An earthquake early warning station network abnormal real-time monitoring alarm method and positioning system

Abstract

The application relates to the technical field of earthquake early warning, and discloses an alarm method and a positioning system for abnormal real-time monitoring of an earthquake early warning station network, which comprises the following steps: obtaining M groups of original fluctuation signals based on M seat stations; extracting abnormal data of the original fluctuation signals, including data delay, server time service, network transmission flow and zero point drift; and triggering an alarm if the abnormal data is synchronously over-limit. The original fluctuation signals are obtained based on the M seat stations, four types of key abnormal data, including data delay, server time service, network transmission flow and zero point drift, are extracted, and alarm information is output when the abnormal data is synchronously over-limit, so that the abnormal conditions of the signal collection, transmission and processing links can be comprehensively covered, real-time capture of systematic signal abnormalities of the earthquake early warning system is realized, and the alarm can be triggered quickly.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of earthquake early warning, more particularly, it relates to an alarm method and positioning system for real-time monitoring of abnormality of an earthquake early warning station network. BACKGROUND

[0002] When earthquake monitoring is performed in regions with high mountains and valleys, dense faults and significant terrain undulations, the existing technology faces significant challenges. Traditional seismic source positioning methods highly depend on seismic wave travel time difference calculation and full-field velocity model. However, in such regions, mountain blockage can cause complex and variable seismic wave propagation paths, resulting in significant deviations in travel time difference calculation; at the same time, the near-surface velocity heterogeneity caused by terrain elevation can cause significant differences between the velocity model and the actual situation, further exacerbating positioning errors. SUMMARY

[0003] The present application provides an alarm method and positioning system for real-time monitoring of abnormality of an earthquake early warning station network, which solves the technical problems raised in the background art.

[0004] In a first aspect, an alarm method for real-time monitoring of abnormality of an earthquake early warning station network, comprising:

[0005] S1, obtaining M groups of original wave signals based on M seismic stations;

[0006] S2, extracting abnormal data of the original wave signals, including data delay, server time, network transmission traffic and zero point drift;

[0007] S3, if the abnormal data is synchronous and exceeds the limit, triggering an alarm.

[0008] In a second aspect, a positioning system for real-time monitoring of abnormality of an earthquake early warning station network, applied to the alarm method for real-time monitoring of abnormality of an earthquake early warning station network, and in response to an earthquake alarm, performing the following steps:

[0009] A data acquisition module for monitoring a target mountainous region based on M seismic stations to obtain M groups of original wave signals;

[0010] A data processing module for performing Kirchhoff phase migration processing on each group of original wave signals with a preset sliding window H to obtain the propagation direction and local wavefront curvature of each monitoring station;

[0011] A positioning estimation module for constructing a closed wavefront curvature field based on the local wavefront curvature of the stations corresponding to the outer ring monitoring, and sequentially estimating the wavefront estimation radius, the source plane estimation coordinates and the source center estimation depth of the closed wavefront curvature field;

[0012] The precise positioning module is configured to take the estimated coordinates of the source plane and the estimated depth of the source center as initial positions to construct a three-dimensional grid; for any grid in the three-dimensional grid, the source parameter residual of each grid in the three-dimensional grid is calculated according to the propagation direction and the local wavefront curvature of each monitoring station and in combination with a minimum variance unbiased estimation algorithm, so as to take the grid with the minimum source parameter residual as the final source position.

[0013] Further, the outer ring monitoring, the inner ring monitoring, the fault monitoring and the vertical monitoring include:

[0014] The outer ring monitoring includes: constructing a ring-shaped first monitoring array around the ridges of the target mountain;

[0015] The inner ring monitoring includes: constructing a ring-shaped second monitoring array around the basins of the target mountain;

[0016] The fault monitoring includes: constructing a linear third monitoring array at a fixed horizontal interval based on the active faults of the target mountain;

[0017] The vertical monitoring includes: constructing a linear fourth monitoring array at a fixed vertical interval from the ridges to the basins of the target mountain.

[0018] Further, the Kirchhoff phase migration processing is performed on each group of original fluctuation signals with a preset sliding window H to obtain the propagation direction and the local wavefront curvature of each monitoring station, including:

[0019] Each group of original fluctuation signals is segmented according to the preset sliding window H to obtain K sub-signals, and the mean value and the trend of each sub-signal are removed to obtain a standard fluctuation signal;

[0020] The local wavefront model is initialized as follows:

[0021] A three-dimensional model of the target mountain is established; the three-dimensional model is subjected to spherical fitting processing to obtain a local wavefront model, and the spherical center coordinates of the local wavefront model are determined;

[0022] The standard fluctuation signal is subjected to phase shift processing according to a preset time delay based on the local wavefront model and the spherical center coordinates, the prior P-wave velocity and the sampling interval of the standard fluctuation signal to obtain a migrated fluctuation signal;

[0023] The center coordinates of the ring-shaped first monitoring array are determined, and the center coordinates are connected to each monitoring station in the first monitoring array to form a candidate propagation direction set; M groups of migrated fluctuation signals are superimposed in each candidate propagation direction to determine the superimposed intensity of each candidate propagation direction, and the candidate propagation direction with the highest superimposed intensity is selected as the propagation direction;

[0024] A local coordinate system is constructed with the propagation direction as the normal direction, and the station coordinates of the M monitoring stations are fitted in the local coordinate system to obtain the local wavefront curvature of each monitoring station.

[0025] Further, a closed wavefront curvature field is constructed based on the local wavefront curvature of the corresponding station of the outer ring monitoring, including:

[0026] The geometric curve of the ridge of the target mountain is extracted to construct a monitoring outer ring.

[0027] A virtual monitoring station is inserted at the midpoint of adjacent monitoring stations in the monitoring outer ring, and the local wavefront curvature of the virtual station is the average of the local wavefront curvatures of the adjacent monitoring stations.

[0028] The virtual monitoring station is repeatedly inserted until the number of monitoring stations reaches a preset number threshold to form a local wavefront curvature set.

[0029] The local wavefront curvature set is fitted to obtain a curvature function as the closed wavefront curvature field.

[0030] Further, the closed wavefront curvature field is sequentially estimated to obtain a wavefront estimated radius, a focal plane estimated coordinate, and a focal center estimated depth, including:

[0031] The wavefront estimated radius is determined based on the ring-wise average curvature of the closed wavefront curvature field.

[0032] The wavefront estimated radius is loaded in the monitoring outer ring, and the focal plane estimated coordinate is obtained by minimizing the intersection point of the wavefront estimated radius.

[0033] The mean time of the time stamp of the original wave signal of the first monitoring array and the mean time of the time stamp of the original wave signal of the second monitoring array are determined respectively, and the time difference of the mean time is calculated.

[0034] The vertical propagation time is determined based on the time difference of the mean time.

[0035] The focal center estimated depth is calculated As follows:

[0036]

[0037] Wherein, represents the propagation speed of P wave, represents the vertical propagation time, represents the wavefront estimated radius, represents the average of the Euclidean distance between each monitoring station in the first monitoring array and the focal plane estimated coordinate.

[0038] Further, taking the focal plane estimated coordinate and the focal center estimated depth as initial positions, a three-dimensional grid is constructed, including:

[0039] A three-dimensional grid with a preset step length is constructed in a space range that extends a first preset distance east, west, south and north from the focal plane estimated coordinate and extends a second preset distance vertically upward and downward from the focal center estimated depth.

[0040] Further, according to the propagation direction and the local wave front curvature of each monitoring station, a focal parameter residual of each grid in the three-dimensional grid is calculated by using a minimum variance unbiased estimation algorithm, so that the grid with the minimum focal parameter residual is taken as the final focal position, including:

[0041] For any grid in the three-dimensional grid, a weight matrix is formed according to the observation variance of the propagation direction and the local wave front curvature of each monitoring station, and a focal parameter residual corresponding to the grid is calculated by using a minimum variance unbiased estimation algorithm, and the grid with the minimum residual is selected as the final focal position.

[0042] The beneficial effects of the present application are that:

[0043] 1. The original wave signal is obtained based on M stations, four types of key abnormal data of data delay, server time, network transmission flow and zero point drift are extracted, and an alarm is output when the abnormal data is synchronized and exceeds the limit, which can comprehensively cover the abnormal conditions of signal collection, transmission and processing links, realize real-time capture of earthquake systematic signal abnormalities, and quickly trigger an alarm.

[0044] 2. By constructing an outer ring, an inner ring, a fault and a vertical station monitoring network with mountainous terrain adaptability, and combining Kirchhoff phase migration processing under a sliding window to extract the propagation direction and the local wave front curvature, a whole process mechanism of closed wave front curvature field inversion of the focal position is established, which can stably estimate the plane position and the depth of the focal position without relying on the traditional travel time difference and the full field velocity model, and further realizes focal solution through a three-dimensional grid fine positioning process, and improves the reaction speed, positioning accuracy and reliability of the earthquake early warning system in a complex terrain area. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 is a flowchart of an alarm method for real-time monitoring of an earthquake early warning station network anomaly of the present application;

[0046] Figure 2 is a module diagram of a positioning system for real-time monitoring of an earthquake early warning station network anomaly of the present application. DETAILED DESCRIPTION

[0047] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that discussions of these implementations are merely provided to enable those skilled in the art to better understand so as to best use the subject matter described herein, and variations of elements can be made dependant on the elements function and arrangement without departing from the scope of the present specification. Various examples can omit, substitute, or add various procedures or components as appropriate, and the specifications of features described with respect to some examples can be combined in others. Additionally, features described with respect to some examples can be combined in other examples.

[0048] Embodiment one:

[0049] As shown in the figure, an alarm method for real-time monitoring of abnormality of a seismic early warning station network comprises: Figure 1

[0050] S1, obtaining M groups of original fluctuation signals based on M stations;

[0051] S2, extracting abnormal data of the original fluctuation signals, including data delay, server time setting, network transmission flow, and zero point drift;

[0052] S3, triggering an alarm if the abnormal data is synchronous and out of limits.

[0053] In detail, data delay refers to the time delay of the original fluctuation signal from the completion of collection by the station sensor to the transmission to the data processing center through the communication network. When an earthquake occurs, the strong seismic wave will interfere with the communication link, or the station transmission equipment will be temporarily overloaded due to vibration, resulting in a significant extension of the signal transmission time, which is out of the normal preset range, forming a data delay anomaly.

[0054] In detail, server time setting refers to the synchronization deviation of the local time of each monitoring station and the standard time of the center server. When an earthquake occurs, the violent vibration of the earth's crust may affect the time synchronization module of the station or the server, resulting in an increase in the deviation of the original fluctuation signal timestamp recorded by each station and the server standard time, which is out of the normal accuracy range, forming a server time setting anomaly.

[0055] In detail, network transmission flow refers to the amount of data transmitted per unit time and stability between the station and the data processing center. When an earthquake occurs, the station sensor will collect violent fluctuation signals (sudden change in amplitude and frequency), resulting in an instantaneous surge in data volume; at the same time, the earthquake may cause vibration failure of network equipment (such as routers and switches), resulting in transmission interruption or data loss, causing a sharp drop in flow. This abnormal fluctuation (surge or sharp drop) of flow is out of the normal range, which is a network transmission flow anomaly.

[0056] ​In detail, the zero-point drift indicates that the output signal of the seismic sensor deviates from the reference zero value when the seismic sensor is static (without obvious vibration). When an earthquake occurs, the strong ground vibration can cause the internal mechanical structure of the sensor (such as the pendulum, piezoelectric element) to temporarily deviate or the performance to be disturbed, causing the deviation to exceed the normal threshold, forming a zero-point drift anomaly, and affecting the accurate analysis of the original wave signal.

[0057] In detail, when the four types of abnormal data of data delay, server time service, network transmission flow and zero point all exceed their respective preset thresholds within the same period, the system determines that it is a systematic signal anomaly caused by seismic activity, and then outputs a seismic warning. Specifically, the generation of data delay, server time service deviation, network transmission flow fluctuation and zero-point drift is directly related to the occurrence of an earthquake: the crust vibration, electromagnetic interference or equipment affected by the earthquake caused by the earthquake can simultaneously affect the signal transmission efficiency, time synchronization accuracy, data transmission stability and sensor baseline. Synchronous overrun indicates that the four types of abnormalities are not caused by single device failure or accidental interference, but are systematic chain reactions caused by seismic activity. At this time, the alarm is triggered, which can accurately respond to a real earthquake event.

[0058] Embodiment two:

[0059] As shown in Figure 2 A positioning system for real-time monitoring of anomalies in a seismic early warning station network, applied to the method for real-time monitoring and alarming of anomalies in a seismic early warning station network, in response to a seismic warning, and performing the following steps:

[0060] A data acquisition module for acquiring M sets of original wave signals by performing outer ring monitoring, inner ring monitoring, fault monitoring and vertical monitoring on a target mountain based on M seismic stations;

[0061] A data processing module for performing Kirchhoff phase migration processing on each set of original wave signals with a preset sliding window H to obtain the propagation direction and local wavefront curvature of each monitoring station;

[0062] A positioning estimation module for constructing a closed wavefront curvature field based on the local wavefront curvature of the stations corresponding to the outer ring monitoring, and sequentially estimating the wavefront estimation radius, the focal plane estimation coordinates and the focal center estimation depth of the closed wavefront curvature field;

[0063] A precise positioning module for constructing a three-dimensional grid with the focal plane estimation coordinates and the focal center estimation depth as the initial position, calculating the focal parameter residual of each grid in the three-dimensional grid according to the propagation direction and local wavefront curvature of each monitoring station, and combining the least variance unbiased estimation algorithm to calculate the focal parameter residual of each grid in the three-dimensional grid, and taking the grid with the smallest focal parameter residual as the final focal position.

[0064] In an embodiment of the present application, the outer ring monitoring, the inner ring monitoring, the fault monitoring and the vertical monitoring include:

[0065] The outer ring monitoring includes: constructing a ring-shaped first monitoring array around the ridges of the target mountain;

[0066] The inner ring monitoring includes: constructing a ring-shaped second monitoring array around the basins of the target mountain;

[0067] The fault monitoring includes: constructing a linear third monitoring array at a fixed horizontal interval based on the active faults of the target mountain;

[0068] The vertical monitoring includes: constructing a linear fourth monitoring array at a fixed vertical interval from the ridges to the basins of the target mountain.

[0069] In detail, the ring-shaped monitoring array is constructed around the ridges of the target mountain, the ridgeline contour is extracted by the digital elevation model, the closed curve (monitoring outer ring) is fitted, and the detection station is deployed at an equal distance according to the average arc length (such as 8km). In the environment of high altitude of the ridge and shallow buried bedrock, the power spectrum of the environmental noise is smaller than that of the valley noise. The ring-shaped layout ensures that the seismic P wave in any incident direction is observed synchronously by multiple monitoring stations.

[0070] The ring-shaped array is constructed around the basins of the mountain basement, the basin boundary is fitted by combining the digital elevation model and remote sensing image, and the monitoring station is deployed at an equal distance according to the average arc length (such as 5km). The low-speed sedimentary layer of the basin amplifies the high-frequency component of the P wave.

[0071] Along the strike of the active fault of the target mountain, a linear array is constructed at a fixed horizontal interval (such as 2km); the fault central line is extracted from the geological map and the digital elevation model, and the monitoring station is equally divided after calculating the total length of the fault. The linear array can collect the travel time difference of the rupture front, the rupture velocity and the extension direction.

[0072] From the ridges to the basins, a vertical profile array is constructed at a fixed vertical interval (such as 0.625m). The elevation zone is divided by calculating the vertical difference from the ridges to the basins; the profile line perpendicular to the ridges is selected, and the coordinates of the monitoring stations corresponding to each elevation zone are extracted.

[0073] The mountain terrain is rugged, and a single array cannot take into account all-directional perception, spectrum coverage, fault tracking and vertical profiling. The spatial complementation and data redundancy of the original waveform data collected by the four-layer array are realized.

[0074] It should be noted that the original wave signal refers to the three-component (east-west, north-south, vertical) ground vibration time history data collected directly by the sensor of the monitoring station without processing.

[0075] In an embodiment of the present application, the Kirchhoff phase migration processing is performed on each group of original wave signals with a preset sliding window H to obtain the propagation direction and local wave front curvature of each monitoring station, including:

[0076] Each group of original wave signals is segmented according to the preset sliding window H to obtain K sub-signals, and the mean value and trend of each sub-signal are removed to obtain a standard wave signal;

[0077] The local wave front model is initialized as follows:

[0078] A three-dimensional model of the target mountain is established, the three-dimensional model is subjected to spherical fitting processing to obtain a local wave front model, and the spherical center coordinates of the local wave front model are determined;

[0079] Based on the local wave front model, the spherical center coordinates, the prior P-wave velocity and the sampling interval of the standard wave signal, the phase of the standard wave signal is shifted by a preset time delay to obtain a migrated wave signal;

[0080] The center coordinates of the annular first monitoring array (i.e. the centroid point of the outer monitoring ring) are determined, and lines are drawn from the center coordinates to each monitoring station in the first monitoring array to form a candidate propagation direction set; M groups of migrated wave signals are superimposed in each candidate propagation direction to determine the superimposed intensity of each candidate propagation direction, and the candidate propagation direction with the highest superimposed intensity is selected as the propagation direction;

[0081] A local coordinate system is constructed with the propagation direction as the normal direction, and the station coordinates of the M monitoring stations are subjected to surface fitting in the local coordinate system to obtain the local wave front curvature of each monitoring station.

[0082] It should be noted that the superimposed intensity of the candidate propagation direction is calculated according to the following formula:

[0083]

[0084] Wherein, represents the time sampling point index, represents the component index of the migrated wave signal, represents the monitoring station index, represents the migrated wave signal of the i th monitoring station, component and time .

[0085] It should be noted that the local coordinate system includes three axes respectively , and ;

[0086] Wherein, represents an arbitrarily selected non-zero vector, , represents the propagation direction.

[0087] The coordinates of the mth monitoring station in the local coordinate system are represented as The surface fitting is as follows:

[0088]

[0089] wherein, represents the local wavefront curvature of the mth monitoring station, which is solved by the least square method .

[0090] In detail, the original wave signal is sliced with a fixed window length H (such as 1 second) to generate K sub-signals. The direct current offset (de-meaning) and linear drift (de-trending) of each sub-signal are eliminated, and the standard wave signal is output to improve the signal-to-noise ratio of the subsequent phase migration. Among them, de-meaning and de-trending can be direct current offset correction algorithm (de-meaning) and linear regression de-trending algorithm (de-trending) respectively.

[0091] Based on the digital elevation model, station coordinates and topographic and geological data, a three-dimensional space grid model containing topographic relief is constructed. In the local mountain scale, the terrain surface is approximated by a sphere through the least square method, and the local wavefront model is obtained, and the spherical center coordinates are obtained.

[0092] The preset time delay of the jth standard wave signal is calculated as follows:

[0093] wherein,

[0094] represents the station coordinates of the jth monitoring station, represents the spherical center coordinates, represents the propagation speed of P wave, which is preferably 8 kilometers per second; The standard wave signal is time-domain shifted according to

[0095] to generate a migrated wave signal. In an embodiment of the present application, the local wavefront curvature of the corresponding station of the outer ring monitoring is constructed to form a closed wavefront curvature field, which comprises:

[0096] The geometric curve of the ridge of the target mountain is extracted to construct a monitoring outer ring (to ensure the geometric closure of the outer ring, and to provide spatial boundary constraints for the curvature field);

[0097]

[0098] ​A virtual monitoring station is inserted at the midpoint of adjacent monitoring stations in the monitoring outer ring, and the local wavefront curvature of the virtual station is the average of the local wavefront curvatures of the adjacent monitoring stations (encrypted sampling points are used to suppress curvature aliasing caused by excessively large azimuth intervals, and to improve subsequent fitting accuracy);

[0099] The virtual monitoring station is repeatedly inserted until the number of monitoring stations reaches a preset number threshold, to form a local wavefront curvature set (providing a sufficient number of curvature samples to meet the grid density requirement of continuous function fitting, and avoiding undersampling errors);

[0100] The local wavefront curvature set is fitted to obtain a curvature function as a closed wavefront curvature field (discrete curvature samples are converted into a smooth and closed analytical field, providing a differentiable mathematical basis for subsequent wavefront radius estimation and source positioning).

[0101] It should be noted that the contour tracking algorithm is used to extract discrete ridge points from the digital elevation model. The discrete ridge points are connected into a closed geometric curve through spline interpolation. The monitoring stations in the outer ring are projected onto the closed geometric curve (monitoring outer ring).

[0102] The preset number threshold is based on expert setting.

[0103] The periodic cubic spline interpolation algorithm is used to fit the local wavefront curvature set to generate a closed wavefront curvature field , , represents any point in the closed geometric curve, is a continuous and differentiable function.

[0104] In an embodiment of the present application, the wavefront estimated radius, the source plane estimated coordinates and the source center estimated depth are sequentially estimated based on the closed wavefront curvature field, including:

[0105] The wavefront estimated radius is determined based on the ring-wise average curvature of the closed wavefront curvature field (the ring-wise average is used to eliminate local curvature fluctuations, and the wavefront estimated radius is used to represent the overall wavefront curvature);

[0106] The wavefront estimated radius is loaded in the monitoring outer ring, and the source plane estimated coordinates are obtained by minimizing the intersection points of the wavefront estimated radius;

[0107] The mean time of the time stamps of the original wave signals of the first monitoring array and the mean time of the time stamps of the original wave signals of the second monitoring array are determined respectively, and the time difference between the mean times is calculated (the mean time difference between the inner and outer rings is used to suppress the noise interference of a single station, and the vertical time is sensitive to the source depth, providing a key time parameter for depth inversion);

[0108] The vertical propagation time is determined based on the time difference between the mean times;

[0109] Computing the focal center estimated depth As follows:

[0110]

[0111] Wherein, represents the propagation speed of P wave, represents the vertical propagation time, represents the wave front estimated radius, represents the average value of the Euclidean distance of each monitoring station in the first monitoring array and the focal plane estimated coordinates.

[0112] It should be noted that the ring average curvature is calculated based on the ring integral average , the wave front estimated radius is converted to: ;

[0113] It should be noted that one end of the wave front estimated radius is connected to the monitoring station, and the other end is located inside the closed geometric curve; by adjusting the position of the other end of each wave front estimated radius to minimize the intersection point of the wave front estimated radius (genetic algorithm is realized). Then sequentially connect the other end of all the wave front estimated radius to form a closed geometric figure, determine the centroid of the closed geometric figure based on the centroid method as the focal plane estimated coordinates.

[0114] It should be noted that the vertical propagation time is half of the time difference of the average time.

[0115] It should be noted that, determined by spherical geometry.

[0116] In one embodiment of the present application, a three-dimensional grid is constructed with the focal plane estimated coordinates and the focal center estimated depth as the initial position, including:

[0117] A three-dimensional grid with a preset step length is constructed in the space range centered on the focal plane estimated coordinates and extending a first preset distance east, west, south and north, and centered on the focal center estimated depth and extending a second preset distance vertically up and down.

[0118] It should be noted that the focal plane estimated coordinates are ;

[0119] Horizontal range calculation:

[0120] East-west: left boundary , right boundary ;

[0121] North-south: lower boundary , upper boundary ​;

[0122] Form a horizontal two-dimensional rectangular area based on the east-west and north-south directions:

[0123]

[0124] in, Indicates half of the first preset distance.

[0125] Vertical range calculation:

[0126] Nether , upper bound ;

[0127] The vertical range is: .

[0128] The preset step sizes include: x-direction step size , y-direction step length , z-direction step size ;

[0129] In a horizontal 2D rectangular area, by step and Divide evenly; by step length within the vertical range Divide evenly to form a three-dimensional grid.

[0130] In one embodiment of the present invention, based on the propagation direction and local wavefront curvature of each monitoring station, the minimum square error unbiased estimation algorithm is used to calculate the source parameter residual of each grid in the three-dimensional grid, so that the grid with the smallest source parameter residual is used as the final source position, including:

[0131] For any grid in the three-dimensional grid, a weight matrix is ​​formed according to the observed variance of the propagation direction and local wavefront curvature of each monitoring station. The minimum variance unbiased estimation algorithm is used to calculate the residual of the source parameters corresponding to the grid, and the grid with the smallest residual is selected as the final source position.

[0132] Specifically, obtain the line connecting the station coordinates of each monitoring station to the center coordinates of the first monitoring array, and calculate the cosine similarity between the line direction and the propagation direction. ; , is a positive integer;

[0133] The station , local wavefront curvature and and The corresponding observation variance and , using the inverse of the observed variance as the weight, construct a diagonal weight matrix , ; wherein the observation variance and is obtained based on a historical time period;

[0134] for the grid , the cosine similarity of each station theory under the assumption of the source is calculated by ray tracing and local wavefront curvature .

[0135] The difference between the observed value and the predicted value is combined into a residual vector .

[0136] The grid residual is obtained by weighting the residual vector by a diagonal weight matrix .

[0137] The grid residual of all grids is traversed, and the grid with the smallest grid residual is selected as the final source location.

[0138] A method for real-time monitoring and alarming of abnormality of a seismic early warning station network, the positioning system for realizing any one of the real-time monitoring of abnormality of the seismic early warning station network comprises:

[0139] Step 1, based on M stations, outer ring monitoring, inner ring monitoring, fault monitoring and vertical monitoring are performed on a target mountain to obtain M groups of original wave signals;

[0140] Step 2, for each group of original wave signals, Kirchhoff phase migration processing is performed with a preset sliding window H to obtain the propagation direction and local wavefront curvature of each monitoring station;

[0141] Step 3, based on the local wavefront curvature of the stations corresponding to the outer ring monitoring, a closed wavefront curvature field is constructed, and the wavefront estimation radius, the source plane estimation coordinates and the source center estimation depth are estimated in turn based on the closed wavefront curvature field;

[0142] Step 4, taking the source plane estimation coordinates and the source center estimation depth as the initial position, a three-dimensional grid is constructed; for any grid in the three-dimensional grid, the source parameter residual of each grid in the three-dimensional grid is calculated according to the propagation direction and the local wavefront curvature of each monitoring station, combined with the least variance unbiased estimation algorithm, and the grid with the smallest source parameter residual is taken as the final source location.

[0143] The above describes the embodiments of the embodiments, but the embodiments are not limited to the specific embodiments described above, and the specific embodiments described above are only illustrative and not limiting, and those skilled in the art can make many forms under the inspiration of the embodiments, which all belong to the protection of the embodiments.

Claims

1. A positioning system for real-time monitoring of anomalies in a network of earthquake early warning stations, characterized by, The positioning system comprises: a data acquisition module configured to perform outer ring monitoring, inner ring monitoring, fault monitoring and vertical monitoring on a target mountain based on M stations to obtain M groups of original wave signals; a data processing module configured to perform Kirchhoff phase migration processing on each group of original wave signals based on a preset sliding window H to obtain a propagation direction and a local wave front curvature of each monitoring station; a positioning estimation module configured to construct a closed wave front curvature field based on the local wave front curvature of the stations corresponding to the outer ring monitoring, and sequentially estimate a wave front estimation radius, a focal plane estimation coordinate and a focal center estimation depth based on the closed wave front curvature field; a precise positioning module configured to construct a three-dimensional grid based on the focal plane estimation coordinate and the focal center estimation depth as initial positions; for any grid in the three-dimensional grid, the propagation direction and the local wave front curvature of each monitoring station are combined to calculate a focal parameter residual of each grid in the three-dimensional grid by using a minimum variance unbiased estimation algorithm, and a grid with the smallest focal parameter residual is taken as a final focal position; The positioning system is also configured to implement a monitoring and alarming method, and the method comprises the following steps: S1, obtaining M groups of original wave signals based on M stations; S2, extracting abnormal data of the original wave signals, including data delay, server time, network transmission flow and zero point drift; S3, if the abnormal data is synchronous and exceeds the limit, triggering a seismic alarm.

2. The positioning system for real-time monitoring of anomalies of an earthquake early warning station network according to claim 1, characterized in that, The outer ring monitoring, the inner ring monitoring, the fault monitoring and the vertical monitoring comprise: The outer ring monitoring comprises: constructing a ring-shaped first monitoring array around the ridges of the target mountain; The inner ring monitoring comprises: constructing a ring-shaped second monitoring array around the basins of the target mountain; The fault monitoring comprises: constructing a linear third monitoring array at a fixed horizontal interval based on the active faults of the target mountain; The vertical monitoring comprises: constructing a linear fourth monitoring array at a fixed vertical interval from the ridges to the basins of the target mountain. 3.The positioning system for real-time monitoring of anomalies of an earthquake early warning station network according to claim 2, characterized in that, The Kirchhoff phase migration processing on each group of original wave signals based on a preset sliding window H to obtain a propagation direction and a local wave front curvature of each monitoring station comprises: segmenting each group of original wave signals according to the preset sliding window H to obtain K sub-signals, and performing mean removal and trend removal processing on each sub-signal to obtain a standard wave signal; initializing a local wave front model, as follows: establishing a three-dimensional model of the target mountain; performing spherical fitting processing on the three-dimensional model to obtain a local wave front model, and determining the spherical center coordinates of the local wave front model; performing phase shift processing on the standard wave signal according to a preset time delay based on the local wave front model, the spherical center coordinates, the prior P-wave velocity and the sampling interval of the standard wave signal to obtain a migration wave signal; determining the center coordinates of the ring-shaped first monitoring array, connecting the center coordinates to each monitoring station in the first monitoring array to form a candidate propagation direction set, and superimposing the M groups of migration wave signals in each candidate propagation direction to determine the superimposed intensity of each candidate propagation direction, and selecting the candidate propagation direction with the highest superimposed intensity as the propagation direction; constructing a local coordinate system with the propagation direction as the normal direction, and performing surface fitting on the station coordinates of the M monitoring stations in the local coordinate system to obtain the local wave front curvature of each monitoring station.

4. The positioning system for real-time monitoring of anomalies of an earthquake early warning station network according to claim 3, characterized in that, The closed wave front curvature field is constructed based on the local wave front curvature of the corresponding station in the outer ring, and comprises: A monitoring outer ring is constructed by extracting the geometric curve of the ridge of the target mountainous area; A virtual monitoring station is inserted at the midpoint of adjacent monitoring stations in the monitoring outer ring, and the local wave front curvature of the virtual station is the average of the local wave front curvatures of the adjacent monitoring stations; The virtual monitoring station is repeatedly inserted until the number of monitoring stations reaches a preset number threshold, to form a set of local wave front curvatures; The set of local wave front curvatures is fitted to obtain a curvature function as the closed wave front curvature field.

5. The positioning system for real-time monitoring of anomalies of an earthquake early warning station network according to claim 4, characterized in that, The closed wave front curvature field is sequentially estimated to obtain a wave front estimated radius, a focal plane estimated coordinate and a focal center estimated depth, comprising: The wave front estimated radius is determined based on the ring-wise average curvature of the closed wave front curvature field; The wave front estimated radius is loaded in the monitoring outer ring, and the focal plane estimated coordinate is obtained by minimizing the intersection point of the wave front estimated radius; The mean time of the time stamp of the original wave signal of the first monitoring array and the mean time of the time stamp of the original wave signal of the second monitoring array are determined respectively, and the time difference between the mean times is calculated; The vertical propagation time is determined based on the time difference between the mean times; Computing a seismic source center estimated depth As follows: ； wherein, represents the propagation velocity of P-waves, represents the vertical propagation time, represents the wavefront estimation radius, represents the average of the Euclidean distances of each monitoring station in the first monitoring array from the estimated coordinates of the hypocenter plane.

6. The positioning system for real-time monitoring of anomalies of an earthquake early warning station network according to claim 5, characterized in that, The three-dimensional grid is constructed based on the focal plane estimated coordinate and the focal center estimated depth, comprising: A three-dimensional grid with a preset step length is constructed in the space range centered on the focal plane estimated coordinate and extending a first preset distance to the east, west, south and north, and centered on the focal center estimated depth and extending a second preset distance vertically up and down.

7. The positioning system for real-time monitoring of anomalies of an earthquake early warning station network according to claim 6, characterized in that, The focal source parameter residual of each grid in the three-dimensional grid is calculated according to the propagation direction and the local wave front curvature of each monitoring station, and the grid with the minimum focal source parameter residual is selected as the final focal source position, comprising: For any grid in the three-dimensional grid, a weight matrix is formed according to the observation variance of the propagation direction and the local wave front curvature of each monitoring station, and the least squares unbiased estimation algorithm is used to calculate the focal source parameter residual of the grid, and the grid with the minimum residual is selected as the final focal source position.

Images