Emergency broadcast intelligent triggering system based on multi-source seismic data fusion

Abstract

The invention provides an emergency broadcast intelligent triggering system based on multi-source seismic data fusion, and relates to the technical field of seismic monitoring and early warning, and the system comprises a data receiving module which is used for receiving multi-source seismic early warning information in real time, including seismic source parameter data, strong vibration observation data and real-time early warning parameters, and transmitting the multi-source seismic early warning information through a built-in communication protocol; and outputting the standardized multi-source seismic data. And the localization calculation module is used for carrying out real-time analysis on the standardized multi-source seismic data, extracting a P-wave initial parameter, a seismic motion acceleration predicted value and a seismic source fracture feature, carrying out dynamic calibration on an analysis result by combining actual measurement data of a built-in intensity meter, and outputting a multi-dimensional seismic data stream verified by the intensity meter. Through multi-source earthquake data fusion, real-time analysis and calibration, dynamic risk modeling and intelligent broadcast scheduling, generation and efficient propagation of earthquake early warning information are realized, emergency response timeliness and pertinence are improved, and earthquake disaster risks are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of earthquake monitoring and early warning, and particularly to an intelligent trigger system for emergency broadcasting based on multi-source earthquake data fusion. Background Art

[0002] The existing technology faces serious limitations due to the contradiction between static trigger mechanisms and dynamic earthquake evolution. Traditional systems rely on pre-set fixed trigger conditions (such as a single magnitude or intensity threshold) and cannot dynamically adjust according to the real-time changes in the earthquake rupture process, leading to the following key problems:

[0003] During the earthquake rupture process, source parameters (such as magnitude, epicenter location, rupture direction) may change as the rupture propagates. For example, in the initial stage of an earthquake, due to limited station data, the system may underestimate the magnitude. If the subsequent rupture expansion causes a significant increase in the actual magnitude, the traditional system, unable to dynamically correct the trigger threshold, still uses the initial low threshold condition, resulting in high-risk areas not being covered in a timely manner; conversely, if the magnitude is misjudged too high in the initial stage, the system will continuously send redundant early warnings to low-risk areas, causing a decline in public trust.

[0004] The hazard avoidance prompts generated by existing broadcast systems are mostly fixed templates and do not dynamically adapt in combination with local building structure characteristics, population density distribution, and real-time intensity data. For example, the earthquake avoidance strategies for high-rise buildings and low-rise brick-concrete structures are significantly different, but it is impossible to automatically match the differentiated guidance content through AI technology. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent trigger system for emergency broadcasting based on multi-source earthquake data fusion, which can automatically generate hazard avoidance prompts adapted to the target area and improve the information reach efficiency.

[0006] To solve the above technical problems, the technical solution of the present invention is as follows:

[0007] In a first aspect, an intelligent trigger system for emergency broadcasting based on multi-source earthquake data fusion includes:

[0008] A data receiving module, configured to receive multi-source earthquake early warning information in real time, including source parameter data, strong motion observation data, and real-time early warning parameters, and output standardized multi-source earthquake data through a built-in communication protocol;

[0009] A localization calculation module, configured to perform real-time analysis on the standardized multi-source earthquake data, extract P-wave initial parameters, predicted values of ground motion acceleration, and source rupture characteristics, dynamically calibrate the analysis results in combination with the measured data of a built-in seismometer, and output a multi-dimensional earthquake data stream verified by the seismometer;

[0010] A data processing module, which is used to input the multi-dimensional seismic data stream verified by the seismometer into a preset geographic grid coordinate system, generate a dynamically updated magnitude-epicenter location probability distribution matrix by jointly analyzing P-wave parameters, predicted values of ground motion acceleration, and source rupture characteristics, and calculate a dynamic trigger threshold to generate a hierarchical early warning trigger instruction;

[0011] An emergency broadcast module, which is used to match the earthquake avoidance guidance strategy for the target area according to the hierarchical early warning trigger instruction and the emergency response pre-plan library of the preset AI chip, and generate multimedia early warning information including earthquake countdown parameters, predicted intensity distribution, and AI-optimized hazard avoidance tips;

[0012] A collaborative broadcast control module, which is used to distribute the multimedia early warning information to voice broadcast terminals, digital display terminals, and public information screens through the IP network protocol, realize the priority transmission of voice broadcasts and the hierarchical rendering of dynamic information, and at the same time, based on the real-time feedback of terminal response data, optimize the pre-plan matching logic of the AI chip and the determination conditions of the dynamic trigger threshold.

[0013] Furthermore, perform real-time analysis on the standardized multi-source seismic data, extract the initial P-wave parameters, predicted values of ground motion acceleration, and source rupture characteristics, dynamically calibrate the analysis results in combination with the measured data of the built-in seismometer, and output a multi-dimensional seismic data stream verified by the seismometer, including:

[0014] Perform real-time scanning on the seismic waveform data through the sliding time window analysis method, and identify the initial moment of the P-wave and extract the initial P-wave parameters according to the preset waveform slope threshold and energy change rate threshold;

[0015] Based on the three-dimensional coordinates of the earthquake source in the initial P-wave parameters and the spatial distribution of geographic grid units, and combining the physical laws of ground motion attenuation and historical statistical relationships, calculate the basic predicted value of ground motion acceleration for each grid unit;

[0016] Obtain the measured data of real-time ground motion acceleration through the seismometer built in the localization calculation module, dynamically compare the basic predicted value with the measured data, and if the deviation exceeds the preset error threshold, then correct the magnitude-distance attenuation parameter and the geological medium correction coefficient to generate a calibrated predicted value interval of ground motion acceleration;

[0017] According to the spatial distribution characteristics of the calibrated predicted value interval of ground motion acceleration, and combining the temporal variation law of the initial P-wave parameters, extract the parameters of the source rupture direction, rupture speed, and rupture length;

[0018] Fuse the calibrated predicted value interval of ground motion acceleration, the initial P-wave parameters, and the source rupture characteristics, add the seismometer verification mark, and generate a multi-dimensional seismic data stream.

[0019] Further, based on the three-dimensional source coordinates in the initial P-wave parameters and the spatial distribution of geographical grid cells, combined with the physical laws of ground motion attenuation and historical statistical relationships, calculate the basic predicted values of ground motion acceleration for each grid cell, including:

[0020] According to the spatial distance between the three-dimensional source coordinates and the center point of the geographical grid cell, calculate the length of the seismic wave propagation path, and combined with the distribution data of the geological medium type corresponding to the path, extract the geological density and elastic modulus parameters;

[0021] Based on the geological density and elastic modulus parameters, combined with the physical laws of ground motion attenuation, construct a magnitude-distance attenuation relationship model, and the magnitude-distance attenuation relationship model defines the logarithmic attenuation characteristics of the acceleration attenuation reference values at different magnitudes with the propagation path length;

[0022] According to the historical ground motion statistical data set, statistically calibrate the attenuation reference values to generate a mapping table of acceleration attenuation reference values under different magnitude and distance combinations;

[0023] Bind the mapping table of acceleration attenuation reference values to the spatial distribution of the geographical grid coordinate system to generate the basic predicted values of ground motion acceleration for each grid cell.

[0024] Further, obtain the measured data of real-time ground motion acceleration through the accelerometer built in the localization calculation module, dynamically compare the basic predicted values with the measured data, and if the deviation exceeds the preset error threshold, correct the magnitude-distance attenuation parameters and the geological medium correction coefficient to generate a calibrated predicted value interval of ground motion acceleration, including:

[0025] Based on the spatial distribution of geographical grid cells, map the measured data of real-time ground motion acceleration to the corresponding grid cells;

[0026] Compare the basic predicted values of ground motion acceleration for each grid cell with the measured data of the same grid cell point by point, calculate the prediction deviation rate, and if the prediction deviation rate exceeds the preset error threshold, adjust the magnitude-distance attenuation parameters and the geological medium correction coefficient according to the deviation direction to obtain the corrected attenuation parameters and correction coefficient;

[0027] Based on the corrected attenuation parameters and correction coefficient, recalculate the predicted values of ground motion acceleration to generate a calibrated predicted value interval including the confidence interval range.

[0028] Further, input the multi-dimensional seismic data stream verified by the accelerometer into the preset geographical grid coordinate system, and generate a dynamically updated magnitude-epicenter position probability distribution matrix by jointly analyzing the P-wave parameters, the predicted values of ground motion acceleration and the source rupture characteristics, and calculate the dynamic trigger threshold to generate a hierarchical early warning trigger instruction, including:

[0029] Based on the geographic grid coordinate system, calculate the seismic wave propagation path distance from the earthquake source to each grid cell, and combine the geological medium distribution parameters corresponding to the path to extract geological density and elastic modulus data;

[0030] According to the calibrated range of predicted ground motion acceleration values, and combining the earthquake source rupture direction and propagation speed parameters, construct the spatio-temporal constraint conditions for magnitude and epicenter location. The spatio-temporal constraint conditions include the magnitude upper limit, rupture propagation range, and time window;

[0031] Based on the geological parameters and spatio-temporal constraint conditions, through the Monte Carlo sampling method, iteratively fuse the earthquake source probability distribution, prior probability of ground motion intensity, and spatio-temporal constraint conditions to update the joint probability density of magnitude-epicenter location for each grid cell, and generate a dynamically updated magnitude-epicenter location probability distribution matrix;

[0032] From the dynamically updated magnitude-epicenter location probability distribution matrix, extract the set of grid cells where the magnitude probability exceeds the preset intensity threshold, and mark them as high-probability regions;

[0033] According to the population density level and building seismic resistance level parameters corresponding to the high-probability regions in the external database, calculate the regional comprehensive risk coefficient;

[0034] Real-time monitor the dynamic trigger threshold and the growth rate of magnitude probability. When the trigger threshold reaches the preset critical value and the growth rate exceeds the growth rate threshold, generate a hierarchical warning trigger instruction including the target area code and warning level.

[0035] Furthermore, based on the geological parameters and spatio-temporal constraint conditions, through the Monte Carlo sampling method, iteratively fuse the earthquake source probability distribution, prior probability of ground motion intensity, and spatio-temporal constraint conditions to update the joint probability density of magnitude-epicenter location for each grid cell, and generate a dynamically updated magnitude-epicenter location probability distribution matrix, including:

[0036] According to the magnitude upper limit and rupture propagation range in the spatio-temporal constraint conditions, define the parameter random generation interval for Monte Carlo sampling;

[0037] Perform random sampling within the parameter random generation interval to generate multiple groups of candidate parameter combinations of magnitude, epicenter location, and ground motion intensity;

[0038] For each group of candidate parameter combinations, calculate the probability value in the earthquake source probability distribution, the ground motion intensity probability value, and the compliance weight under the spatio-temporal constraint conditions;

[0039] Fuse the earthquake source probability value, the ground motion intensity probability value, and the compliance weight to obtain the joint probability density corresponding to the parameter combination;

[0040] Statistically analyze the joint probability density of all candidate parameter combinations, calculate the mean probability density within each grid cell, map the mean to the corresponding grid cell in the geographic grid coordinate system, update the values in the magnitude-epicenter location probability distribution matrix, and when the number of Monte Carlo sampling reaches the preset threshold, output the updated magnitude-epicenter location probability distribution matrix.

[0041] Furthermore, according to the graded warning trigger instruction and the emergency response plan library of the preset AI chip, match the earthquake avoidance guidance strategy for the target area, and generate multimedia warning information including earthquake countdown parameters, predicted intensity distribution, and AI-optimized risk avoidance tips, including:

[0042] According to the warning level in the graded warning trigger instruction, match the corresponding response strategy template from the emergency response plan library to determine the priority, language style, and geographic coverage of the broadcast content;

[0043] Based on the seismic wave propagation speed parameter and the updated magnitude-epicenter location probability distribution matrix, calculate the propagation time of the seismic wave reaching each grid cell within the geographic coverage area, and generate regionalized countdown parameters synchronized with the time axis;

[0044] Spatially correlate and match the predicted intensity distribution data of the high-probability area with the road network topology and shelter coordinates in the external database to generate intensity distribution spatial data superimposed with road traffic status and shelter resource identifiers;

[0045] According to the building structure feature database of the target area, analyze the building type and seismic resistance capacity through the AI chip, and combine the predicted intensity level in the intensity distribution spatial data to dynamically generate earthquake avoidance action guidance content adapted to the building type;

[0046] Synchronously fuse the regionalized countdown parameters, intensity distribution spatial data, and earthquake avoidance action guidance content according to the time stamp and spatial position to generate multimedia warning information including voice broadcast, text prompt, and dynamic visualization elements.

[0047] Furthermore, spatially correlate and match the predicted intensity distribution data of the high-probability area with the road network topology and shelter coordinates in the external database to generate intensity distribution spatial data superimposed with road traffic status and shelter resource identifiers, including:

[0048] Based on the geographic grid coordinate system, align the predicted intensity distribution data, road network topology, and shelter coordinates of the high-probability area in terms of spatial position to generate an associated dataset under a unified coordinate system;

[0049] Based on the road network topology in the associated dataset and combined with real-time traffic data, analyze the traffic states of each road segment in high-probability areas, and generate road traffic state data including road grades, congestion levels, and availability indicators; based on the coordinates of shelters in the associated dataset and combined with the capacity database, mark the real-time available capacity of each shelter, and generate shelter resource availability indicator data;

[0050] Fuse the intensity distribution prediction data, road traffic state data, and shelter resource availability indicator data according to the spatial positions of geographical grid cells to generate intensity distribution spatial data with overlaid road traffic states and shelter resource indicators.

[0051] Furthermore, distribute the multimedia warning information to voice broadcast terminals, digital display terminals, and public information screens through the IP network protocol, realize the priority transmission of voice broadcasts and the hierarchical rendering of dynamic information. At the same time, based on the real-time feedback of terminal response data, optimize the scenario matching logic and dynamic trigger threshold determination conditions of the AI chip, including:

[0052] Assign transmission priorities to voice broadcast data through the IP network protocol, and dynamically adjust the rendering levels of text prompts and dynamic information elements based on the transmission status of voice broadcasts;

[0053] Receive the broadcast completion status of voice broadcast terminals, the rendering delay time of digital display terminals, and the network load data of public information screens in real time, and generate a terminal response dataset;

[0054] According to the broadcast delay and rendering efficiency data in the terminal response dataset, adjust the matching weights of emergency response scenarios in the AI chip, optimize the generation logic of earthquake avoidance guidance strategies, and dynamically correct the determination conditions of dynamic trigger thresholds based on the network load fluctuation data in the terminal response dataset.

[0055] In a second aspect, a computer-readable storage medium stores a program that, when executed by a processor, implements the system described above.

[0056] The above solution of the present invention has at least the following beneficial effects:

[0057] Dynamically calibrate the analysis results through the measured data of the built-in seismometer, correct the prediction deviations caused by geological condition differences or propagation path errors (such as reducing the acceleration prediction error by more than 30% through calibration in soft soil layer areas), and ensure the reliability of the output data. By extracting initial P-wave parameters, source rupture characteristics, etc., and combining with the geographical grid coordinate system, generate a dynamically updated magnitude-epicenter position probability distribution matrix, which reflects the spatio-temporal changes of earthquake risks in real time. For example, during the propagation process of the source rupture, the matrix is updated every 2 seconds, winning precious time for emergency response.

[0058] Calculate the comprehensive risk coefficient based on parameters such as regional population density and building seismic resistance level, and dynamically adjust the early warning trigger conditions in combination with the magnitude probability growth rate to reduce the false alarm rate while ensuring that high-risk areas are preferentially triggered for early warning (for example, the trigger threshold in densely populated areas is reduced by 20%). The emergency broadcast module automatically matches the plan template according to the early warning level and generates personalized risk avoidance strategies. For example, in areas with dense masonry structure buildings, it preferentially prompts "Hide in the corner of load-bearing walls"; for high-rise building groups, it emphasizes "Avoid elevators and take shelter nearby" to improve the scenario adaptability of the guidance strategy. Generate dynamic early warning information including earthquake countdown, intensity distribution map, and shelter resource identification, and present it in multiple modalities of voice, text, and visualization layer to meet the needs of different users (for example, people with hearing impairments can obtain early warnings through visual information).

[0059] Assign the highest transmission priority to voice broadcasts through the IP network protocol to ensure that key information such as "earthquake countdown" and "emergency risk avoidance instructions" can still be preferentially delivered in case of network congestion (for example, the voice data transmission delay is controlled within 500 milliseconds). Dynamically adjust the plan matching logic based on terminal feedback data (such as broadcast delay and screen load). For example, automatically simplify the dynamic map rendering in areas with poor network to enhance information accessibility; at the same time, correct the trigger threshold and give early warnings in advance to cope with transmission delays. Brief Description of the Drawings

[0060] Figure 1 It is a schematic diagram of an emergency broadcast intelligent trigger system based on multi-source seismic data fusion provided by an embodiment of the present invention. Detailed Description of the Invention

[0061] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0062] As Figure 1 shown, an embodiment of the present invention proposes an emergency broadcast intelligent trigger system based on multi-source seismic data fusion, including:

[0063] A data receiving module for receiving multi-source earthquake early warning information in real time, including seismic source parameter data, strong motion observation data, and real-time early warning parameters, and outputting standardized multi-source seismic data through a built-in communication protocol;

[0064] The localization calculation module is used to perform real-time parsing on the standardized multi-source seismic data, extract the initial P-wave parameters, predicted values of ground motion acceleration, and source rupture characteristics, dynamically calibrate the parsing results by combining with the measured data of the built-in accelerometer, and output a multi-dimensional seismic data stream verified by the accelerometer;

[0065] The data processing module is used to input the multi-dimensional seismic data stream verified by the accelerometer into a preset geographic grid coordinate system, generate a dynamically updated magnitude-epicenter location probability distribution matrix by jointly analyzing the P-wave parameters, predicted values of ground motion acceleration, and source rupture characteristics, calculate the dynamic trigger threshold, and generate a hierarchical early warning trigger instruction;

[0066] The emergency broadcast module is used to match the earthquake avoidance guidance strategy for the target area according to the hierarchical early warning trigger instruction and the emergency response plan library of the preset AI chip, and generate multimedia early warning information including earthquake countdown parameters, predicted intensity distribution, and AI-optimized hazard avoidance tips;

[0067] The collaborative broadcast control module is used to distribute the multimedia early warning information to voice broadcast terminals, digital display terminals, and public information screens through the IP network protocol, realize the priority transmission of voice broadcasts and the hierarchical rendering of dynamic information, and at the same time optimize the plan matching logic of the AI chip and the determination conditions of the dynamic trigger threshold based on the real-time feedback of terminal response data.

[0068] In the embodiment of the present invention, various types of data such as source parameters and strong motion observations are collected in real time and output in a standardized manner, which avoids the limitations of a single data source and enables the multi-dimensional acquisition of seismic information. For example, by combining source parameters and strong motion observation data, the earthquake intensity and influence range can be more accurately judged. The multi-source data is parsed in real time to extract key parameters, and the measured data of the built-in accelerometer is used for dynamic calibration. For example, in complex geological areas, the prediction deviation caused by geological conditions can be corrected through calibration, ensuring that the output multi-dimensional seismic data stream is true and reliable, providing strong support for accurate early warning.

[0069] By jointly analyzing multi-dimensional data, a dynamically updated magnitude-epicenter location probability distribution matrix is generated, and a dynamic trigger threshold is calculated. This data-depth analysis-based method can comprehensively consider various factors, more scientifically evaluate earthquake risks, and compared with traditional fixed-threshold early warnings, can reduce false alarm and missed alarm rates, making the early warning more in line with actual earthquake situations. According to the classified early warning trigger instructions and the emergency response plan library, combined with the actual situation of the target area, personalized multimedia early warning information is generated. For example, for different building types and population distribution areas, appropriate earthquake avoidance guidance is provided, enabling the public to quickly obtain risk avoidance information suitable for their own situations and improving the effectiveness of risk avoidance measures. Optimize the plan matching and threshold determination according to the terminal response data. For example, in areas with network congestion, prioritize voice broadcasts and adjust the early warning trigger strategy to ensure the efficient transmission of early warning information and enhance the adaptability and reliability of the entire early warning system.

[0070] In a preferred embodiment of the present invention, the standardized multi-source seismic data is analyzed in real time, the initial P-wave parameters, the predicted value of ground motion acceleration, and the source rupture characteristics are extracted, and the analysis results are dynamically calibrated in combination with the measured data of the built-in seismometer, and a multi-dimensional seismic data stream verified by the seismometer is output, which may include:

[0071] The seismic waveform data is scanned in real time through the sliding time window analysis method, and according to the preset waveform slope threshold and energy change rate threshold, the initial moment of the P-wave is identified, and the initial P-wave parameters are extracted;

[0072] Based on the three-dimensional source coordinates in the initial P-wave parameters and the spatial distribution of geographical grid units, combined with the physical laws of ground motion attenuation and historical statistical relationships, the basic predicted value of ground motion acceleration for each grid unit is calculated, specifically including:

[0073] According to the spatial distance between the three-dimensional source coordinates and the center point of the geographical grid unit, the seismic wave propagation path length is calculated, and combined with the geological medium type distribution data corresponding to the path, the geological density and elastic modulus parameters are extracted;

[0074] Based on the geological density and elastic modulus parameters, combined with the physical laws of ground motion attenuation, a magnitude-distance attenuation relationship model is constructed, and the magnitude-distance attenuation relationship model defines the logarithmic attenuation characteristics of the acceleration attenuation reference value at different magnitudes with the propagation path length;

[0075] According to the historical ground motion statistical data set, the attenuation reference value is statistically calibrated to generate an acceleration attenuation reference value mapping table for different magnitude and distance combinations;

[0076] Bind the acceleration attenuation reference value mapping table to the spatial distribution of the geographical grid coordinate system to generate the basic predicted value of ground motion acceleration for each grid unit;

[0077] Through the accelerometer built in the localization calculation module, the measured data of real-time ground motion acceleration is obtained. The basic prediction value is dynamically compared with the measured data. If the deviation exceeds the preset error threshold, the magnitude-distance attenuation parameter and the geological medium correction coefficient are corrected to generate a calibrated prediction value interval of ground motion acceleration, specifically including:

[0078] Based on the spatial distribution of geographical grid cells, the measured data of real-time ground motion acceleration is mapped to the corresponding grid cells;

[0079] The basic prediction value of ground motion acceleration of each grid cell is compared point by point with the measured data of the same grid cell, and the prediction deviation rate is calculated. If the prediction deviation rate exceeds the preset error threshold, the magnitude-distance attenuation parameter and the geological medium correction coefficient are adjusted according to the deviation direction to obtain the corrected attenuation parameter and correction coefficient;

[0080] Based on the corrected attenuation parameter and correction coefficient, the prediction value of ground motion acceleration is recalculated to generate a calibrated prediction value interval including the confidence interval range;

[0081] According to the spatial distribution characteristics of the calibrated prediction value interval of ground motion acceleration, combined with the temporal variation law of the initial P-wave parameters, the parameters of the earthquake source rupture direction, rupture velocity and rupture length are extracted;

[0082] The calibrated prediction value interval of ground motion acceleration, the initial P-wave parameters and the earthquake source rupture characteristics are fused, and after adding the accelerometer verification flag, a multi-dimensional seismic data stream is generated.

[0083] In the embodiment of the present invention, the seismic monitoring point collects data at a frequency of 50 - 200 times per second to form a time series waveform. The sliding time window method is adopted, the window width is set to 1 second, and the sliding step is 0.1 second to scan the waveform data. In each 1-second window, by calculating the amplitude difference between adjacent data points, the waveform slope is obtained. For example, if the amplitude of the first data point in a window is 2 and the amplitude of the 100th data point is 5, the slope calculation is ; at the same time, the waveform amplitudes are squared and summed, and the energy change rate is obtained by comparing the total energy change of adjacent windows. According to historical earthquake data, the waveform slope threshold is set to 0.03 and the energy change rate threshold is set to 1.5. When the waveform slope in a certain window exceeds 0.03 and the energy change rate reaches 1.5, the start time of the window is determined as the initial P-wave time. After determining the initial time, parameters such as the time stamp (accurate to milliseconds), amplitude (such as 3.2 μGal), and frequency components (such as the main frequency 8 Hz) at this time are extracted.

[0084] After obtaining the three-dimensional coordinates of the earthquake source (longitude 110°, latitude 30°, depth 10 km) and the coordinates of the center point of the geographical grid cell (longitude 110.1°, latitude 30.1°, the earth's surface), the calculation of the earthquake wave propagation path length begins. First, the longitude and latitude coordinates need to be converted into three-dimensional rectangular coordinates. This is like converting the "address" on the earth's surface into the "exact coordinates" in three-dimensional space. Taking the center of the earth as the origin, a spatial rectangular coordinate system is established. According to the geographical coordinate conversion rules, the longitude and latitude values of the earthquake source and the center point of the geographical grid cell, as well as the earthquake source depth (the earth's surface depth is regarded as 0), are substituted into the rules for calculation, so as to obtain their corresponding X, Y, and Z coordinate values in the three-dimensional rectangular coordinate system.

[0085] After obtaining the three-dimensional rectangular coordinates, use the three-dimensional space distance formula to calculate the straight-line distance between the earthquake source and the center point of the geographical grid cell in three-dimensional space. This distance is like digging a straight tunnel inside the earth to connect the earthquake source and the grid center point. However, in fact, the earthquake wave propagates along the vicinity of the earth's surface, not straight through the interior of the earth. Therefore, this straight-line distance is not the final propagation path length. Next, according to the average radius of the earth of 6371 km, the straight-line distance calculated just now is corrected. Because the earth is a sphere, the actual shortest path between the earthquake source and the grid center point on the earth's surface is different from the straight-line distance in three-dimensional space. Through the principles and methods of spherical trigonometry, a spherical triangle is formed by the center of the earth, the earthquake source point, and the grid center point. Using the relationships between the sides and angles of this triangle, the straight-line distance is adjusted, and finally the earthquake wave propagation path length that more conforms to the actual situation is obtained, about 14.1 km.

[0086] While calculating the propagation path length, according to the trajectory of the propagation path on the earth's surface and underground, judge the geological regions it crosses. This process is like doing a "geological scan" for the propagation path to determine which types of strata it passes through. When it is judged that the propagation path passes through the granite area, the corresponding geological parameters of this area will be extracted from the geological database, such as the density of granite 2700 kg / m³ and the elastic modulus 60 GPa.

[0087] Construct an attenuation relationship model:

[0088] Based on the geological density and elastic modulus, combined with the earthquake ground motion attenuation law in seismological research, establish a magnitude-distance relationship. For example, set the initial acceleration reference value of a magnitude 6 earthquake at 10 km to be 200 Gal. As the distance increases by 1 km each time, it decays according to the logarithmic law, and the decay amplitude is related to the geological parameters. For different magnitudes, the corresponding initial reference values and decay rates are determined respectively to construct a complete attenuation relationship model framework.

[0089] Call the historical dataset containing 5,000 earthquake records in the past 10 years, and compare the acceleration benchmark values calculated by the model for different magnitude-distance combinations with the actually measured acceleration values. For example, when the model calculates that the acceleration at 10 km from a magnitude 6 earthquake is 200 Gal, while the actual records are mostly between 180 - 220 Gal. Adjust the model parameters according to the deviation. After repeated calibration, generate an acceleration attenuation benchmark value mapping table covering magnitudes 3 - 8 and distances 1 - 100 km. Associate the mapping table with the geographical grid coordinate system, and each grid cell corresponds to a unique magnitude-distance combination. For example, a certain grid cell corresponds to a magnitude 5 earthquake and a distance of 20 km. Obtain the attenuation benchmark value of 120 Gal by querying the mapping table, and use this as the basic predicted value of the ground motion acceleration for this grid cell. The accelerometer samples acceleration data at a sampling frequency of 100 Hz, and distributes the data to the corresponding geographical grid cell according to the longitude and latitude coordinates of the monitoring point. For example, the data of the monitoring point with coordinates (110.2°, 30.2°) is distributed to the grid cell with the corresponding coordinates. Compare the basic predicted value of each grid cell with the measured data point by point, and calculate the prediction deviation rate. For example, if the basic predicted value of a certain grid cell is 150 Gal and the measured average value is 160 Gal, the deviation rate is ≈6.25%. If the deviation rate exceeds the preset threshold of 5%, adjust the model parameters according to the deviation direction. When the actual value is greater than the predicted value, increase the magnitude-distance attenuation parameter and at the same time fine-tune the geological medium correction coefficient. For example, adjust the elastic modulus correction coefficient of the granite area from 1.0 to 1.05.

[0090] Based on the corrected parameters, recalculate the acceleration prediction values of each grid cell, and determine the confidence interval through statistical methods. Conduct 100 simulation calculations for the same grid cell, and take the numerical range at the 95% confidence level. For example, if the calculation result is 155 Gal and the confidence interval is ±5 Gal, then the calibrated prediction value interval is [150, 160] Gal. Analyze the spatial distribution of the calibrated acceleration prediction values. If it is found that the prediction values of multiple adjacent grid cells in a certain direction increase significantly, such as from east to west, the acceleration prediction value increases from 100 Gal to 200 Gal, and the initial time of the P-wave is delayed by 0.5 - 1 second in this direction. According to the seismic wave propagation speed (such as 6 km / s), it can be inferred that the source rupture direction is from east to west, and the rupture speed is about 6 km / s. Estimate the rupture length by calculating the distance corresponding to the change range of the prediction values in this direction. Rupture length = rupture speed × prediction value change time window; where the rupture speed is based on the speed of the source rupture surface expansion (unit: km / s), and the value here is 6 km / s, indicating that the rupture surface expands 6 kilometers eastward and westward per second; the prediction value change time window is the time interval (unit: s) required for the vibration prediction parameter (such as acceleration) to change from the starting point to stability. After determining this time window through the monitoring data and multiplying it by the rupture speed, the expansion distance of the rupture surface in this direction, that is, the rupture length, can be obtained (the estimated result is about 30 km). Integrate the calibrated acceleration prediction value interval (such as [150, 160] Gal), the initial P-wave parameters (time, amplitude, frequency), and the source rupture characteristics (direction, speed, length), add the intensity meter verification mark, and form a multi-dimensional data stream including time (such as 2024-01-01 10:00:00.000), location (grid coordinates), intensity prediction, and source characteristics.

[0091] Extract the initial P-wave parameters through the sliding time window analysis method and threshold judgment, which can accurately identify the P-wave; construct and calibrate the attenuation relationship model by combining historical data, and dynamically calibrate the prediction values using the measured data, effectively improving the accuracy of ground motion acceleration prediction and reducing errors. Calculate the ground motion acceleration prediction values personalized according to the geological medium parameters and spatial distribution of different geographical grid cells, and can correct the parameters in real time according to the measured data, which can better adapt to different geological conditions and earthquake situations and improve the reliability of prediction. Extracting the source rupture characteristics can not only predict the ground motion intensity but also understand the source rupture situation, providing more comprehensive and valuable information for earthquake research and emergency decision-making.

[0092] In a preferred embodiment of the present invention, input the multi-dimensional seismic data stream verified by the intensity meter into the preset geographical grid coordinate system. By jointly analyzing the P-wave parameters, the ground motion acceleration prediction values, and the source rupture characteristics, generate a dynamically updated magnitude-epicenter location probability distribution matrix, and calculate the dynamic trigger threshold to generate a hierarchical early warning trigger instruction, which may include:

[0093] Based on the geographical grid coordinate system, calculate the distance of the seismic wave propagation path from the earthquake source to each grid cell, and combine the geological medium distribution parameters corresponding to the path to extract the geological density and elastic modulus data;

[0094] According to the calibrated predicted value interval of ground motion acceleration, combine the earthquake source rupture direction and propagation speed parameters to construct the spatio-temporal constraint conditions for magnitude and epicenter location, and the spatio-temporal constraint conditions include magnitude upper limit, rupture propagation range and time window;

[0095] Based on the geological parameters and spatio-temporal constraint conditions, through the Monte Carlo sampling method, iteratively fuse the earthquake source probability distribution, the prior probability of ground motion intensity and the spatio-temporal constraint conditions to update the joint probability density of magnitude-epicenter location for each grid cell, and generate a dynamically updated magnitude-epicenter location probability distribution matrix, specifically including:

[0096] According to the magnitude upper limit and rupture propagation range in the spatio-temporal constraint conditions, define the parameter random generation interval for Monte Carlo sampling;

[0097] Perform random sampling within the parameter random generation interval to generate multiple groups of candidate parameter combinations of magnitude, epicenter location and ground motion intensity;

[0098] For each group of candidate parameter combinations, calculate the probability value in the earthquake source probability distribution, the ground motion intensity probability value and the compliance weight under the spatio-temporal constraint conditions;

[0099] Fuse the earthquake source probability value, the ground motion intensity probability value and the compliance weight to obtain the joint probability density corresponding to the parameter combination;

[0100] Statistically analyze the joint probability density of all candidate parameter combinations, calculate the mean value of the probability density within each grid cell, and map the mean value to the corresponding grid cell in the geographical grid coordinate system to update the value of the magnitude-epicenter location probability distribution matrix. When the number of Monte Carlo sampling times reaches the preset threshold, output the updated magnitude-epicenter location probability distribution matrix;

[0101] Extract the set of grid cells with magnitude probability exceeding the preset intensity threshold from the dynamically updated magnitude-epicenter location probability distribution matrix, and mark them as high-probability regions;

[0102] According to the population density level and building seismic resistance level parameters corresponding to the high-probability regions in the external database, calculate the regional comprehensive risk coefficient;

[0103] Real-time monitor the dynamic trigger threshold and the growth rate of magnitude probability. When the trigger threshold reaches the preset critical value and the growth rate exceeds the growth rate threshold, generate a hierarchical warning trigger instruction including the target area code and warning level.

[0104] In an embodiment of the present invention, the earth's surface and underground space are divided according to a unified standard to construct a three-dimensional geographic grid coordinate system. To balance calculation accuracy and efficiency, each grid cell is set as a cube with a side length of 1 kilometer. Taking a certain earthquake as an example, assuming the three-dimensional coordinates of the earthquake source are longitude 110°, latitude 30°, and depth 10 km, hundreds of thousands of grid cells covering the study area will be processed one by one. When calculating the distance from the earthquake source to the center point of the grid cell, the first step is coordinate transformation. With the center of the earth as the origin, the longitude, latitude, and depth information of the earthquake source and the center point of the grid cell are substituted using the method of converting geographic coordinates to rectangular coordinates. For example, for the earthquake source point, through specific conversion rules, its longitude and latitude are converted into the X, Y, and Z coordinate values in the three-dimensional rectangular coordinate system, and the characteristics of the earth as an approximate sphere are considered during the conversion process to ensure the accuracy of the coordinates.

[0105] After completing the coordinate transformation, since the earth is not a standard sphere, the plane distance formula cannot be directly used. The coordinates are corrected twice using the average radius of the earth, 6371 km.

[0106] The specific operation is to fine-tune the transformed rectangular coordinates according to the spherical curvature of the earth to make the coordinate values more conform to the true shape of the earth. Subsequently, using the three-dimensional space distance formula, the straight-line distance from the earthquake source to the center point of each grid cell is calculated. However, seismic waves actually propagate near the earth's surface, so the straight-line distance needs to be further converted into the actual propagation path distance along the earth's surface through the principle of spherical trigonometry. For example, by calculating the great circle arc length between two points on the sphere, a more accurate propagation distance value is obtained.

[0107] While calculating the propagation path distance, along the propagation path trajectory, corresponding parameters are extracted from a professional geological database, which integrates multi-source data such as geological exploration and geophysical exploration and details the geological types, geological densities, and elastic moduli of different regions. If the propagation path passes through a granite area, the density of granite (ranging from 2600 - 2800 kg / m³, taking 2700 kg / m³ here) and the elastic modulus (about 60 - 70 GPa, taking 60 GPa here) are accurately extracted; if it passes through a sedimentary rock area, the corresponding parameters of the sedimentary rock are obtained.

[0108] Construct time-space constraint conditions:

[0109] Based on the predicted range of ground motion acceleration calibrated in the early stage, combined with the known source rupture direction (such as from east to west) and propagation speed (6 km / s), start to construct the spatio-temporal constraint conditions for magnitude and epicenter location. When determining the magnitude upper limit, deeply analyze historical earthquake data and professional seismological theories. Taking a certain seismic zone as an example, retrieve the earthquake records in this area over the past century, analyze the maximum magnitude, the accumulation of geological tectonic stress, and combine factors such as the trend of plate movement. Comprehensively judge and set the magnitude upper limit for this earthquake. After evaluation, if the historical maximum magnitude is 7.8 and there is no significant abnormality in the current geological stress state, set the magnitude upper limit to 8.0, while reserving a certain safety margin. When determining the rupture propagation range, calculate according to the source rupture speed and time. Assume that 10 seconds after the earthquake occurs, with a rupture speed of 6 km / s, the source rupture surface theoretically expands 60 kilometers along the east-west direction. However, in actual calculations, consider the hindering or promoting effect of geological structure differences on rupture. For example, when encountering a hard rock layer, the rupture speed may decrease, thereby correcting the expansion range and delineating a more realistic rupture influence area.

[0110] When determining the time window, comprehensively consider factors such as seismic wave propagation speed and source rupture speed. By analyzing the statistical data of historical seismic wave propagation speed in this area and combining the source characteristics of this earthquake, set 30 seconds from the moment of earthquake occurrence as the effective time window. Within these 30 seconds, the system believes that the influence of seismic wave propagation and source rupture on the surrounding area is in a critical change stage. Beyond this time, the influence tends to be stable or other complex changes occur, and it is no longer included in the core analysis period.

[0111] Define the sampling interval:

[0112] Based on the spatio-temporal constraint conditions, determine the parameter random generation intervals for Monte Carlo sampling. In terms of magnitude, combined with the magnitude upper limit of 8.0, considering that the earthquake occurrence probability decreases as the magnitude increases, set the random generation interval as [5.0, 8.0], covering most possible magnitude situations; the random generation range of the epicenter location is strictly limited within the previously determined rupture propagation area to ensure that the sampling location is of practical significance; the random generation interval of ground motion intensity is set according to the distribution law of ground motion intensity in this area under different magnitudes, distances, and geological conditions in historical data, combined with the current geological parameters, to set a reasonable range, such as [20 Gal, 300 Gal].

[0113] Within the defined intervals, start large-scale random sampling. Each sampling generates a set of parameter combinations of magnitude, epicenter location, and ground motion intensity. For example, one sampling may obtain a combination of magnitude 6.5, epicenter at a certain grid cell (longitude 110.2°, latitude 30.1°), and ground motion intensity 120 Gal. Continuously sample to generate a large number of candidate parameter combinations.

[0114] Calculate the probability value and weight:

[0115] For each set of candidate parameter combinations, calculate three key values respectively:

[0116] Seismic source probability value: Based on historical earthquake data and geological structure conditions, the seismic source probability distribution is determined to evaluate the possibility of an earthquake occurring at the epicenter position in the parameter combination. For example, in the area near the geological fault zone, the seismic source probability value is relatively high, with a probability of 0.7 - 0.9; while in the geologically stable area, the probability value is relatively low, being 0.1 - 0.3.

[0117] Seismic ground motion intensity probability value: Based on the established seismic ground motion attenuation model, which comprehensively considers the physical laws of seismic wave propagation. In the model, the magnitude directly determines the initial energy of the seismic wave. For example, for every 1 - level increase in magnitude, the energy carried by the seismic wave increases exponentially, which means that under the same propagation conditions, the reference value of the seismic ground motion intensity generated by a higher - magnitude earthquake in the target area is larger. By calculating the actual propagation distance from the seismic source to the target grid cell, the propagation path of the seismic wave is determined. The area closer to the seismic source receives relatively less energy loss of the seismic wave and has a higher seismic ground motion intensity; conversely, the farther the distance, the more energy is dissipated during propagation, and the seismic ground motion intensity decreases. Taking the propagation path passing through different geological regions as an example: Suppose the propagation path passes through a granite formation with a geological density of about 2700 kg / m³ and an elastic modulus of up to 60 GPa. Such a rock structure is dense, with weak absorption and scattering effects on seismic waves, fast seismic wave propagation speed, and slow energy attenuation. When the seismic wave of a magnitude - 5 earthquake passes through this area and propagates 10 km, based on the seismic ground motion attenuation model calculation, the expected seismic ground motion intensity generated in this area is 120 Gal. By analyzing historical earthquake cases with the same magnitude, distance, and similar geological conditions, it is statistically found that the number of times the actual seismic ground motion intensity appears in the range of 110 - 130 Gal accounts for 70% of the total number of times. Thus, the seismic ground motion intensity probability value in this case is determined to be 0.7, which is at a relatively high level. If the propagation path passes through a silty clay layer with a density of about 1800 kg / m³ and an elastic modulus of only 5 GPa. The soft soil makes the seismic wave prone to waveform distortion and energy dissipation during propagation, with a significant reduction in the seismic wave propagation speed and rapid energy attenuation. Similarly, for a magnitude - 5 earthquake propagating 10 km, the expected seismic ground motion intensity in this area may be only 60 Gal. By analyzing the same - type historical earthquake data, it is found that the number of times the actual seismic ground motion intensity appears in the range of 50 - 70 Gal accounts for 30% of the total number of times. Therefore, the seismic ground motion intensity probability value in this case is set to 0.3, which is significantly lower than that in the hard - rock area.

[0118] It will traverse all target grid cells and repeat the above analysis process for each combination of the specific magnitude, epicenter location, propagation distance, and geological conditions of each grid cell. Through the statistical analysis of a large amount of historical earthquake data and the theoretical calculation combined with the ground motion attenuation model, the probability of the ground motion intensity appearing within a certain range under each combination is comprehensively determined, and finally, the ground motion intensity probability values ranging from 0.2 to 0.8 are obtained.

[0119] Compliance weight: It is used to judge whether the parameter combination meets the spatio-temporal constraint conditions. If the magnitude exceeds the upper limit, the epicenter location is not within the rupture propagation range, or the time is not within the set time window, the weight is directly set to 0; for combinations that fully meet the conditions, the weight ranges from 0.8 to 1.0 according to the degree of fit with the constraint conditions. For example, for a combination that is exactly at the center of the rupture propagation range and the magnitude is close to the average expectation, the weight is taken as 1.0; for a combination close to the edge of the range, the weight is taken as 0.8.

[0120] Fusion to obtain the joint probability density:

[0121] Fuse the source probability value, ground motion intensity probability value, and compliance weight. For example, by the method of weighted multiplication, the three values are comprehensively calculated to obtain the joint probability density of each group of parameter combinations. Suppose the source probability value of a certain group of parameter combinations is 0.7, the ground motion intensity probability value is 0.6, and the compliance weight is 0.9, then the joint probability density is 0.7×0.6×0.9 = 0.378, and this value intuitively reflects the possibility of this earthquake scenario occurring.

[0122] Update the probability distribution matrix:

[0123] Statistically analyze the joint probability densities of all candidate parameter combinations, and calculate the mean value of the joint probability densities of all combinations within each grid cell. For example, if a grid cell contains 1000 groups of candidate parameter combinations, add up the joint probability densities of these 1000 groups and divide by 1000 to obtain the mean value. Subsequently, map this mean value to the grid cell corresponding to the geographical grid coordinate system and update the values in the magnitude-epicenter location probability distribution matrix. The system continuously repeats the sampling, calculation, and update processes. When the number of sampling times reaches the preset 100,000 times, stop the calculation and output the finally updated probability distribution matrix, clearly showing the earthquake occurrence probability of each grid cell at different magnitudes.

[0124] High-probability area identification and risk coefficient calculation:

[0125] From the updated probability distribution matrix, screen out the grid cells with earthquake magnitude probabilities exceeding the preset intensity threshold (such as 0.6) and mark them as high-probability areas. After determining the high-probability areas, retrieve the population density levels and building seismic resistance level parameters of the corresponding areas from an external database. The population density levels are divided into high density (more than 5,000 people per square kilometer), medium density (1,000 - 5,000 people), and low density (less than 1,000 people); the building seismic resistance levels are divided into Class A (strong seismic design), Class B (medium seismic design), and Class C (ordinary design) according to the structural type and construction standards. When calculating the comprehensive risk coefficient of the area, a weighted calculation method is used. For example, a weight of 0.6 is assigned to the population density level and a weight of 0.4 is assigned to the building seismic resistance level. For a certain high-probability area, if the population density level is high density (corresponding value 3) and the building seismic resistance level is Class C (corresponding value 1), then the comprehensive risk coefficient is 3×0.6 + 1×0.4 = 2.2. The higher the risk coefficient, the greater the risks faced by the people and buildings in this area during an earthquake.

[0126] Real-time monitor the dynamic trigger threshold and the growth rate of earthquake magnitude probability. The dynamic trigger threshold is adjusted dynamically according to the comprehensive risk coefficient of the area, historical earthquake data, and emergency response requirements. In areas with dense population, weak building seismic resistance, and high earthquake risks, the dynamic trigger threshold may be set at 0.7; in relatively safe areas, the threshold may be set at 0.8. The growth rate of earthquake magnitude probability reflects the change speed of the probability over time. For example, a growth of 0.01 per second is regarded as a change unit. When the dynamic trigger threshold reaches the preset critical value (such as 0.7) and the growth rate of earthquake magnitude probability exceeds the growth rate threshold (such as 0.02 per second), it is determined that the earthquake risk has reached the warning standard. At this time, generate a hierarchical warning trigger instruction containing the target area code (such as a unique code set according to the grid cell number) and the warning level. The warning levels are divided into three levels: Level 1 warning (extremely high risk) requires immediate evacuation of personnel; Level 2 warning (relatively high risk) prompts to make preparations for taking shelter; Level 3 warning (medium risk) reminds to pay attention to subsequent dynamics to ensure that the warning information is accurately sent to the corresponding area to guide the emergency response work.

[0127] By considering the propagation path distance, geological parameters, and spatio-temporal constraints, and combining the Monte Carlo sampling method, a magnitude-epicenter location probability distribution matrix is calculated by integrating various factors. Compared with single-factor analysis, it can more accurately reflect the likelihood and influence range of an earthquake, reducing prediction errors. It can update the probability distribution matrix and dynamic trigger threshold in real time, and adjust the early warning strategy in a timely manner according to the actual situation of earthquake development and newly obtained data, adapting to the uncertainty and complexity of the earthquake process, and improving the timeliness and effectiveness of the early warning system. By calculating the regional comprehensive risk coefficient and generating hierarchical early warning trigger instructions, it can provide more targeted decision-making basis for emergency management departments. According to different early warning levels, corresponding emergency measures are taken, rescue resources are rationally allocated, the emergency response efficiency is improved, and the losses caused by earthquake disasters are reduced.

[0128] In a preferred embodiment of the present invention, according to the hierarchical early warning trigger instruction and the emergency response plan library of the preset AI chip, the earthquake avoidance guidance strategy for the target area is matched, and a multimedia early warning message including earthquake countdown parameters, intensity distribution prediction, and AI-optimized hazard avoidance tips is generated, which may include:

[0129] According to the early warning level in the hierarchical early warning trigger instruction, the corresponding response strategy template is matched from the emergency response plan library to determine the priority, language style, and geographical coverage of the broadcast content;

[0130] Based on the earthquake wave propagation speed parameter and the updated magnitude-epicenter location probability distribution matrix, the propagation time of the earthquake wave reaching each grid unit within the geographical coverage is calculated, and a regionalized countdown parameter synchronized with the time axis is generated;

[0131] The intensity distribution prediction data of the high-probability area is spatially associated and matched with the road network topology and shelter coordinates in the external database to generate intensity distribution spatial data superimposed with road traffic status and shelter resource identifiers, specifically including:

[0132] Based on the geographical grid coordinate system, the intensity distribution prediction data, road network topology, and shelter coordinates of the high-probability area are spatially aligned to generate an associated data set in a unified coordinate system;

[0133] Based on the road network topology in the associated data set and combined with real-time traffic data, the traffic status of each road segment in the high-probability area is analyzed to generate road traffic status data including road grade, congestion degree, and availability identifier; based on the shelter coordinates in the associated data set and combined with the capacity database, the real-time available capacity of each shelter is marked to generate shelter resource availability identifier data;

[0134] The intensity distribution prediction data, road traffic status data and refuge resource availability identification data are integrated according to the spatial location of the geographic grid unit to generate intensity distribution spatial data with superimposed road traffic status and refuge resource identification;

[0135] According to the database of building structure characteristics in the target area, the AI ​​chip analyzes the building type and earthquake resistance, and combines the intensity level predicted in the intensity distribution spatial data to dynamically generate earthquake avoidance guidance content adapted to the building type;

[0136] The regional countdown parameters, intensity distribution spatial data and vibration avoidance guidance content are synchronously integrated according to timestamps and spatial positions to generate multimedia warning information including voice broadcasts, text prompts and dynamic visualization elements.

[0137] In an embodiment of the present invention, the emergency response plan library stores policy templates in advance according to the warning level (level one, level two, and level three). Taking the level one warning as an example, after receiving the graded warning trigger instruction containing the "level one warning", the corresponding template is retrieved from the plan library. The template has preset the priority of the broadcast content: first broadcast the earthquake countdown and the core risk avoidance instructions, and then scroll the intensity distribution and evacuation instructions; the language style is set to a quick and concise short sentence, such as "Immediately hide in the corner of the load-bearing wall!"; the geographical coverage range is based on the target area code in the instruction, and all grid cells in the high probability area are locked to ensure that the information coverage has no dead angles. For the level two warning, the template will adjust the content priority, first explain the earthquake risk level, and then gradually guide risk avoidance. The language style is relatively gentle, and the coverage range can be narrowed to the grid cells with the most concentrated risks.

[0138] Regionalized countdown parameter generation:

[0139] Based on the average propagation velocity parameter of seismic waves (such as 6km / s) and the updated magnitude-epicenter location probability distribution matrix, the propagation time is calculated for each grid cell within the geographic coverage area. The specific steps are as follows:

[0140] Get the source distance corresponding to each grid cell from the probability distribution matrix (for example, a grid cell is 15 km away from the source); divide the distance by the propagation speed to get the time it takes for the earthquake wave to reach the grid cell (15 km ÷ 6 km / s = 2.5 seconds); take the time of the earthquake as the starting point (such as 2025-05-21 14:30:00), add the propagation time, and generate the countdown end time of the grid cell (14:30:02.5); summarize the countdown time of all grid cells by geographical area, and generate regional countdown parameters with a time axis interval of 1 second, for example: "In the next 2 seconds, area A (grids 1-10) will be affected by the earthquake; in the next 3 seconds, area B (grids 11-20) will enter the warning range".

[0141] Based on the geographic grid coordinate system (such as the WGS84 coordinate system), the intensity distribution prediction data of high-probability areas (such as the predicted intensity of grid cells 1-5 is level VII), road network topology data (such as the main road X1 in grid cell 3), and shelter coordinate data (such as the shelter S1 in grid cell 4, with coordinates of longitude 110.1° and latitude 30.1°) are spatially aligned. Through the coordinate conversion algorithm, data from different sources are unified into the same grid cell, for example, the coordinates of the main road X1 and shelter S1 are mapped to the corresponding grid cells 3 and 4, forming an associated data set containing location and attributes.

[0142] Combined with real-time traffic data (such as vehicle density monitored by radar), the road network topology in the associated dataset is analyzed:

[0143] For the main road X1, if the real-time traffic density exceeds 80 vehicles / km, it is considered to be congested, marked in red, and the availability indicator is "pass with caution";

[0144] For the secondary trunk road Y1, the traffic density is less than 30 vehicles / km, which is considered to be unobstructed, marked in green, and the availability indicator is "priority passage";

[0145] Finally, the road traffic status data including road grade (main road / secondary road), congestion level (red / yellow / green), and availability mark are generated. For example, the main road X1 in grid unit 3 is displayed as "main road-red-pass with caution".

[0146] Retrieve the capacity database of shelters (e.g., the total capacity of shelter S1 is 500 people), combine it with the real-time registration data (e.g., 200 people have been accommodated), calculate the real-time available capacity (500-200=300 people), and mark it as "Shelter S1-Available capacity 300 people". Repeat this process for all shelters to generate shelter resource availability identification data, such as shelter S1 in grid unit 4 is displayed as "blue icon-300 people available". Spatially fuse the intensity distribution prediction data (e.g., grid unit 3-4 is level VII), road traffic status data (e.g., main road X1 is congested), and shelter resource availability identification data (e.g., shelter S1 is available for 300 people) by grid unit. For example, in grid unit 3, the superimposed visualization result is generated: the background color is orange representing level VII intensity, the main road X1 shows a red congestion line, and the nearby shelter S1 shows a blue available icon and capacity value, forming intuitive intensity distribution spatial data.

[0147] The target area building structural characteristics database stores seismic capacity data of different building types, such as:

[0148] Masonry structure (seismic fortification intensity VI): In the predicted intensity VII area, the AI ​​chip determines that the building has a high risk of collapse;

[0149] Frame structure (seismic fortification intensity VIII): In the predicted intensity VII area, the structure is judged to be basically safe but attention should be paid to secondary disasters.

[0150] Combined with the spatial data of intensity distribution, the AI ​​chip dynamically generates adaptation guidance content:

[0151] For users of masonry structures: "Your building is a masonry structure, and the predicted intensity is Level VII. Please hide in the corner of the load-bearing wall immediately and protect your head with tables and chairs!";

[0152] To users of frame structures: "The building you are in is a frame structure, and the predicted intensity is Level VII. Please avoid windows, hide under a desk, and wait for evacuation instructions!"

[0153] Using timestamp and spatial location as indexes, regional countdown parameters, intensity distribution spatial data, and earthquake avoidance guidance content are synchronously integrated:

[0154] Timeline synchronization: When the countdown is 2 seconds, the voice broadcasts "The earthquake will reach area A in 2 seconds!" At the same time, the grid unit of area A on the map flashes orange (indicating the VIIth level of intensity), and a text prompt of the corresponding masonry structure's earthquake avoidance action pops up;

[0155] Spatial location synchronization: In the grid unit 3-4 area, the dynamic visualization layer shows the congested main road X1 (red line) and the available shelter S1 (blue icon), and the voice broadcasts "In the grid unit 3-4 area, please avoid the main road X1 and go to the shelter S1 (available for 300 people) first!".

[0156] The final generated multimedia warning information includes:

[0157] Voice broadcast: emergency instructions played in order of priority (such as countdown, evasive action);

[0158] Text prompt: Intensity level and evacuation instructions for each region (e.g. “Grid cells 1-5: Intensity level VII, it is recommended to go to Shelter S2”);

[0159] Dynamic visualization: real-time updated intensity distribution map, road traffic status layer, and refuge resource signs.

[0160] By matching the early warning level with the pre - plan template, "classified response and precise measures" are achieved, shortening the information generation time and ensuring that high - risk areas receive emergency instructions first. For example, the delay of core information broadcast under level - one early warning can be controlled within 1 second. The regionalized countdown parameters are synchronized with the real - time seismic wave propagation path, enabling users to intuitively perceive the threat arrival time (such as "the earthquake will arrive in 3 seconds"). Combined with the dynamic intensity map, "time + space" two - dimensional evacuation guidance is realized, improving the efficiency of evacuation decision - making. By integrating intensity prediction, road traffic, and shelter resource data, users can quickly obtain comprehensive information on "where it is dangerous, which roads can be taken, and where to take shelter". For example, in high - intensity areas, congested roads are automatically avoided, and the shortest safe evacuation route is recommended. The AI - based adaptation tips according to building types avoid "one - size - fits - all" suggestions and improve the scientific nature of evacuation measures. For example, users of masonry structures receive targeted reinforcement suggestions, and users of frame structures receive guidance on preventing secondary disasters.

[0161] In a preferred embodiment of the present invention, the multimedia early warning information is distributed to voice broadcast terminals, digital display terminals, and public information screens through the IP network protocol, realizing the priority transmission of voice broadcasts and the hierarchical rendering of dynamic information. At the same time, based on the real - time feedback of terminal response data, the pre - plan matching logic and dynamic trigger threshold determination conditions of the AI chip can be optimized, which may include:

[0162] Assign a transmission priority to the voice broadcast data through the IP network protocol, and dynamically adjust the rendering levels of text prompts and dynamic information elements based on the transmission status of the voice broadcast;

[0163] Real - time receive the broadcast completion status of the voice broadcast terminal, the rendering delay time of the digital display terminal, and the network load data of the public information screen to generate a terminal response data set;

[0164] According to the broadcast delay and rendering efficiency data in the terminal response data set, adjust the matching weights of the emergency response pre - plans in the AI chip, optimize the generation logic of the earthquake - resistant guidance strategy, and dynamically correct the determination conditions of the dynamic trigger threshold based on the network load fluctuation data in the terminal response data set.

[0165] In an embodiment of the present invention, the voice broadcast data, text prompts, and dynamic information elements (such as map layers, shelter identifiers) in the multimedia early warning information are classified and encapsulated through the IP network protocol (such as TCP / IP). For the voice broadcast data, the highest transmission priority is assigned to it, similar to the "emergency channel" mechanism in network communication. For example, mark the voice data as "priority 1" in the IP data packet header to ensure that it can still preferentially occupy bandwidth resources for transmission in case of network congestion.

[0166] During the transmission process, the transmission status of voice broadcast data (such as latency and packet loss rate) is monitored in real time. If it is detected that the voice transmission latency exceeds a preset threshold (such as 500 milliseconds), the rendering levels of text prompts and dynamic information elements are automatically adjusted. For example, temporarily reduce the refresh frequency of the dynamic map (from 5 times per second to 2 times per second) or simplify the text content (such as omitting some non-critical shelter guides) to release network resources and ensure the continuity of voice data. Conversely, if the voice transmission is stable, the rendering level of full-scale information is restored to ensure that users obtain complete dynamic information.

[0167] Receive feedback data from various terminals in real time:

[0168] Voice broadcast terminal: Feedback the "broadcast completion status", that is, the timestamp difference between the sending and playing completion of each voice instruction (such as the sending time of instruction A is 14:30:02, the playing completion time is 14:30:03, and the elapsed time is 1 second);

[0169] Digital display terminal: Feedback the "rendering latency time", that is, the time taken for text or images to be fully displayed from reception (such as the loading time of a grid unit intensity map is 800 milliseconds);

[0170] Public information screen: Feedback the "network load data", including the current number of connected terminals and bandwidth occupancy rate (such as a public information screen in a certain area is connected to 500 devices, and the bandwidth occupancy reaches 70%).

[0171] Integrate the above data into a terminal response dataset, for example:

[0172] Voice terminal A: Instruction ID-001, broadcast time 1.2 seconds, latency normal;

[0173] Digital terminal B: Map rendering time 1.5 seconds, exceeding the threshold by 0.5 seconds;

[0174] Public information screen C: Bandwidth occupancy rate 85%, approaching the congestion threshold (90%).

[0175] AI chip optimization and threshold correction:

[0176] Analyze the "broadcast latency" and "rendering efficiency" in the terminal response data. For example, if there is a widespread problem of excessive rendering latency in digital display terminals in a certain area (such as the average elapsed time exceeds 1 second), it indicates that the network environment in this area is poor and it is difficult to support the real-time display of complex dynamic information. At this time, the AI chip will automatically reduce the weight of "dynamic map rendering" in the emergency response plan for this area, preferentially generate simple text instructions (such as "Immediately go to the nearest shelter"), and increase the number of repetitions of voice broadcasts (such as changing from a single broadcast to a loop broadcast 3 times) to ensure the transmission of key information.

[0177] Based on the network load fluctuation data of the public information screen, dynamically adjust the early warning trigger logic. For example, when the network load rate in a certain area continuously exceeds 80%, it indicates that the terminal reception ability has declined, which may cause delays in early warning information. At this time, the system will pre-correct the dynamic trigger threshold, and lower the "magnitude probability growth rate acceleration threshold" from the default "0.02 per second" to "0.015 per second", that is, allow earlier triggering of early warnings, reserve more time for network transmission, and avoid the failure of early warnings due to transmission delays.

[0178] Through the voice priority transmission mechanism, ensure that core information such as earthquake countdown and evacuation instructions can still be quickly delivered in case of network congestion. For example, in high-network-load scenarios such as subways and shopping malls, the voice broadcast delay can be controlled within 1 second. Adjust the information rendering level in real time according to the terminal feedback, and automatically simplify non-critical content (such as reducing map accuracy) in areas with poor networks to avoid the lack of user information due to data transmission failures. For example, in scenarios with limited network bandwidth in remote mountainous areas, it can automatically switch to the "pure voice + simple text" mode to ensure the early warning coverage rate. Optimize the AI pre-plan matching logic based on the terminal response data to make the evacuation strategy more suitable for the actual propagation conditions. For example, in areas with a high proportion of the elderly population, if the voice terminal broadcast completion rate reaches 95%, the system will automatically increase the detail level of voice instructions; in areas concentrated with young groups, if the digital terminal rendering efficiency is high, strengthen the functions of dynamic maps and AR evacuation guides. Dynamically correct the trigger threshold through network load data to avoid early warning delays due to transmission delays. For example, in temporary high-load scenarios such as large event venues, the system pre-lowers the trigger threshold, and can advance the early warning release time by 2-3 seconds to buy precious time for personnel evacuation. From information distribution to terminal feedback to strategy optimization, form a complete emergency response closed loop, improve the self-adaptability and reliability of the system, especially in complex network environments or sudden traffic peak scenarios, and ensure the stability and effectiveness of the early warning system.

[0179] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which when run on a computer, cause the computer to execute the system as described above. All implementation manners in the above system embodiment are applicable to this embodiment and can achieve the same technical effects.

[0180] The above is the preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. An emergency broadcast intelligent triggering system based on multi-source seismic data fusion, characterized in that, Including: A data receiving module, which is used to receive multi-source earthquake early warning information in real time, including source parameter data, strong motion observation data and real-time early warning parameters, and output standardized multi-source earthquake data through a built-in communication protocol; A localization calculation module, which is used to perform real-time parsing on the standardized multi-source earthquake data, extract P-wave initial parameters, ground motion acceleration prediction values and source rupture characteristics, dynamically calibrate the parsing results in combination with the measured data of the built-in accelerometer, and output a multi-dimensional earthquake data stream verified by the accelerometer; A data processing module, which is used to input the multi-dimensional earthquake data stream verified by the accelerometer into a preset geographical grid coordinate system, generate a dynamically updated magnitude-epicenter location probability distribution matrix by jointly analyzing P-wave parameters, ground motion acceleration prediction values and source rupture characteristics, and calculate a dynamic trigger threshold to generate a hierarchical early warning trigger instruction; An emergency broadcast module, which is used to match the earthquake avoidance guidance strategy for the target area according to the hierarchical early warning trigger instruction and the emergency response plan library of the preset AI chip, and generate a multimedia early warning information including earthquake countdown parameters, intensity distribution prediction and AI-optimized avoidance tips; A collaborative broadcast control module, which is used to distribute the multimedia early warning information to voice broadcast terminals, digital display terminals and public information screens through the IP network protocol, realize the priority transmission of voice broadcast and the hierarchical rendering of dynamic information, and optimize the pre-plan matching logic of the AI chip and the dynamic trigger threshold determination condition based on the real-time feedback of terminal response data.

2. The emergency broadcast intelligent trigger system based on multi-source seismic data fusion according to claim 1, wherein Performing real-time parsing on the standardized multi-source earthquake data, extracting P-wave initial parameters, ground motion acceleration prediction values and source rupture characteristics, dynamically calibrating the parsing results in combination with the measured data of the built-in accelerometer, and outputting a multi-dimensional earthquake data stream verified by the accelerometer, including: Performing real-time scanning on the seismic waveform data through the sliding time window analysis method, and identifying the initial time of the P-wave and extracting the P-wave initial parameters according to the preset waveform slope threshold and energy change rate threshold; Based on the three-dimensional source coordinates in the P-wave initial parameters and the spatial distribution of geographical grid units, combining the physical laws of ground motion attenuation and historical statistical relationships, calculating the basic ground motion acceleration prediction value of each grid unit; Obtaining the measured data of real-time ground motion acceleration through the accelerometer built in the localization calculation module, dynamically comparing the basic prediction value with the measured data, if the deviation exceeds the preset error threshold, then correcting the magnitude-distance attenuation parameter and the geological medium correction coefficient to generate a calibrated ground motion acceleration prediction value interval; According to the spatial distribution characteristics of the calibrated ground motion acceleration prediction value interval, combining the time series change law of the P-wave initial parameters, extracting the source rupture direction, rupture speed and rupture length parameters; Fusing the calibrated ground motion acceleration prediction value interval, P-wave initial parameters and source rupture characteristics, adding an accelerometer verification mark, and generating a multi-dimensional earthquake data stream.

3. The emergency broadcast intelligent triggering system based on multi-source seismic data fusion according to claim 2, wherein Based on the three-dimensional source coordinates in the P-wave initial parameters and the spatial distribution of geographical grid units, combining the physical laws of ground motion attenuation and historical statistical relationships, calculating the basic ground motion acceleration prediction value of each grid unit, including: Calculate the seismic wave propagation path length based on the spatial distance between the three-dimensional coordinates of the seismic source and the center point of the geographical grid unit, and extract the geological density and elastic modulus parameters by combining the distribution data of the geological medium types corresponding to the paths; Based on the geological density and elastic modulus parameters, and in combination with the physical laws of ground motion attenuation, construct a magnitude-distance attenuation relationship model, where the magnitude-distance attenuation relationship model defines the logarithmic attenuation characteristics of the acceleration attenuation reference values at different magnitudes with the propagation path length; According to the historical ground motion statistical data set, statistically calibrate the attenuation reference values to generate a mapping table of acceleration attenuation reference values for different magnitude and distance combinations; Bind the mapping table of acceleration attenuation reference values to the spatial distribution of the geographical grid coordinate system to generate the basic predicted values of ground motion acceleration for each grid unit.

4. The emergency broadcast intelligent trigger system based on multi-source seismic data fusion according to claim 3, characterized in that Obtain the measured real-time ground motion acceleration data through the seismometer built in the localization calculation module, and dynamically compare the basic predicted values with the measured data. If the deviation exceeds the preset error threshold, then correct the magnitude-distance attenuation parameters and the geological medium correction coefficient to generate a calibrated ground motion acceleration prediction value interval, including: Based on the spatial distribution of the geographical grid unit, map the measured real-time ground motion acceleration data to the corresponding grid unit; Compare the basic predicted values of ground motion acceleration for each grid unit with the measured data of the same grid unit point by point, calculate the prediction deviation rate. If the prediction deviation rate exceeds the preset error threshold, then adjust the magnitude-distance attenuation parameters and the geological medium correction coefficient according to the deviation direction to obtain the corrected attenuation parameters and correction coefficient; Based on the corrected attenuation parameters and correction coefficient, recalculate the ground motion acceleration prediction values to generate a calibrated prediction value interval including the confidence interval range.

5. The emergency broadcast intelligent triggering system based on multi-source seismic data fusion according to claim 4, characterized in that, Input the multi-dimensional seismic data stream verified by the seismometer into the preset geographical grid coordinate system. By jointly analyzing the P-wave parameters, the ground motion acceleration prediction values, and the seismic source rupture characteristics, generate a dynamically updated magnitude-epicenter position probability distribution matrix, and calculate the dynamic trigger threshold to generate a hierarchical warning trigger instruction, including: Based on the geographical grid coordinate system, calculate the seismic wave propagation path distance from the seismic source to each grid unit, and extract the geological density and elastic modulus data by combining the geological medium distribution parameters corresponding to the paths; According to the calibrated ground motion acceleration prediction value interval, and in combination with the seismic source rupture direction and the propagation speed parameters, construct the spatio-temporal constraint conditions of magnitude and epicenter position, where the spatio-temporal constraint conditions include the magnitude upper limit, the rupture propagation range, and the time window; Based on the geological parameters and the spatio-temporal constraint conditions, through the Monte Carlo sampling method, iteratively fuse the seismic source probability distribution, the prior probability of ground motion intensity, and the spatio-temporal constraint conditions to update the joint probability density of magnitude and epicenter position for each grid unit, and generate a dynamically updated magnitude-epicenter position probability distribution matrix; Extract the set of grid units with magnitude probabilities exceeding the preset intensity threshold from the dynamically updated magnitude-epicenter position probability distribution matrix, and mark them as high-probability regions; Calculate the regional comprehensive risk coefficient according to the population density level and the building seismic resistance level parameters corresponding to the high-probability regions in the external database; Real-time monitor the dynamic trigger threshold and the magnitude probability growth rate. When the trigger threshold reaches the preset critical value and the growth rate exceeds the growth rate threshold, generate a hierarchical warning trigger instruction including the target area code and the warning level.

6. The intelligent triggering system for emergency broadcasting based on multi-source seismic data fusion according to claim 5, characterized in that, Based on geological parameters and spatio-temporal constraint conditions, through the Monte Carlo sampling method, iteratively fuse the seismic source probability distribution, the prior probability of ground motion intensity, and spatio-temporal constraint conditions to update the joint probability density of magnitude-epicenter location for each grid cell, and generate a dynamically updated magnitude-epicenter location probability distribution matrix, including: Define the parameter random generation interval for Monte Carlo sampling according to the magnitude upper limit and the rupture propagation range in the spatio-temporal constraint conditions; Perform random sampling within the parameter random generation interval to generate multiple groups of candidate parameter combinations of magnitude, epicenter location, and ground motion intensity; For each group of candidate parameter combinations, calculate the probability value in the seismic source probability distribution, the ground motion intensity probability value, and the compliance weight under spatio-temporal constraint conditions; Fuse the seismic source probability value, the ground motion intensity probability value, and the compliance weight to obtain the joint probability density of the corresponding parameter combination; Statistically analyze the joint probability density of all candidate parameter combinations, calculate the mean probability density within each grid cell, and map the mean value to the corresponding grid cell in the geographical grid coordinate system to update the values in the magnitude-epicenter location probability distribution matrix. When the number of Monte Carlo sampling times reaches the preset threshold, output the updated magnitude-epicenter location probability distribution matrix.

7. The emergency broadcast intelligent trigger system based on multi-source seismic data fusion according to claim 6, characterized in that According to the hierarchical warning trigger instruction and the emergency response plan library of the preset AI chip, match the earthquake avoidance guidance strategy for the target area, and generate a multimedia warning message including earthquake countdown parameters, predicted intensity distribution, and AI-optimized hazard avoidance tips, including: According to the warning level in the hierarchical warning trigger instruction, match the corresponding response strategy template from the emergency response plan library to determine the priority, language style, and geographical coverage of the broadcast content; Based on the seismic wave propagation speed parameter and the updated magnitude-epicenter location probability distribution matrix, calculate the propagation time of seismic waves reaching each grid cell within the geographical coverage area, and generate regionalized countdown parameters synchronized with the time axis; Spatially correlate and match the predicted intensity distribution data of the high-probability area with the road network topology and shelter coordinates in the external database to generate intensity distribution spatial data superimposed with road traffic status and shelter resource identifiers; According to the building structure feature database of the target area, analyze the building type and seismic resistance ability through the AI chip, and combine with the predicted intensity level in the intensity distribution spatial data to dynamically generate earthquake avoidance action guidance content adapted to the building type; Synchronously fuse the regionalized countdown parameters, intensity distribution spatial data, and earthquake avoidance action guidance content according to the time stamp and spatial position to generate a multimedia warning message including voice broadcast, text prompt, and dynamic visualization elements.

8. The emergency broadcast intelligent trigger system based on multi-source seismic data fusion according to claim 7, characterized in that Spatially correlate and match the predicted intensity distribution data of the high-probability area with the road network topology and shelter coordinates in the external database to generate intensity distribution spatial data superimposed with road traffic status and shelter resource identifiers, including: Based on the geographic grid coordinate system, align the intensity distribution prediction data of the high-probability area, the road network topology, and the coordinates of the shelters in terms of spatial position to generate an associated dataset under the unified coordinate system; Based on the road network topology in the associated dataset and combined with real-time traffic data, analyze the traffic conditions of each road segment in the high-probability area to generate road traffic condition data including road grade, congestion level, and availability identification; Based on the coordinates of the shelters in the associated dataset and combined with the capacity database, mark the real-time available capacity of each shelter to generate shelter resource availability identification data; Fuse the intensity distribution prediction data, road traffic condition data, and shelter resource availability identification data according to the spatial position of the geographic grid cells to generate intensity distribution spatial data with superimposed road traffic conditions and shelter resource identifications.

9. The emergency broadcast intelligent trigger system based on multi-source seismic data fusion according to claim 8, characterized in that Distribute the multimedia warning information to the voice broadcast terminal, digital display terminal, and public information screen through the IP network protocol to achieve priority transmission of voice broadcasts and hierarchical rendering of dynamic information. At the same time, based on the real-time feedback of the terminal response data, optimize the pre-plan matching logic and dynamic trigger threshold determination conditions of the AI chip, including: Assign a transmission priority to the voice broadcast data through the IP network protocol, and dynamically adjust the rendering levels of text prompts and dynamic information elements based on the transmission status of the voice broadcast; Receive in real-time the broadcast completion status of the voice broadcast terminal, the rendering delay time of the digital display terminal, and the network load data of the public information screen to generate a terminal response dataset; According to the broadcast delay and rendering efficiency data in the terminal response dataset, adjust the matching weights of the emergency response pre-plans in the AI chip, optimize the generation logic of the earthquake avoidance guidance strategy, and based on the network load fluctuation data in the terminal response dataset, dynamically correct the determination conditions of the dynamic trigger threshold.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the system according to any one of claims 1 to 9.

Images