analyze the lower limit of frequency, To analyze the upper limit of frequency, At the starting time, For the analysis time, The energy density at the time-frequency point; Using the lower limit of the analysis frequency to the mid-frequency analysis frequency as the low-frequency range, the total energy of the low-frequency band is calculated, and the total energy of the low-frequency band is divided by the total energy of the entire frequency band to calculate the energy ratio of the P-wave low-frequency band. The formula for calculating the proportion of P-wave low-frequency energy is: ; In the formula, The proportion of P-wave low-frequency energy. This is the intermediate frequency analysis frequency.

[0034] The S-wave low-frequency energy ratio refers to the ratio of the energy carried by the S-wave low-frequency vibration in the total energy of the entire seismic wave in a seismic event. Different S-wave low-frequency energy ratios correspond to different earthquake types. In this embodiment, the method for analyzing the energy proportion of the low-frequency band of the S-wave is consistent with the analysis logic of the energy proportion of the low-frequency band of the P-wave. The difference is that in the energy proportion of the low-frequency band of the S-wave, the arrival time of the S-wave is taken as the starting time, and a new preset interception time is extended to mark the analysis time. The time period between the starting time and the analysis time is recorded as the analysis window.

[0035] Earthquake classification models are artificial intelligence models that take type features as input data and earthquake type as output data, thus serving as a tool to determine the specific earthquake type corresponding to an earthquake event. Earthquake type indicates whether an earthquake event is caused by natural geological structures or artificial conditions. Earthquake types include natural earthquakes and non-natural earthquakes.

[0036] Earthquake classification models do not exist directly in existing technologies; they are obtained by training a dual-branch deep learning model with a large amount of data. Specifically, when training the earthquake classification model, multiple labeled earthquake type features are extracted from the historical database as training samples. Specifically, the label for natural earthquakes is 00, and the label for non-natural earthquakes is 11. The type features are used as input data, and the corresponding earthquake type labels are used as output data to construct a training dataset. The training dataset is then input into a support vector machine classifier for iterative training until the classification accuracy reaches a preset accuracy threshold, thus obtaining the earthquake type classification model.

[0037] In this embodiment, when determining the earthquake type of an earthquake event, the parsed type features are input into the earthquake classification model, and the corresponding earthquake type label is output. When the output earthquake type label is 00, the earthquake type of the earthquake event is a natural earthquake; When the output earthquake type label is 11, the earthquake event is a non-natural earthquake.

[0038] The intelligent distribution determination module extracts the magnitude regional characteristics of earthquake events, including the epicenter level and epicenter region affiliation, and determines the distribution nature of earthquake events based on the magnitude regional characteristics, and determines whether to perform intelligent distribution operation. Not all monitored earthquake events require information distribution. For smaller earthquakes, the actual impact is limited, and over-distribution can lead to information redundancy and resource waste. Therefore, the necessity of distribution should be determined comprehensively based on the earthquake's magnitude, the region where the epicenter is located, and the extent of its impact.

[0039] Regional magnitude characteristics are used to represent the specific changes in earthquakes corresponding to seismic events and serve as a direct basis for determining whether to distribute the earthquake event as necessary. Regional characteristics of earthquake magnitude include epicenter magnitude and epicenter regional affiliation.

[0040] The epicenter magnitude refers to the maximum amplitude of a seismic event and is used to provide a numerical reference for the negative impact of an earthquake on buildings and other facilities. The epicenter attribution refers to the relationship between the location of an earthquake event and the target area. Specifically, the epicenter attribution includes within the province, neighboring provinces, and within the country; respectively, it means that the earthquake epicenter and the target area belong to the same province, the earthquake epicenter is located in a neighboring province of the target area, and the earthquake epicenter is located in another province within the target area.

[0041] In this embodiment, the epicenter level and epicenter region affiliation are obtained by querying multi-source raw data of the earthquake event.

[0042] Distribution properties are used to indicate whether an earthquake event needs to be distributed to the outside world, thereby judging the negative impact of the earthquake event on the target area, and serving as a prerequisite for subsequent collection and summarization of relevant earthquake information; Specifically, the nature of distribution includes necessary distribution and unnecessary distribution.

[0043] The method for determining the distribution properties is as follows: Necessary analyses were conducted on the epicenter magnitude and epicenter region attribution of the earthquake events. When the epicenter area belongs to the same province and the epicenter magnitude is greater than 0, it means that the epicenter of the earthquake event is located in the same province as the target area. In this case, both the epicenter area and the epicenter magnitude are recorded as necessary features. When the epicenter area belongs to a neighboring province and the epicenter level is greater than or equal to the preset first level threshold, it means that the epicenter of the earthquake event is located in a neighboring province of the target area and the epicenter level meets the requirements for necessary distribution. In this case, both the epicenter area and the epicenter level are recorded as necessary features. The preset first level threshold is the minimum epicenter level required to determine that the earthquake event in the neighboring province is necessary for distribution, which can provide a threshold for the epicenter level recorded as a necessary feature. For example, the epicenter level is 3.0. When the epicenter region is located within China and the epicenter magnitude is greater than or equal to the preset second-level threshold, it indicates that the epicenter of the earthquake event is located in another province within China, and the epicenter magnitude meets the requirements for necessary distribution. In this case, both the epicenter region and the epicenter magnitude are recorded as necessary features. The preset second-level threshold refers to the minimum epicenter magnitude that meets the requirements for necessary distribution when the epicenter is located in another province within China. This threshold provides a threshold for the epicenter magnitude that is recorded as a necessary feature. For example, the epicenter magnitude is 4.0. The number of necessary features in an earthquake event is counted. If the number of necessary features is 2, the necessary property of the earthquake event is necessary distribution; if the number of necessary features is less than 2, the necessary property of the earthquake event is unnecessary distribution.

[0044] Intelligent distribution operation refers to the specific operations for distributing earthquake information related to earthquake events based on the nature of the distribution, and can serve as a guide for subsequent operations when necessary distribution is required. In this embodiment, when determining whether to perform the intelligent distribution operation, if the necessary nature of the earthquake event is necessary distribution and subsequent operations need to be performed, then the intelligent distribution operation is determined to be performed; if the necessary nature of the earthquake event is unnecessary distribution and subsequent operations do not need to be performed, then the intelligent distribution operation is determined not to be performed.

[0045] By quantifying the epicenter level, subsequent earthquake event distribution decisions become more precise and flexible, ensuring that differentiated distribution strategies can be implemented based on the specific circumstances of different earthquake events.

[0046] The earthquake event distribution module constructs earthquake information based on source parameters, fills the earthquake information into the event template to generate distribution events, and arranges the distribution events into distribution queues based on event distribution criteria when the distribution center is in a healthy state, and intelligently distributes the distribution events in the distribution queues. Once it is determined that intelligent distribution is required, the relevant information of the earthquake event needs to be organized into standardized distribution events to facilitate subsequent unified distribution.

[0047] Earthquake information is used to summarize the information data that needs to be distributed in earthquake events that require intelligent distribution, thereby providing a comprehensive summary representation of earthquake events from multiple dimensions. Source parameters are parameters used to represent the location parameters and geological parameters of an earthquake event. Specifically, the source parameters include the time of occurrence, the name of the epicenter, the coordinates of the epicenter, the depth of the epicenter, and the distance from the epicenter.

[0048] The time of occurrence of an earthquake refers to the first moment when an earthquake event is detected; the epicenter name refers to the specific province name corresponding to the earthquake event; the epicenter coordinates refer to the latitude and longitude of the earthquake event; the focal depth refers to the depth of the seismic impact caused by the earthquake event; and the epicenter distance refers to the minimum straight-line distance between the center of the earthquake event and the administrative boundary of the target area.

[0049] The focal depth varies depending on the type of earthquake. When the earthquake is a natural earthquake, a focal depth exists; when the earthquake is a non-natural earthquake, a focal depth does not exist. In this embodiment, the epicenter distance is calculated using the Distaz algorithm. If the epicenter coordinates are located in the open sea and the epicenter distance exceeds a preset distance threshold, the epicenter distance is not calculated, and therefore this specific parameter does not exist. The preset distance threshold refers to the maximum epicenter distance that would negatively impact the target area, thus providing a distance-dimensional constraint on the location of earthquake events.

[0050] Earthquake information is a summary of information obtained by aggregating focal parameters, earthquake type, epicenter magnitude, and epicenter region. Earthquake information serves as the core data for earthquake early warning and emergency response. Only by fully and accurately including this information can sufficient basis be provided for subsequent emergency decision-making.

[0051] The distribution event is a task that can summarize the earthquake information of earthquake events and provide it to users for distribution. Distribution events are not generated out of thin air. The earthquake information needs to be filled into the corresponding content of the event template to obtain the required distribution event. The event template is a dynamic structured data template. The event template contains several information fill positions, and each information fill position corresponds to a specific earthquake information. The advantages of using dynamic templates are: first, they ensure that earthquake events of different epicenter magnitudes and earthquake types can obtain appropriate information presentation formats; second, they avoid information redundancy, and for earthquake events with smaller epicenter magnitudes, information fields can be simplified, improving the efficiency of intelligent distribution.

[0052] In this embodiment, when generating a distribution event, the epicenter magnitude and earthquake type are used as the filtering criteria to select the corresponding event template from the template database. According to the arrangement order of the information filling positions in the event template, the earthquake type, epicenter area affiliation, earthquake time, epicenter place name, epicenter coordinates, epicenter magnitude, focal depth, and epicenter distance are sequentially filled into the corresponding information filling positions to generate a complete distribution event. If the focal depth does not exist, the filling position corresponding to the focal depth is removed.

[0053] Once the distribution event is constructed, it can be sent to the distribution center and await further intelligent distribution processing.

[0054] The distribution center is the core hub for the intelligent distribution of events. Its health status directly affects the reliability and timeliness of the distribution of events. Therefore, before distribution, it is necessary to first confirm that the distribution center is in a healthy state. The method for determining the health status of a distribution center is as follows: Based on a preset detection cycle, a heartbeat detection signal is periodically sent to the distribution center, and the timestamp of the distribution center's response to the heartbeat detection signal is recorded. The preset detection cycle refers to the interval between two consecutive heartbeat detection signals, which can be used as an example of the time span for the distribution center to determine its health status. The preset detection cycle is 10 seconds. The heartbeat detection signal is a small probe data packet that is periodically sent to confirm whether the distribution center is still operating normally. Get the timestamp of the distribution center at the current moment, and record the duration between the current timestamp and the timestamp of the most recent heartbeat detection signal response as the heartbeat interval value; When the heartbeat interval value is less than or equal to the preset heartbeat timeout threshold, the distribution center is determined to be in a healthy state; When the heartbeat interval value exceeds the preset heartbeat timeout threshold, the distribution center is determined to be in an unhealthy state. At this time, the earthquake network center will switch the distribution event to the backup distribution channel to ensure the continuous availability of the intelligent distribution service. This mechanism of flexibly switching between the primary distribution channel and the backup distribution channel ensures the availability of the intelligent distribution service for the distribution event.

[0055] It should be noted that the preset heartbeat timeout threshold is the specific duration of the longest allowed silence period or response deadline set by the distribution center, and serves as the maximum value of the heartbeat interval when the distribution center is in a healthy state.

[0056] After confirming that the distribution center is in a healthy state, it is necessary to sort the events according to the event distribution criteria to obtain the distribution queue, and then realize the intelligent and orderly distribution of the events based on the sorting result of the distribution queue. Event distribution criteria are used to define the order in which multiple events are distributed, ensuring that multiple events are distributed in an orderly manner according to their severity. The event distribution criteria are as follows: mandatory distribution is the primary criterion, and epicenter level is the secondary criterion. The method for intelligent event distribution is as follows: The mandatory state of each dispatch event is retrieved one by one. Dispatch events with a mandatory state of mandatory dispatch are arranged before those with a non-mandatory state of mandatory dispatch to obtain the basic queue. The mandatory state is used to indicate whether a dispatch event is in a mandatory dispatch state. The mandatory state includes mandatory dispatch and non-mandatory dispatch. The events in the base queue are arranged sequentially according to the epicenter level from largest to smallest to generate a distribution queue, and each distribution event is marked with a unique event ID. Following the order of the distribution queue from front to back, the first distribution event is recorded as the current event, and the earthquake type, epicenter region affiliation, and epicenter magnitude of the current event are retrieved. If the current earthquake is a natural earthquake and the epicenter is located within the province, the earthquake information of the current event will be sent to all user groups simultaneously. If the current earthquake type is a natural earthquake, the epicenter area belongs to a neighboring province, and the epicenter level is greater than or equal to the preset first level threshold, the earthquake information of the current event will be sent to the first user group and the second user group. If the current earthquake is a natural earthquake, the epicenter is located within China, and the epicenter level is greater than or equal to the preset second-level threshold, the earthquake information of the current event will be sent to the second user group and the third user group. If the earthquake type of the current event is a non-natural earthquake, and the epicenter level is greater than or equal to the preset second level threshold, the earthquake information of the current event will be sent to the second user group. Iterate through all dispatch events in the dispatch queue until all dispatch events have been intelligently dispatched.

[0057] User groups are groups of personnel used to receive earthquake information about earthquake events. User groups include a first user group, a second user group, and a third user group. Specifically, the first user group consists of members of the leadership team, the second user group consists of members of the duty team, and the third user group consists of members of the emergency response team.

[0058] It should be noted that during intelligent distribution, it is also necessary to verify whether the event ID of the currently distributed event has been distributed recently. If it has been distributed, the distribution event is skipped to prevent the same earthquake event from being triggered multiple times and causing duplicate transmission.

[0059] For distribution events under mandatory conditions, the distribution center will initiate a voice broadcast warning process in parallel with the intelligent distribution, broadcasting the distribution event directly via audio equipment connected to the earthquake network center to ensure that the duty team can receive and understand the earthquake event in a timely manner.

[0060] The event log summary module constructs event logs for earthquake events and imports these logs into a historical database for data storage and updates. The event record of an earthquake event is not only a summary of the handling of this earthquake event, but also an important data foundation for subsequent earthquake research and optimization of early warning strategies.

[0061] The event log includes earthquake information, event ID, distribution time, user grouping, and distribution result.

[0062] Distribution time refers to the timestamp when an earthquake event is intelligently distributed to user groups.

[0063] Distribution results refer to the specific results of the intelligent distribution of earthquake events to user groups, including successful distribution and distribution failure.

[0064] Historical databases are used to record and store information related to all earthquake events, providing a reasonable and reliable recording and storage platform for earthquake information processing and intelligent distribution. By importing event records into the historical database through incremental updates and intelligent archiving mechanisms, the historical database can continuously accumulate data on earthquake events while ensuring the traceability and retrieval of the data. This provides complete, traceable, and multi-dimensional data support for subsequent optimization of earthquake early warning strategies and earthquake scientific research in target areas.

[0065] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An event-driven real-time seismic information processing and intelligent distribution system, applied to a seismic network center, characterized in that: include: The earthquake event monitoring module is used to extract data fluctuation characteristics from multi-source raw data of the target area. The data fluctuation characteristics include fluctuation coefficients and standard magnitude values. Based on the data fluctuation characteristics, it determines whether an earthquake event has occurred in the target area. The earthquake type determination module is used to parse type features from the phase files of earthquake events and input the type features into the earthquake classification model to determine the earthquake type of the earthquake event. The intelligent distribution determination module is used to extract the magnitude regional characteristics of earthquake events, including the epicenter level and epicenter region affiliation, and to determine the distribution nature of earthquake events based on the magnitude regional characteristics, and to determine whether to perform intelligent distribution operation. Distribution characteristics include necessary distribution and unnecessary distribution; The method for determining whether to perform intelligent distribution is as follows: When the epicenter area belongs to the province and the epicenter level is greater than 0, both the epicenter area and the epicenter level are recorded as necessary features. When the epicenter area belongs to a neighboring province and the epicenter level is greater than or equal to the preset first level threshold, both the epicenter area affiliation and the epicenter level are recorded as necessary features. When the epicenter area is classified as domestic and the epicenter level is greater than or equal to the preset second-level threshold, both the epicenter area classification and the epicenter level are recorded as necessary features. The number of necessary features in an earthquake event is counted. If the number of necessary features is 2, the necessary property is determined to be necessary distribution, and the intelligent distribution operation is executed. If the number of necessary features is less than 2, the necessary properties are deemed unnecessary to distribute, and the intelligent distribution operation is not performed. The earthquake event distribution module is used to construct earthquake information based on the source parameters, fill the earthquake information into the event template to generate distribution events, and arrange the distribution events into distribution queues based on the event distribution criteria when the distribution center is in a healthy state, and intelligently distribute the distribution events in the distribution queues. The event log summary module is used to build event logs for earthquake events and import the event logs into a historical database for data storage and updates.

2. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 1, characterized in that, The wave coefficient and the standard value of magnitude are extracted as follows: Using the current time as the first sampling point, and with a preset sampling step size as the interval, A sampling points with interval distribution are deduced by going backward along the time line; Collect multi-source raw data from A sampling points one by one, mark the fields of the multi-source raw data, and arrange the raw values ​​of the multi-source raw data with the same fields into a data sequence; Starting from the second sampling point, the original values ​​of two adjacent sampling points are used as comparison objects, and the fluctuation coefficient of the data sequence is calculated by the fluctuation calculation function; The multi-source raw data from C monitoring stations were analyzed sequentially, and the raw values ​​of magnitude from the same sampling point were extracted to obtain C magnitude values. The average magnitude is calculated by summing the C magnitude values. The average magnitude and the C magnitude values ​​are then substituted into the standard deviation formula to calculate the standard magnitude.

3. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 2, characterized in that, The method for determining whether an earthquake event has occurred is as follows: The number of data sequences whose volatility coefficients exceed a preset volatility threshold is counted to obtain the volatility exceedance amount; When the fluctuation threshold is greater than or equal to 1 and the magnitude standard value is greater than the preset magnitude standard threshold, it is determined that an earthquake event has occurred in the target area. When the fluctuation threshold is less than 1, or the magnitude standard value is less than or equal to the preset magnitude standard threshold, it is determined that no earthquake event has occurred in the target area.

4. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 3, characterized in that, Type characteristics include waveform arrival time difference, P-wave amplitude ratio, P-wave main period, S-wave main period, P-wave low-frequency energy ratio, and S-wave low-frequency energy ratio.

5. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 4, characterized in that, The analytical method for the proportion of P-wave low-frequency energy is as follows: The analysis time is marked after extending the P-wave arrival time by a preset interception period. The time period between the start time and the analysis time is recorded as the analysis window, and the waveform data of the acceleration value within the analysis window is recorded as the acceleration window signal. A short-time Fourier transform is performed on the acceleration window signal, using the Hanning window as the sliding window function; Set the window length and overlap step, slide along the time axis frame by frame, calculate the local spectrum of all frames, and stack the local spectra of all frames in chronological order to generate a time-spectrum graph; The effective response frequency band and sampling frequency of the accelerometer are queried from the technical parameter table. The lower limit of the analysis frequency is set with one-quarter of the effective response frequency band as the first standard, and the upper limit of the analysis frequency is set with half of the sampling frequency as the second standard. The intermediate frequency analysis frequency is also customized. Using the lower limit to the upper limit of the analysis frequency as the first constraint and the analysis window as the second constraint, the time spectrum is double-integrated to calculate the total energy of the entire frequency band. Using the lower limit of the analysis frequency to the mid-frequency analysis frequency as the low-frequency range, the total energy of the low-frequency band is calculated, and the total energy of the low-frequency band is divided by the total energy of the entire frequency band to calculate the energy ratio of the P-wave low-frequency band.

6. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 5, characterized in that, Earthquake types include natural earthquakes and non-natural earthquakes; Label natural earthquakes as 00 and non-natural earthquakes as 11; When the output earthquake type label is 00, the earthquake event is a natural earthquake; when the output earthquake type label is 11, the earthquake event is a non-natural earthquake.

7. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 6, characterized in that, The focal parameters include the time of occurrence, epicenter location, epicenter coordinates, focal depth, and epicenter distance; earthquake information is a summary of focal parameters, earthquake type, epicenter magnitude, and epicenter regional affiliation.

8. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 7, characterized in that, The method for determining the health status of a distribution center is as follows: Based on a preset detection cycle, a heartbeat detection signal is periodically sent to the distribution center, and the timestamp of the distribution center's response to the heartbeat detection signal is recorded. Get the timestamp of the distribution center at the current moment, and record the duration between the current timestamp and the timestamp of the most recent heartbeat detection signal response as the heartbeat interval value; When the heartbeat interval value is less than or equal to the preset heartbeat timeout threshold, the distribution center is determined to be in a healthy state; When the heartbeat interval value is greater than the preset heartbeat timeout threshold, the distribution center is determined to be in an unhealthy state.

9. The event-driven real-time earthquake information processing and intelligent distribution system according to claim 8, characterized in that, The event distribution criteria are as follows: mandatory distribution is the primary criterion, and epicenter level is the secondary criterion. The method for intelligent event distribution is as follows: Query the mandatory status of all dispatch events one by one, and arrange the dispatch events with the mandatory status of mandatory dispatch before the dispatch events with the mandatory status of non-mandatory dispatch to obtain the basic queue; The distribution events in the basic queue are arranged in descending order of epicenter level to generate a distribution queue. Each distribution event is marked with a unique event ID. The first distribution event in the distribution queue is recorded as the current event. If the current earthquake is a natural earthquake and the epicenter is located within the province, the earthquake information of the current event will be sent to all user groups simultaneously. If the current earthquake type is a natural earthquake, the epicenter area belongs to a neighboring province, and the epicenter level is greater than or equal to the preset first level threshold, the earthquake information of the current event will be sent to the first user group and the second user group. If the current earthquake is a natural earthquake, the epicenter is located within China, and the epicenter level is greater than or equal to the preset second-level threshold, the earthquake information of the current event will be sent to the second user group and the third user group. If the earthquake type of the current event is a non-natural earthquake, and the epicenter level is greater than or equal to the preset second level threshold, the earthquake information of the current event will be sent to the second user group. Iterate through all dispatch events in the dispatch queue until all dispatch events have been intelligently dispatched.

Citation Information

Patent Citations

  • Earthquake early warning information grading automatic pushing system and method based on region division

    CN121838425A

  • Distributed earthquake early warning cloud monitoring network system and method

    CN119479204A