A probabilistic tsunami early warning system based on data analysis

Abstract

This invention relates to a probabilistic tsunami early warning system based on data analysis. The system includes: a seismic data acquisition unit for real-time acquisition of monitoring information transmitted from various seismic stations in the target area; a seismic data preprocessing unit for preprocessing the monitoring information; a hypocenter location determination unit for acquiring multiple sets of estimated hypocenter locations, combinations of magnitudes corresponding to each estimated hypocenter location, and probability values ​​for each combination; a tsunami early warning analysis unit for obtaining the tsunami time series and water level station time series corresponding to each combination along the coast, and obtaining preliminary tsunami early warning results; and a tsunami parameter analysis unit for correcting the preliminary tsunami early warning results based on water level data, obtaining the final tsunami early warning result for publication. This tsunami early warning system, based on limited data after an earthquake, quickly provides the tsunami hazard distribution of the predicted coastal section, thereby facilitating decision-makers to formulate defense plans based on different probabilities.

Description

Technical Field

[0001] This invention belongs to the field of computer technology, and in particular relates to a probabilistic tsunami early warning system based on data analysis. Background Technology

[0002] Tsunamis are extremely destructive disasters. Since 80% of historical tsunami disasters are caused by undersea earthquakes, how to quickly predict the damage caused by earthquakes and tsunamis along coastlines has become a global hot topic. The main principle behind tsunami disaster prediction is that the waves generated by undersea earthquakes (the source of tsunamis) travel faster than tsunami waves.

[0003] Existing tsunami warning systems mostly rely on single source parameters for their warnings. However, in the short period following an earthquake, there are multiple possibilities for source parameters. It is necessary to deduce the possible combinations of source parameters based on limited data and quickly issue tsunami warning information to the public and decision-making departments. There is an urgent need for an intelligent tsunami warning system to accomplish the above tasks. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the aforementioned problems in the prior art, this invention provides a probabilistic tsunami early warning system based on data analysis.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0008] This invention provides a probabilistic tsunami early warning system based on data analysis, comprising:

[0009] The seismic data acquisition unit is used to acquire monitoring information transmitted by various seismic stations in the target area in real time. The monitoring information of each seismic station includes: basic information of the seismic station and monitored seismic wave information.

[0010] The seismic data preprocessing unit is used to preprocess the monitoring information transmitted by seismic stations in real time and obtain preprocessed early warning data.

[0011] The source location determination unit is used to obtain multiple sets of estimated source locations, combinations of magnitudes corresponding to each estimated source location, and probability values ​​for each combination based on the preprocessed early warning data.

[0012] The tsunami warning analysis unit is used to simulate and calculate each combination based on the tsunami numerical model, obtain the tsunami time series of the coastal output points and the time series of water level stations corresponding to each combination; and perform statistical analysis on the time series of all combinations to obtain preliminary results of tsunami warning.

[0013] The tsunami parameter analysis unit is used to acquire water level data from various water level monitoring points in the target area, and to revise the preliminary tsunami warning results based on the water level data to obtain updated tsunami warning results.

[0014] Optionally, the basic information of the seismic station includes: the longitude, latitude, and depth of the seismic station;

[0015] The monitored seismic wave information includes: whether seismic wave information was received, and the arrival time and waveform information of the P-wave and S-wave when seismic wave information was received.

[0016] Optionally, the seismic data preprocessing unit is specifically used for:

[0017] Data without seismic wave information in the monitoring information is removed, and abnormal information in the seismic wave information is removed to obtain preprocessed early warning data;

[0018] Specifically, removing outliers from seismic wave information includes:

[0019] Group any two station data points from the monitoring information to obtain... Group 1; where n is the total number of selected data;

[0020] For each first group, calculate the anomaly degree E, which reflects the anomalies of the transverse and longitudinal waves. f =max{E f (S),E f (P)},

[0021] Anomaly degree E f All first groups exceeding the abnormal threshold are designated as abnormal first groups;

[0022] If there is data that simultaneously belongs to the preset number of abnormal first groups, then the monitoring information will be removed;

[0023] Where f is the identifier of the first group, E f (S) represents the shear wave anomaly of the f-th first group, E f (P) represents the P-wave anomaly of the f-th first group. Let f be the longitude of the seismic monitoring station corresponding to one of the selected data points in the first group. Let f be the longitude of the seismic monitoring station corresponding to the other selected data in the first group. Let f be the latitude of the seismic monitoring station corresponding to one of the selected data points in the first group. Let f be the latitude of the seismic monitoring station corresponding to the other selected data in the first group. The depth of the seismic monitoring station corresponding to one of the selected data in the f-th first group. The depth of the seismic monitoring station corresponding to the other selected data in the f-th first group. The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to one of the selected data in the f-th first group is [the time when the seismic wave was detected]. The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to the other selected data in the f-th first group is [the time when the seismic wave was detected]. The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to one of the selected data in the f-th first group. v represents the time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the other selected data in the f-th first group. s v is the wave velocity of the transverse wave in a seismic wave. p This represents the wave velocity of the longitudinal wave in a seismic wave.

[0024] Optionally, the source location determination element is specifically used for:

[0025] The estimated location of the earthquake source is obtained from the preprocessed early warning data;

[0026] Specifically, four data points from the preprocessed early warning data are grouped together sequentially to obtain... The second group, where k is the total number of the early warning data to be processed;

[0027] For each second group, the estimated location of the corresponding seismic source is calculated using the following system of equations:

[0028]

[0029] Where g is the identifier for the second group. Let be the longitude of the estimated location of the earthquake source corresponding to the g-th second group. Let g be the latitude of the estimated location of the earthquake source corresponding to the g-th second group. The depth of the estimated location of the seismic source corresponding to the g-th second group. Let g be the longitude of the earthquake monitoring station corresponding to one of the unprocessed early warning data in the g-th second group. The longitude of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group is given. The longitude of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group is given. The longitude of the earthquake monitoring station corresponding to the last unprocessed early warning data in the g-th second group. Let be the latitude of the seismic monitoring station corresponding to one of the unprocessed early warning data in the g-th second group. Let be the latitude of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group. Let be the latitude of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group. Let be the latitude of the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group. Let be the depth of the seismic monitoring station corresponding to a single early warning data point in the g-th second group. The depth of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group. The depth of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group. Let be the depth of the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group. Let be the distance between the earthquake monitoring station and the epicenter corresponding to one of the unprocessed early warning data in the g-th second group. Let be the distance between the earthquake monitoring station and the epicenter corresponding to the second unprocessed early warning data in the g-th second group. Let be the distance between the earthquake monitoring station and the epicenter corresponding to the third unprocessed early warning data in the g-th second group. denoted as the distance between the earthquake monitoring station and the epicenter corresponding to the last pending early warning data in the g-th second group. The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to one of the pending early warning data in the g-th second group is denoted as . The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to one of the pending early warning data in the g-th second group is given. The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to the second unprocessed early warning data in the g-th second group is denoted as . The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group is given. The time when the shear wave of the seismic wave was detected by the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group is given. The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group is given. The time when the shear wave of the seismic wave was detected by the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group is given. v represents the time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the last pending early warning data in the g-th second group. s v is the wave velocity of the shear waves in the seismic region. p Let V be the wave velocity of the P-wave of the seismic waves within the region. This is the square root of the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to one of the unprocessed early warning data in the g-th second group. It is the square root of the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group. It is the square root of the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group. It is the square root of the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group.

[0030] Optionally, the source location determination element is specifically used for:

[0031] Based on the obtained estimated source locations, calculate the difference E between the theoretical P-wave arrival time at the g-th estimated source location (i.e., the earthquake location result) and the actual P-wave arrival time recorded at that station. gi ;

[0032] The probability P of the estimated location of each seismic source is determined based on the travel time difference between each estimated location and all stations within the target area. g ;

[0033] Furthermore, based on the waveform information in the seismic wave data, the magnitude of each seismic station is estimated, and statistical analysis is performed on the magnitudes of all seismic stations to obtain the probability P of the magnitude. m ;

[0034] By combining the estimated location and magnitude of each earthquake source, multiple combinations of estimated location and magnitude are obtained, i.e., earthquake parameter combinations. The probability value P of each combination is then obtained. gm .

[0035] Optionally, P gm =P g P m ;

[0036] Among them, t i (P) represents the arrival time of the P-wave recorded by the i-th station. Let v be the distance from the g-th earthquake location result to the i-th station. p This represents the P-wave velocity of the seismic waves within the region.

[0037] N m Let m be the number of stations with magnitude m, and N be the total number of seismic stations.

[0038] Optionally, the tsunami warning analysis unit is specifically used for:

[0039] The database stores the tsunami time series of all pre-acquired earthquake parameter combinations at the forecast shoreline output points and tsunami water level recording stations;

[0040] The earthquake parameter combinations obtained from the earthquake location determination unit are matched with those in the database to obtain the tsunami time series and water level station time series of each output point along the coast corresponding to each combination obtained from the earthquake location determination unit.

[0041] By performing statistical analysis on the tsunami time series and water level station time series at each output point, we can obtain tsunami time series with different probabilities and weighted average tsunami time series.

[0042] The time series of tsunamis with different probabilities and the weighted average time series of tsunamis are used as the preliminary results for tsunami warning.

[0043] Optionally, the tsunami parameter analysis unit is specifically used for:

[0044] Obtain water level data around each seismic station in the target area;

[0045] The water level data is preprocessed.

[0046] Traverse the preprocessed water level data and determine the mean value of the preprocessed water level data included in each data window. The data window includes m preprocessed water level data, where m is the maximum number of water level data in the preset window.

[0047] The maximum mean is determined as the maximum amplitude of the tsunami, and the start time of the window corresponding to the maximum mean is determined as the arrival time of the tsunami.

[0048] The preliminary tsunami warning results are reweighted based on the maximum tsunami amplitude and the tsunami amplitude near the time of tsunami arrival, and the final tsunami warning result is selected.

[0049] Optionally, the tsunami parameter analysis unit is specifically used for:

[0050] For any water level station i with recorded data, take (m / 2-1) points before and after the station with the maximum tsunami wave amplitude as the center, and obtain a time series ts of length m. i ;

[0051] For any combination of source parameters j, i.e., the combination of seismic parameters j obtained in the seismic location determination unit, taking (m / 2-1) points before and after the maximum tsunami amplitude as the center, a time series ts of length m is obtained. j And determine whether the following conditions are met:

[0052] ∑(ts i -ts j )<ε

[0053] Where ε is a preset value. Based on the distribution information of the warning area and water level stations, if the conditions are met, the combination of source parameters is retained; if the conditions are not met, the combination of source parameters is removed and the weights of each group of source parameters are recalculated to correct the tsunami time series with different probabilities and the weighted average tsunami time series, so as to obtain the final tsunami warning result.

[0054] Optionally, the system further includes:

[0055] The publishing unit is used to generate and publish tsunami forecast reports based on tsunami warning results and predefined tsunami forecast templates.

[0056] (III) Beneficial Effects

[0057] The probabilistic tsunami early warning system based on data analysis of the present invention can quickly determine the possible range and probability of the impact of a tsunami based on a limited amount of data and information after an earthquake, and automatically generate tsunami early warning information based on the analysis results and release it to the public, thus ensuring the timeliness of tsunami early warning while providing decision-makers with a theoretical basis for evacuation. Attached Figure Description

[0058] Figure 1 A schematic diagram of the structure of a probabilistic tsunami early warning system based on data analysis provided by the present invention;

[0059] Figure 2 This is a schematic diagram of another probabilistic tsunami early warning system based on data analysis provided by the present invention. Detailed Implementation

[0060] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0061] Example 1

[0062] like Figure 1 As shown, the present invention provides a probabilistic tsunami warning system based on data analysis. This tsunami warning system can be located in any computing device and specifically includes the following units:

[0063] The seismic data acquisition unit is used to acquire monitoring information transmitted by various seismic stations in the target area in real time. The monitoring information of each seismic station includes: basic information of the seismic station and monitored seismic wave information. For example, the basic information of the seismic station includes: longitude, latitude and depth of the seismic station; the monitored seismic wave information includes: whether seismic wave information has been received, and the arrival time and waveform information of P wave and S wave when seismic wave information is received.

[0064] The seismic data preprocessing unit is used to preprocess the monitoring information transmitted by seismic stations in real time and obtain preprocessed early warning data.

[0065] The source location determination unit is used to obtain multiple sets of estimated source locations, combinations of magnitudes corresponding to each estimated source location, and probability values ​​for each combination based on the preprocessed early warning data.

[0066] The tsunami warning analysis unit is used to simulate and calculate each combination based on the tsunami numerical model, obtain the tsunami time series of the coastal output points and the time series of water level stations corresponding to each combination; and perform statistical analysis on the time series of all combinations to obtain preliminary results of tsunami warning.

[0067] The tsunami parameter analysis unit is used to acquire water level data from various water level monitoring points in the target area, and to revise the preliminary tsunami warning results based on the water level data to obtain updated tsunami warning results.

[0068] In addition, the system in this embodiment also includes a publishing unit; the publishing unit is specifically used to generate and publish a tsunami forecast based on the tsunami warning results and a predefined tsunami forecast template.

[0069] The probabilistic tsunami warning system in this embodiment is based on a limited amount of data and information after an earthquake. It quickly determines the possible range and probability of a tsunami's impact and automatically generates tsunami warning information based on the analysis results and releases it to the public. This ensures the timeliness of tsunami warnings while providing decision-makers with a theoretical basis for evacuation.

[0070] Example 2

[0071] To better understand the solution of Embodiment 1 above, the solution of Embodiment 1 above will be described in detail below with reference to the accompanying drawings, and the function of each unit will be explained in detail.

[0072] The seismic data acquisition unit is specifically used to acquire monitoring information transmitted by various seismic stations in the target area in real time. The monitoring information of each seismic station includes: basic information of the seismic station and monitored seismic wave information.

[0073] Understandably, after an undersea earthquake, due to the rapid propagation speed of seismic waves, multiple nearby seismic monitoring stations (i.e., seismic stations) will be able to quickly receive the seismic waves, including S-waves (transverse waves) and P-waves (longitudinal waves). Seismic stations can automatically pick up seismic wave information based on historical waveforms using pre-trained machine learning models. The seismic data acquisition unit then receives basic information from multiple seismic stations, as well as the picked / monitored seismic wave information, transmitted by the seismic stations.

[0074] The basic information for any seismic station (such as seismic monitoring station b) includes: the longitude of any seismic monitoring station (such as the longitude x of seismic monitoring station b). b ), the latitude of any seismic monitoring station (e.g., the latitude y of seismic monitoring station b) b ) and the depth of any seismic monitoring station (e.g., the depth z of seismic monitoring station b) b ).

[0075] The seismic wave information monitored by any seismic monitoring station includes whether seismic wave information was detected.

[0076] If seismic wave information is detected, the seismic wave information detected by any seismic monitoring station also includes: the time when the shear wave of the seismic wave was detected (e.g., the time t when the shear wave of the seismic wave was detected by seismic monitoring station b). b (S)), the time when the P-wave of the seismic wave was detected (e.g., the time t when the P-wave of the seismic wave was detected by seismic monitoring station b). b (P)).

[0077] In practice, the seismic data acquisition unit continuously acquires basic information from seismic monitoring stations in the target area and monitored seismic wave information to ensure the comprehensiveness and timeliness of tsunami warning data.

[0078] The earthquake data preprocessing unit is specifically used to preprocess the monitoring information transmitted by earthquake stations in real time and obtain preprocessed early warning data.

[0079] In this embodiment, the seismic data preprocessing unit is a unit that continuously acquires data. Whenever data is acquired from the seismic data acquisition unit, the newly acquired data is immediately summarized and analyzed with the currently acquired data to effectively remove outliers and ensure the accuracy and reliability of subsequent early warning analysis.

[0080] In this embodiment, data without seismic wave information is removed from the monitoring information, and abnormal information in the seismic wave information is removed to obtain preprocessed early warning data, which is the early warning data to be processed.

[0081] The selection process begins by choosing monitoring data that includes earthquake data, specifically data containing monitored seismic wave information. Since tsunamis are generally caused by earthquakes, data without seismic data is useless for tsunami warnings because it does not cause earthquakes. The second step is to select data without obvious anomalies. Anomalies here are determined by the difference between the calculated seismic wave velocity and the velocity in the data. If this difference is particularly large, it indicates that the data values ​​are abnormal. The specific outlier removal process for the selected data is as follows:

[0082] 1.1, Group any two selected data points together to obtain... Group 1.

[0083] Where n is the total number of selected data.

[0084] For example, if the selected data are data 11, data 12, data 13, and data 14, then n = 4, and the first group has a total of 6 groups, namely {data 11, data 12}, {data 11, data 13}, {data 11, data 14}, {data 12, data 13}, {data 12, data 14}, and {data 13, data 14}.

[0085] Each data point includes the longitude of the corresponding seismic station (e.g., the longitude x of seismic station b). b The latitude of the corresponding seismic station (e.g., the latitude y of seismic station b). b The depth of the corresponding seismic station (e.g., the depth z of seismic station b). b ), the time and velocity of the shear waves detected by seismic station b (e.g., the time t of the shear waves detected by seismic station b). b (S)), the time and velocity of the P-wave detected by the seismic station b (e.g., the time t of the P-wave detected by the seismic station b). b (P)).

[0086] 1.2 For each first group, calculate the anomaly degree E. f =max{E f (S),E f (P)}.

[0087] Where f is the identifier of the first group.

[0088] E f (S) represents the shear wave anomaly of the f-th first group.

[0089]

[0090] E f (P) represents the P-wave anomaly of the f-th first group.

[0091]

[0092] Let f be the longitude of the seismic monitoring station corresponding to one of the selected data points in the first group. The longitude of the seismic monitoring station corresponding to the other selected data in the f-th first group. Let f be the latitude of the seismic monitoring station corresponding to one of the selected data points in the first group. The latitude of the seismic monitoring station corresponding to the other selected data in the f-th first group. The depth of the seismic monitoring station corresponding to one of the selected data in the f-th first group. The depth of the seismic monitoring station corresponding to the other selected data in the f-th first group.

[0093] The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to one of the selected data in the f-th first group is [the time when the seismic wave was detected]. The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to the other selected data in the f-th first group.

[0094] The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to one of the selected data in the f-th first group. v represents the time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the other selected data in the f-th first group. s v is the wave velocity of the shear waves in the seismic region. p This represents the P-wave velocity of the seismic waves within the region.

[0095] Each group contains two data points: data point 1 and data point 2. Data point 1 is labeled with a superscript number of 1, and data point 2 is labeled with a superscript number of 2. This embodiment does not specify which data point is chosen as data point 1.

[0096] For example, in the first group {data 11, data 12}, if data 11 is data number 1, then if data 12 is data number 2, then the longitude of the corresponding seismic monitoring station in data 11 is... The latitude of the corresponding earthquake monitoring station is The depth of the corresponding seismic monitoring station is The time when the shear wave of the seismic wave was detected was Time of detection of P-waves in seismic waves The longitude of the corresponding earthquake monitoring station in Data 12 is The latitude of the corresponding earthquake monitoring station is The depth of the corresponding seismic monitoring station is The time when the shear wave of the seismic wave was detected was

[0097] If data 12 is data number 1, then if data 11 is data number 2, then the longitude of the corresponding seismic monitoring station in data 12 is... The latitude of the corresponding earthquake monitoring station is The depth of the corresponding seismic monitoring station is The time when the shear wave of the seismic wave was detected was Time of detection of P-waves in seismic waves The wave velocity of the P-wave detected in the seismic wave was: The longitude of the corresponding earthquake monitoring station in Data 11 is The latitude of the corresponding earthquake monitoring station is The depth of the corresponding seismic monitoring station is Time of detection of P-waves in seismic waves

[0098] in, It is the distance between the seismic monitoring stations corresponding to the two data points in the first group. It refers to the time when the seismic monitoring station detected the seismic shear wave corresponding to the two data points in the first group. It is the propagation speed of the seismic shear wave between the seismic monitoring stations corresponding to the two data points in the first group. It is also the wave velocity of the shear wave calculated based on the correlation data values ​​of the two data points. This is the difference between the calculated shear wave velocity and the estimated shear wave velocity in the region. Ideally, this value should be 1, at which point E f (S) = 0 indicates that the data values ​​of the two data points in the first group are very accurate. The greater the distance between this ratio and 1 (it can be a positive distance, i.e., greater than 1, in which case the calculated shear wave velocity is greater than the estimated shear wave velocity, or it can be a negative distance, i.e., less than 1, in which case the calculated shear wave velocity is less than the estimated shear wave velocity), the greater E. f The larger the value of (S), the more abnormal the data.

[0099] Similarly, E f (P) represents the degree of anomaly in the P-wave velocity from the perspective of the P-wave angle, E f The larger the value of (S), the more abnormal the data.

[0100] E f =max{E f (S),E f (P)}, which is E f This reflects the most abnormal wave velocity conditions in both transverse and longitudinal waves.

[0101] 1.3, the anomaly degree E f All first groups exceeding the abnormal threshold are designated as the abnormal first group.

[0102] Ideally, E f (S) and E f (P) are all small values ​​close to 0, but due to the sensitivity and accuracy of the monitoring equipment, as well as various factors in the propagation process of seismic waves, E can be affected. f (S)>0 and / or E f (P)>0, which means E f >0. Typically, a normal range of discrepancies, i.e., an anomaly threshold, is set based on the distribution of stations and geological structures in the region. If E f If the deviation is less than or equal to the outlier threshold, it is considered acceptable, and the two data points in the first group are still considered accurate. f If the deviation exceeds the abnormal threshold, it is considered unacceptable, and two data points in this group are considered abnormal.

[0103] 1.4 If there is data that simultaneously belongs to the preset number of abnormal first groups, then remove it.

[0104] Step 1.3 yields the first group of all anomalies, each containing two data points. At this point, it's impossible to determine which data point is the anomaly. However, if a data point appears in a preset number of first anomaly groups, then it is considered an anomaly.

[0105] For example, if the preset quantity is 3, then if a data point appears in all three anomalous groups, then that data is considered anomalous. In other words, the monitoring data from the seismic station that appears most frequently (or has a frequency greater than a specified value) in the anomalous group is considered anomalous data.

[0106] Abnormal data is removed from the warning data to be processed using the above method, resulting in the warning data to be processed.

[0107] The unprocessed early warning data in this embodiment does not include invalid data (i.e., data without earthquake data) or abnormal data. Therefore, tsunami early warning can be made using unprocessed early warning data to ensure the accuracy of the final early warning.

[0108] The source location determination unit is specifically used to obtain multiple sets of estimated source locations, combinations of magnitudes corresponding to each estimated source location, and probability values ​​for each combination, based on the preprocessed early warning data.

[0109] The source location determination unit in this embodiment is also a continuously calculating and processing unit. Whenever data transmitted by the earthquake data preprocessing unit is acquired, the newly acquired data is immediately summarized and analyzed with the currently acquired data to determine the source location. Alternatively, the source location determination unit can read the latest data acquired by the earthquake data preprocessing unit within a certain period of time, and then summarize the latest data with the currently acquired data to determine the source location.

[0110] The specific process may include:

[0111] 2.1 Calculate the estimated location of the earthquake source, i.e., the earthquake location result, based on the early warning data to be processed.

[0112] Group any four data points from the pending early warning data in sequence to obtain The second group, where k is the total number of the early warning data to be processed;

[0113] In other words, the prerequisite for executing step 2.1 is that there must be more than or equal to 4 warning data to be processed. If there are fewer than 4 warning data to be processed, you can wait for a period of time (such as 1 minute) before determining whether the number of warning data to be processed is greater than or equal to 4. This is because the seismic data preprocessing unit continuously acquires data, and the source location determination unit will add the latest data acquired by the warning data acquisition unit to the currently acquired data for processing. If the current number of warning data to be processed is less than 4, then as the seismic data preprocessing unit continues to acquire data, the condition that the number of warning data to be processed is greater than or equal to 4 may be met in the future, and then step 2.1 will be executed.

[0114] In addition, even if the number of current pending early warning data is greater than or equal to 4, after the earthquake data preprocessing unit obtains the latest data, the source location determination unit can add it to the current pending early warning data and then execute step 2.1. That is to say, the source location determination unit will repeatedly determine the source location based on the data obtained by the earthquake data preprocessing unit, and then continuously update the source location until it obtains a source location that is considered accurate.

[0115] Additionally, in step 2.1, the warning data to be processed will be divided into... The second group.

[0116] For example, there are 6 data points to be processed for early warning, namely data 21, data 22, data 23, data 24, data 25, and data 26, in which case k = 6.

[0117] The corresponding second group has 15 elements, namely {data 21, data 22, data 23, data 24}, {data 21, data 22, data 23, data 25}, {data 21, data 22, data 23, data 26}, {data 21, data 22, data 24, data 25}, {data 21, data 22, data 24, data 26}, {data 21, data 22, data 25, data 26}, {data 21, data 23, data 24, data 25}, {data {21, data 23, data 24, data 26}, {data 21, data 23, data 25, data 26}, {data 21, data 24, data 25, data 26}, {data 22, data 23, data 24, data 25}, {data 22, data 23, data 24, data 26}, {data 22, data 23, data 25, data 26}, {data 22, data 24, data 25, data 26}, {data 23, data 24, data 25, data 26}.

[0118] 2.2 For each second group, the estimated location of the corresponding seismic source is calculated using the following set of equations:

[0119]

[0120] Where g is the identifier for the second group.

[0121] Let be the longitude of the estimated location of the earthquake source corresponding to the g-th second group. Let g be the latitude of the estimated location of the earthquake source corresponding to the g-th second group. The depth is the estimated location of the seismic source corresponding to the g-th second group.

[0122] h g h represents the number of pending warning data in the g-th second group, such as for group {data21, data22, data23, data24}. g =4.

[0123] Let g be the longitude of the earthquake monitoring station corresponding to one of the unprocessed early warning data in the g-th second group. The longitude of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group is given. The longitude of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group is given. This refers to the longitude of the earthquake monitoring station corresponding to the last unprocessed early warning data in the g-th second group.

[0124] Let be the latitude of the seismic monitoring station corresponding to one of the unprocessed early warning data in the g-th second group. Let be the latitude of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group. Let be the latitude of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group. is the latitude of the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group.

[0125] Let be the depth of the seismic monitoring station corresponding to a single early warning data point in the g-th second group. The depth of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group. The depth of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group. The depth of the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group.

[0126] Let be the distance between the earthquake monitoring station and the epicenter corresponding to one of the unprocessed early warning data in the g-th second group. Let be the distance between the earthquake monitoring station and the epicenter corresponding to the second unprocessed early warning data in the g-th second group. Let be the distance between the earthquake monitoring station and the epicenter corresponding to the third unprocessed early warning data in the g-th second group. denoted as the distance between the earthquake monitoring station and the earthquake source corresponding to the last pending early warning data in the g-th second group.

[0127] The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to one of the pending early warning data in the g-th second group is denoted as . The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to one of the pending early warning data in the g-th second group is given. The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to the second unprocessed early warning data in the g-th second group is denoted as . The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group is given. The time when the shear wave of the seismic wave was detected by the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group is given. The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group is given. The time when the shear wave of the seismic wave was detected by the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group is given. The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the last pending early warning data in the g-th second group.

[0128] Let be the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to one of the unprocessed early warning data in the g-th second group. Let be the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the second unprocessed early warning data in the g-th second group. Let be the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the third unprocessed early warning data in the g-th second group. It is the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the last unprocessed early warning data in the g-th second group.

[0129] 2.3 Based on the above grouping calculations, there are multiple sets of earthquake location results (e.g., 15 sets of estimated source locations in the example above). Calculate the difference E between the theoretical P-wave arrival time of the g-th earthquake location result at the i-th station and the actual P-wave arrival time recorded at that station. gi ;

[0130]

[0131] Among them, t i (P) represents the arrival time of the P-wave recorded by the i-th station. Let v be the distance from the g-th earthquake location result to the i-th station. p This represents the P-wave velocity of the seismic waves within the region.

[0132] 2.4 Calculate the probability of each seismic location result based on the travel time difference between each location result and all stations, i.e.:

[0133]

[0134] 2.5 Simultaneously, based on the waveform data from each seismic station, the magnitude of each station is calculated. N stations yield N magnitudes. These magnitudes are grouped together, and the probability of each magnitude is further statistically analyzed.

[0135]

[0136] Where N m Let m be the number of stations with magnitude m, and N be the total number of stations.

[0137] 2.6. Each earthquake location result is combined with the magnitude to obtain multiple combinations of earthquake location results and magnitudes. For example, in the above example, there are 15 sets of earthquake location results. Assuming there are 3 magnitudes, there are a total of 45 combinations of earthquake parameters. The probability value P for each combination... gmP is the product of the probability of the corresponding location result and the probability of the magnitude: gm =P g P m .

[0138] The tsunami warning analysis unit is specifically used to simulate and calculate each combination based on the tsunami numerical model, obtain the tsunami time series of the coastal output points and the time series of water level stations corresponding to each combination; and perform statistical analysis on the time series of all combinations to obtain preliminary results of tsunami warning.

[0139] In this embodiment, the database pre-stores the sequence corresponding to each simulated combination of earthquake parameters, namely the tsunami time series at the predicted shoreline output point and tsunami level recording station. That is, the database pre-estimates the tsunami time series at the predicted shoreline output point and tsunami level recording station for each possible combination of earthquake parameters using known methods.

[0140] That is, before an earthquake occurs, all possible combinations of earthquake parameters (source location and magnitude) are analyzed based on the results of historical earthquake and geological structure research. The above combinations of earthquake parameters are simulated and calculated using a tsunami numerical model to obtain the tsunami time series of the forecast shoreline output point and surrounding tsunami water level recording stations.

[0141] At this point, the combinations of seismic parameters obtained in the seismic location determination unit (45 sets in the example above) are compared one by one with the combinations of seismic parameters in the database. The numerical simulation results (i.e., time series) corresponding to the closest combination of seismic parameters are determined as the numerical simulation results of each seismic parameter in the current seismic location determination unit.

[0142] The numerical simulation results include: the curves of tsunami amplitude changing with time at each output point on the simulated tsunami warning shoreline (tsunami time series) and the tsunami time series of the simulated water level recording station; the probability of each numerical simulation is the probability of that set of earthquake parameters.

[0143] By performing statistical analysis on the set of tsunami amplitude sequences composed of multiple tsunami time series at each forecast point, we can obtain tsunami time series with different probabilities and weighted average tsunami time series, and thus obtain preliminary results for tsunami warning.

[0144] The preliminary results of the aforementioned tsunami warning are based on simulations derived from seismic wave information. These results may contain biases that could lead to false alarms or other unreliable information. Therefore, further revisions to the preliminary results of the tsunami warning are necessary to improve the reliability and accuracy of the warnings. Through verification or revision, the accuracy and reliability of the data have been ensured.

[0145] The tsunami parameter analysis unit is specifically used to acquire water level data from various water level monitoring points / stations in the target area, and to revise the preliminary tsunami warning results based on the water level data to obtain updated tsunami warning results.

[0146] In this embodiment, the water level data monitored by the water level detection point includes: the longitude and latitude of the water level observation station, whether tsunami information was detected, and the time series of the tsunami.

[0147] In one possible line-of-sight approach, the tsunami parameter analysis unit can be specifically used for:

[0148] Obtain water level data around each seismic station in the target area;

[0149] The above water level data is preprocessed, including outlier handling, missing value handling, and bandpass filtering.

[0150] Traverse the preprocessed water level data and determine the mean value of the preprocessed water level data included in each data window. The data window includes m preprocessed water level data, where m is the maximum number of water level data in the preset window.

[0151] The maximum mean was determined as the maximum amplitude of the tsunami, and the start time of the window corresponding to the maximum mean was determined as the arrival time of the tsunami.

[0152] The earthquake parameters were reweighted based on the maximum tsunami amplitude and the tsunami amplitude near the arrival time of the tsunami, and the tsunami simulation results were selected as the final tsunami warning results.

[0153] In another possible line-of-sight approach, the tsunami parameter analysis unit can be specifically used for:

[0154] For any water level station i that has a recorded data, take (m / 2-1) points before and after the maximum tsunami wave amplitude to obtain a time series ts of length m. i ;

[0155] For any combination of source parameters j, taking the maximum tsunami amplitude generated at that water level station as the center, and taking (m / 2-1) points before and after it respectively, we obtain a time series ts of length m. j And determine whether the following conditions are met:

[0156] ∑(ts i -ts j )<ε

[0157] Where ε is a preset value. Based on the individually determined warning area and water level station distribution information, if the conditions are met, the combination of source parameters is retained; if the conditions are not met, the combination of source parameters is removed and the weights of each group of source parameters are recalculated. This further corrects the tsunami time series with different probabilities and the weighted average tsunami time series, and obtains the final tsunami warning result.

[0158] The aforementioned system uses pre-arriving seismic wave information from the current tsunami earthquake to perform data analysis, obtain the source location results and magnitude information, and then infer the possible affected area and time series of the tsunami. At the same time, it uses the currently monitored water level data to correct the possible affected area and time series of the tsunami, ensuring the accuracy and reliability of the tsunami warning.

[0159] In the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0160] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A probabilistic tsunami warning system based on data analysis, characterized by, The system comprises: An earthquake data acquisition unit configured to acquire monitoring information transmitted by each seismic station in a target region in real time, wherein the monitoring information of each seismic station comprises: basic information of the seismic station and monitored seismic wave information; the basic information comprises: longitude, latitude and depth of the seismic monitoring station; the seismic wave information comprises: whether seismic wave information is monitored, and if so, the seismic wave information further comprises: shear wave time and longitudinal wave time of the monitored seismic wave; An earthquake data preprocessing unit configured to preprocess the monitoring information transmitted by the seismic station in real time to obtain preprocessed early warning data; the preprocessed early warning data does not include invalid data or abnormal data; the earthquake data preprocessing unit is a unit for continuously acquiring and analyzing data; specifically, abnormal data in the seismic wave information is removed, including: Grouping any two station data in the monitoring information, obtaining Group the first group; wherein, Total number of selected data for each first group, compute an abnormality degree reflecting the abnormality of the shear wave and the longitudinal wave , is the abnormality degree of the shear wave of the fth first group, is the longitudinal wave abnormality degree of the fth first group; abnormality All first groups exceeding the abnormal threshold are designated as abnormal first groups; If there are selected data belonging to a preset number of abnormal first groups at the same time, the monitoring information is removed; A seismic source position determination unit configured to acquire a plurality of groups of estimated positions of seismic sources, combinations of magnitudes corresponding to each estimated position of the seismic source, and probability values of each combination based on the preprocessed early warning data; the seismic source position determination unit is a unit for continuously calculating and processing; when data transmitted by the earthquake data preprocessing unit is acquired, the newly acquired data is analyzed together with the data already acquired to determine the position of the seismic source; the preprocessed early warning data needs to be greater than or equal to 4; Specifically, according to the acquired source estimated position, the difference between the theoretical time of the P wave of the first source estimated position, i.e., the earthquake positioning result, arriving at the first i station and the actually recorded P wave arrival time of the station is calculated E gi ; determining a likelihood of the source estimate location based on travel time differences between the source estimate location and all stations within the target region P g ; And, according to waveform information in the seismic wave information, magnitude of each seismic station is estimated, and statistical analysis is performed on the magnitude of all seismic stations to obtain the possibility of the magnitude P m ; Combining each source estimated position and magnitude, a plurality of combinations of source estimated position and magnitude, i.e. combinations of seismic parameters, are obtained, and a probability value of each combination is acquired P gm ; A tsunami early warning analysis unit configured to simulate and calculate each combination based on a tsunami numerical model to obtain a time sequence of a coastal output point tsunami and a time sequence of a water level station corresponding to each combination; and statistically analyze the time sequences of all combinations to obtain a preliminary result of the tsunami early warning; A tsunami parameter analysis unit configured to acquire water level data of each water level detection point in the target region, correct the preliminary result of the tsunami early warning based on the water level data, and obtain an updated result of the tsunami early warning.

2. The system of claim 1, wherein, The basic information of the seismic station comprises: longitude, latitude and depth of the seismic station; The monitored seismic wave information comprises: whether seismic wave information is received, arrival time and waveform information of P wave and S wave of the received seismic wave information.

3. The system of claim 2, wherein, The earthquake data preprocessing unit is specifically configured to: remove data without seismic wave information in the monitoring information, and remove abnormal information in the seismic wave information to obtain the preprocessed early warning data; For the first P-wave anomaly of the first group, ， , is the longitude of the seismic monitoring station corresponding to the selected data in the first group, is the longitude of the seismic monitoring station corresponding to the other selected data in the first group, is the latitude of the seismic monitoring station corresponding to the selected data in the first group, is the latitude of the seismic monitoring station corresponding to the other selected data in the first group, is the depth of the seismic monitoring station corresponding to the selected data in the first group, is the depth of the seismic monitoring station corresponding to the other selected data in the first group, is the time at which the seismic monitoring station corresponding to the selected data in the first group detected the shear wave of the seismic wave, is the time at which the seismic monitoring station corresponding to the other selected data in the first group detected the shear wave of the seismic wave, is the time at which the seismic monitoring station corresponding to the selected data in the first group detected the longitudinal wave of the seismic wave, is the time at which the seismic monitoring station corresponding to the other selected data in the first group detected the longitudinal wave of the seismic wave, is the wave velocity of the shear wave of the seismic wave, is the wave velocity of the longitudinal wave of the seismic wave.​​​​​​​​​​ 4. The system of claim 2, wherein, The seismic source position determination unit is specifically configured to: acquire estimated positions of seismic sources based on the preprocessed early warning data; Specifically, any 4 data in the preprocessed early warning data are sequentially grouped to obtain The second grouping, wherein, is the total number of the to-be-processed early warning data. for each second group, calculate the corresponding estimated position of the seismic source by using the following equation set: ； wherein, is an identification of a second group, is a longitude of an estimated location of a seismic source corresponding to the first second group, is a latitude of an estimated location of a seismic source corresponding to the first second group, is a depth of an estimated location of a seismic source corresponding to the first second group, is a longitude of a seismic monitoring station corresponding to a first to-be-processed early warning data in the first second group, is a longitude of a second to-be-processed early warning data in the first second group, is a longitude of a third to-be-processed early warning data in the first second group, is a longitude of a last to-be-processed early warning data in the first second group, is a latitude of a seismic monitoring station corresponding to a first to-be-processed early warning data in the first second group, is a latitude of a second to-be-processed early warning data in the first second group, is a latitude of a third to-be-processed early warning data in the first second group, is a latitude of a last to-be-processed early warning data in the first second group, is a depth of a seismic monitoring station corresponding to a first to-be-processed early warning data in the first second group, is a depth of a second to-be-processed early warning data in the first second group, is a depth of a third to-be-processed early warning data in the first second group, is a depth of a last to-be-processed early warning data in the first second group, is a distance between a seismic monitoring station corresponding to a first to-be-processed early warning data in the first second group and a seismic source, is a distance between a seismic monitoring station corresponding to a second to-be-processed early warning data in the first second group and a seismic source, is a distance between a seismic monitoring station corresponding to a third to-be-processed early warning data in the first The distance between the earthquake monitoring station and the epicenter corresponding to the third unprocessed early warning data in the second group. For the first The distance between the earthquake monitoring station and the epicenter corresponding to the last pending early warning data in the second group. For the first The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to one of the pending early warning data in the second group. For the first The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to one of the early warning data to be processed in the second group. For the first The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to the second unprocessed early warning data in the second group. For the first The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the second pending early warning data in the second group. For the first The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to the third unprocessed early warning data in the second group. For the first The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the third pending early warning data in the second group. For the first The time when the seismic monitoring station detected the shear wave of the seismic wave corresponding to the last pending early warning data in the second group. For the first The time when the P-wave of the seismic wave was detected by the seismic monitoring station corresponding to the last pending early warning data in the second group. Let V be the wave velocity of the transverse waves of the seismic waves within the region. Let V be the wave velocity of the P-wave of the seismic waves within the region. For the first The square root of the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to one of the early warning data to be processed in the second group. For the first The square root of the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the second unprocessed early warning data in the second group. For the first The square root of the sum of the squares of the longitude, latitude, and depth of the seismic monitoring station corresponding to the third unprocessed early warning data in the second group. For the first The square root of the sum of the squares of the longitude, latitude, and depth of the earthquake monitoring station corresponding to the last pending early warning data in the second group.

5. The system of claim 4, wherein ； ； ； ； in, For the first i When the P-wave arrived as recorded by each station For the first The earthquake location results are up to the first i The distance between each station The velocity of the P-wave in the seismic wave region; N m is the number of stations for an earthquake magnitude of m is the number of stations for an earthquake magnitude of N is the total number of stations.

6. The system of any one of claims 1 to 5, wherein, The tsunami early warning analysis unit is specifically configured to: the database stores time sequences of a tsunami at a predicted coastal output point and a tsunami water level recording station of all pre-acquired seismic parameter combinations; match the seismic parameter combinations acquired by the seismic position determination unit with the database to obtain the time sequence of the coastal output point tsunami and the time sequence of the water level station corresponding to each combination acquired by the seismic position determination unit at each output point along the coast. The tsunami time series of different probabilities and the weighted average tsunami time series can be obtained by statistical analysis of the tsunami time series of each output point and the time series of the water level station. The tsunami time series of different probabilities and the weighted average tsunami time series are used as the preliminary result of the tsunami warning.

7. The system of claim 6, wherein, The tsunami parameter analysis unit is specifically configured to: Obtain water level data around each seismic station in a target area; Preprocess the water level data, Determine the mean value of the preprocessed water level data included in each data window by traversing the preprocessed water level data, wherein the data window includes a predetermined maximum number of preprocessed water level data; Determine the maximum mean value as the maximum amplitude of the tsunami, and determine the starting time of the window corresponding to the maximum mean value as the arrival time of the tsunami; According to the obtained maximum amplitude of the tsunami and the amplitudes of the tsunami near the arrival time of the tsunami, reassign weights to the preliminary result of the tsunami warning and select the final result of the tsunami warning.

8. The system of claim 6, wherein, The tsunami parameter analysis unit is specifically configured to: For any water level station with records i , take (N / 2-1) points before and after the maximum tsunami wave amplitude, respectively, to obtain a time series of length m ts i ;​​ For any combination of source parameters j i.e. the combination of seismic parameters obtained in the seismic position determination unit j , taking (N m / 2-1) points before and after the maximum tsunami wave amplitude as the center, respectively, to obtain a time series of length m ts j ; and determining whether the following conditions are met:​ ； wherein is a preset value, based on the early warning area and water level station distribution information, if the condition is met, the source parameter combination is retained, if the condition is not met, the source parameter combination is rejected and the weight of each group of source parameters is recalculated, to correct the tsunami time series of different probabilities and the weighted average tsunami time series, to obtain the final tsunami warning result.

9. The system of claim 1, wherein, The system further includes: A publishing unit configured to generate and publish a tsunami forecast single according to the tsunami warning result and a predefined tsunami forecast template.

Images