Earthquake probability prediction method and device integrating earthquake electromagnetic anomaly precursors

Abstract

The present invention provides an earthquake probability prediction method and device that integrates earthquake electromagnetic anomaly precursors. The method first extracts local electromagnetic disturbance signals (including two mutually perpendicular horizontal components and one vertical component) within a preset time period from the original time-domain electromagnetic data of three components collected by reference and target observation stations. Signal energy analysis is then performed on the three-component local electromagnetic disturbance signals within the preset time period to determine the local electromagnetic anomaly signal. A comprehensive prediction model is then constructed based on the local electromagnetic anomaly signal and historical earthquake catalog data to predict the real-time probability of an earthquake. The method integrates electromagnetic anomalies and earthquake catalog data to construct a comprehensive prediction model that takes into account both the self-excitation effect of earthquakes and external excitation factors for earthquake probability prediction, significantly improving the accuracy and reliability of short-term earthquake predictions.

Description

Technical Field

[0001] The present invention relates to the technical field of earthquake prediction and electromagnetic signal analysis, and in particular to a method and device for earthquake probability prediction integrating earthquake electromagnetic anomaly precursors. Background Art

[0002] Earthquake prediction is a major challenge in the field of geoscience. Existing earthquake prediction technologies primarily rely on single physical parameters (such as surface deformation) or statistical models (such as the ETAS model), resulting in insufficient prediction accuracy. Traditional earthquake prediction methods have a low comprehensive utilization rate of earthquake precursor signals, particularly lacking quantitative modeling of electromagnetic anomaly signals.

[0003] Electromagnetic anomaly signals are a key source of earthquake precursors, and their identification and extraction technology has been implemented in authorized patent CN116661009B. However, effectively integrating the extracted local electromagnetic disturbance signals from stations into earthquake probability prediction models still faces the following deficiencies: existing earthquake prediction models only consider the self-excitation effects of earthquakes and fail to incorporate external excitation factors such as electromagnetic anomalies, limiting model prediction accuracy; they do not dynamically adjust model prediction probabilities based on real-time electromagnetic observation data, making it impossible to dynamically correct model prediction results; and they lack a multi-model joint optimization mechanism based on the information criterion, resulting in inadequate optimization of model parameters. Summary of the Invention

[0004] In view of this, the object of the present invention is to provide a method and device for earthquake probability prediction that integrates earthquake electromagnetic anomaly precursors to alleviate the above-mentioned problems existing in existing earthquake prediction technologies.

[0005] In a first aspect, an embodiment of the present invention provides an earthquake probability prediction method that integrates earthquake electromagnetic anomaly precursors, comprising: extracting local electromagnetic disturbance signals of the three components within a preset time period from the original time domain electromagnetic data of the three components collected by a reference station and a target observation station within a preset time period; wherein the three components include two horizontal components and one vertical component that are perpendicular to each other; performing signal energy analysis on the local electromagnetic disturbance signals of the three components within the preset time period to determine the local electromagnetic anomaly signal; constructing a comprehensive prediction model based on the local electromagnetic anomaly signal and historical earthquake catalog data; using the comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data to obtain the real-time probability of earthquake occurrence; wherein the start time point of the time period corresponding to the real-time electromagnetic anomaly data is later than the end time point of the preset time period, and the start time point of the time period corresponding to the earthquake catalog data is later than the end time point of the time period corresponding to the historical earthquake catalog data.

[0006] In a second aspect, an embodiment of the present invention further provides an earthquake probability prediction device that integrates earthquake electromagnetic anomaly precursors, comprising: an extraction module for extracting local electromagnetic disturbance signals of three components within a preset time period from the original time domain electromagnetic data of three components collected by a reference station and a target observation station within a preset time period; wherein the three components include two horizontal components and one vertical component perpendicular to each other; a determination module for performing signal energy analysis on the local electromagnetic disturbance signals of the three components within the preset time period to determine the local electromagnetic anomaly signal; a construction module for constructing a comprehensive prediction model based on the local electromagnetic anomaly signal and historical earthquake catalog data; a prediction module for using the comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data to obtain the real-time probability of earthquake occurrence; wherein the start time point of the time period corresponding to the real-time electromagnetic anomaly data is later than the end time point of the preset time period, and the start time point of the time period corresponding to the earthquake catalog data is later than the end time point of the time period corresponding to the historical earthquake catalog data.

[0007] In a third aspect, an embodiment of the present invention further provides an electronic device comprising a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the earthquake probability prediction method that integrates earthquake electromagnetic anomaly precursors as described in the first aspect above.

[0008] Embodiments of the present invention provide a method and device for earthquake probability prediction that integrates earthquake electromagnetic anomaly precursors. The method first extracts local electromagnetic disturbance signals (including two mutually perpendicular horizontal components and one vertical component) within a preset time period from the three-component raw time-domain electromagnetic data collected by reference and target observation stations. Signal energy analysis is then performed on the three-component local electromagnetic disturbance signals within the preset time period to determine the local electromagnetic anomaly signal. A comprehensive prediction model is then constructed based on the local electromagnetic anomaly signal and historical earthquake catalog data to predict the real-time probability of an earthquake. Using this technology, electromagnetic anomalies and earthquake catalog data can be integrated to construct a comprehensive prediction model that takes into account both the earthquake's self-excitation effects and external excitation factors to predict earthquake occurrence probability, significantly improving the accuracy and reliability of short-term earthquake predictions.

[0009] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0010] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 Schematic diagram of a flow chart of an earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors in an embodiment of the present invention;

[0013] Figure 2 This is a flow chart of earthquake probability prediction integrating earthquake electromagnetic anomaly precursors in an embodiment of the present invention;

[0014] Figure 3 1 is an example diagram of the change of magnetic field data and residual error of the Z component in an embodiment of the present invention;

[0015] Figure 4 This is an example diagram of earthquake intensity and earthquake magnitude obtained by the ETAS model and earthquake intensity obtained by the external excitation model in an embodiment of the present invention;

[0016] Figure 5 This is an example diagram of earthquake occurrence probability predicted by the comprehensive prediction model in an embodiment of the present invention;

[0017] Figure 6 Schematic diagram of the structure of an earthquake probability prediction device integrating earthquake electromagnetic anomaly precursors according to an embodiment of the present invention;

[0018] Figure 7 Schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0020] To facilitate understanding of this embodiment, firstly, a method for earthquake probability prediction integrating earthquake electromagnetic anomaly precursors disclosed in an embodiment of the present invention is described in detail. Figure 1As shown, the method may include the following steps:

[0021] Step S102 : extracting local electromagnetic disturbance signals of the three components within a preset time period from the original time domain electromagnetic data of the three components collected by the reference station and the target observation station within a preset time period.

[0022] The three components may include two mutually perpendicular horizontal components (ie, an X component and a Y component) and one vertical component (ie, a Z component).

[0023] Step S104 : performing signal energy analysis on the three components of the local electromagnetic disturbance signal within a preset time period to determine the local electromagnetic abnormal signal.

[0024] Step S106: constructing a comprehensive prediction model based on the local electromagnetic anomaly signal and historical earthquake catalog data.

[0025] For example, the comprehensive prediction model may adopt the following conditional intensity function:

[0026]

[0027] in, is the conditional intensity function, is the background earthquake rate, is the contribution of earthquake triggering to earthquake occurrence rate, The contribution of electromagnetic anomaly triggering to earthquake occurrence rate;

[0028] self-excited term ETAS models can be used, The expression can be:

[0029]

[0030] in, for The time of the earthquake before, is the aftershock generation rate, is the magnitude influencing factor, For the time of occurrence The magnitude of the earthquake, is the lower limit of magnitude, is the time delay, is the decay rate factor;

[0031] External excitation term The Gaussian kernel function can be used. The expression can be:

[0032]

[0033] in, for The time when the earthquake electromagnetic anomaly observed before time appears, is the maximum intensity gain of earthquakes triggered by electromagnetic anomalies, is the lead time for electromagnetic anomalies to trigger earthquakes, The length of the time window for electromagnetic anomalies to trigger earthquakes.

[0034] Step S108: Use a comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data to obtain the real-time probability of earthquake occurrence.

[0035] Among them, the start time point of the period corresponding to the real-time electromagnetic anomaly data is later than the end time point of the preset period, and the start time point of the period corresponding to the earthquake catalog data is later than the end time point of the period corresponding to the historical earthquake catalog data.

[0036] An embodiment of the present invention provides an earthquake probability prediction method that integrates earthquake electromagnetic anomaly precursors. The method first extracts local electromagnetic disturbance signals (including two mutually perpendicular horizontal components and one vertical component) within a preset time period from the original time-domain electromagnetic data of three components collected by a reference station and a target observation station. Signal energy analysis is then performed on the local electromagnetic disturbance signals of the three components within the preset time period to determine the local electromagnetic anomaly signal. A comprehensive prediction model is then constructed based on the local electromagnetic anomaly signal and historical earthquake catalog data to predict the observed real-time electromagnetic anomaly data and earthquake catalog data, thereby obtaining the real-time probability of an earthquake. Using this technology, electromagnetic anomalies and earthquake catalogs can be integrated to construct a comprehensive prediction model that takes into account both the earthquake's self-excitation effects and external excitation factors to predict earthquake occurrence probability, significantly improving the accuracy and reliability of short-term earthquake predictions.

[0037] As a possible implementation, the above step S106 (i.e., constructing a comprehensive prediction model based on local electromagnetic anomaly signals and historical earthquake catalog data) may include:

[0038] Step A1: construct an initial comprehensive prediction model based on local electromagnetic anomaly signals and historical earthquake catalog data.

[0039] In step A2, based on a preset information criterion, the maximum likelihood estimation method is used to optimize the parameters of the initial comprehensive prediction model to obtain a comprehensive prediction model.

[0040] For example, the operation of step A2 (i.e., optimizing the parameters of the initial comprehensive prediction model using the maximum likelihood estimation method based on a preset information criterion) may include:

[0041] In step A21, the following log-likelihood function is constructed with the constraints of maximizing the probability of an earthquake occurring at the time when an earthquake actually occurs and minimizing the probability of an earthquake occurring at the time when an earthquake does not actually occur:

[0042]

[0043] in, is the log-likelihood function, is the end time of the earthquake probability prediction period, is the parameter set of the initial comprehensive prediction model, For the parameter set Under the distribution conditions The intensity of the earthquake at any moment.

[0044] Step A22: Establish an objective function based on a preset information criterion and a log-likelihood function.

[0045] For example, the preset information criterion may adopt the Akaike Information Criterion (AIC), and the objective function value may adopt the AIC value. Accordingly, the following objective function may be established based on the preset information criterion and the log-likelihood function:

[0046]

[0047] in, For the parameter set The number of parameters to be optimized.

[0048] In step A23, the parameters of the initial comprehensive prediction model are iteratively optimized with the goal of minimizing the objective function value until the optimization ends when a preset end condition is reached.

[0049] The principle of using the maximum likelihood estimation method to optimize parameters can be as follows: for the earthquake probability prediction model, it is assumed that the parameters to be estimated constitute the parameter set To implement the maximum likelihood estimation method mathematically, we must first define the likelihood, that is, in the parameter set Under the conditions of observing the known time (such as: ) The probability of an earthquake occurring. The physical meaning of using the maximum likelihood estimation method to optimize parameters is to maximize the probability of an earthquake occurring at all actual earthquake times, while minimizing the probability of an earthquake occurring at times when no earthquakes occur. Constraints can be defined according to this physical meaning to construct a log-likelihood function, which can be expressed as:

[0050]

[0051] In addition, since the number of parameters of different earthquake probability prediction models may vary, in order to compare the effects of different earthquake probability prediction models, the Akaike Information Criterion can be used to optimize the parameters of the earthquake probability prediction models. The AIC value is calculated as follows:

[0052]

[0053] As a penalty term for the number of parameters in the earthquake probability prediction model, it can eliminate the influence of different numbers of parameters and avoid overfitting. In this way, the likelihood function is maximized. This can be transformed into optimizing AIC to minimize the AIC value to find the optimal parameter values ​​for the earthquake probability prediction model. Optimizing the parameters of the earthquake probability prediction model can be achieved using any existing optimization algorithm (such as the Gray Wolf Optimization Algorithm, the Whale Optimization Algorithm, the Sparrow Optimization Algorithm, the Particle Swarm Optimization Algorithm, etc.). The optimization algorithm used here is not limited. The parameters of the earthquake probability prediction model are iteratively calculated using the optimization algorithm to find the parameter value that minimizes the AIC, which serves as the optimal parameter for the earthquake probability prediction model.

[0054] As a possible implementation method, the signal energy of the real-time electromagnetic anomaly data exceeds a preset anomaly threshold; before using the comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data in the above-mentioned step S108, the above-mentioned earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors may also include: calculating the signal energy of the observed real-time electromagnetic data, and judging whether there is electromagnetic anomaly data in the real-time electromagnetic data whose signal energy exceeds the preset anomaly threshold, so as to determine the time corresponding to the real-time electromagnetic anomaly data as the time when the real-time electromagnetic anomaly data exists in the real-time electromagnetic data.

[0055] Continuing with the previous example, an abnormal baseline value can be set in advance. After obtaining real-time electromagnetic data over a period of time through observation (in fact, it can be time series data of electromagnetic indicators changing with time), the signal energy of the real-time electromagnetic data can be calculated, and the signal energy of the real-time electromagnetic data can be compared with the abnormal baseline value to determine whether there is a part in the real-time electromagnetic data whose signal energy exceeds the abnormal baseline value; when the signal energy of the real-time electromagnetic data exceeds the abnormal baseline value, the part of the real-time electromagnetic data whose signal energy exceeds the abnormal baseline value can be used as real-time electromagnetic anomaly data, and the time corresponding to the real-time electromagnetic anomaly data can be marked as the time when the earthquake electromagnetic anomaly occurs, and then the real-time electromagnetic anomaly data marked with the time when the earthquake electromagnetic anomaly occurs can be input into the comprehensive prediction model to participate in the earthquake probability prediction process. The process of predicting the earthquake probability by the comprehensive prediction model can be considered as a dynamic correction of the above-mentioned conditional intensity function The comprehensive prediction model finally outputs the probability of earthquake occurrence.

[0056] The specific process of predicting earthquake occurrence probability by comprehensive prediction model can be as follows: input into comprehensive prediction model The time when the electromagnetic anomaly observed before time and the maximum intensity gain of earthquakes triggered by electromagnetic anomalies and the lead time for electromagnetic anomalies to trigger earthquakes and time window length , substitute the external excitation term of the comprehensive prediction model The expression is calculated to obtain the probability of earthquakes triggered by electromagnetic anomalies at each moment in the future; the catalog data of earthquakes that have occurred (including: The time of the earthquake before , occurred at The magnitude of the earthquake and the lower limit of magnitude ), substituted into the self-excitation term of the comprehensive prediction model The expression is calculated, and then the conditional intensity function is used based on the electromagnetic data at each moment in the future period and the earthquake catalog data that has occurred Calculations are performed to obtain the earthquake condition intensity value at each moment in the future period of time. The earthquake condition intensity value is the probability value of an earthquake occurring at each moment in the future period of time.

[0057] As a possible implementation, the above-mentioned step S104 (i.e., performing signal energy analysis on the local electromagnetic disturbance signals of the three components within a preset time period to determine the local electromagnetic abnormality signal) may include: calculating the signal energy of the local electromagnetic disturbance signals of the three components within the preset time period; judging whether the obtained signal energy exceeds a preset energy threshold, so as to determine the local electromagnetic disturbance signal whose signal energy exceeds the preset energy threshold as a local electromagnetic abnormality signal.

[0058] Taking the vertical component (i.e., Z component) as an example, the signal energy calculation formula of the local electromagnetic disturbance signal of the Z component can be:

[0059]

[0060] in, for The actual observed change of the Z component of the observation station at each moment, For the observatory The estimated change in the Z component at time t, is the total number of data points.

[0061] The signal energy calculation formulas for the local electromagnetic disturbance signals of the X component and the Y component are similar to the signal energy calculation formula for the local electromagnetic disturbance signal of the Z component. The main difference is that the relevant variables of the Z component are replaced by the variables corresponding to the X component and the Y component respectively. The signal energy calculation formulas corresponding to the X component and the Y component are not repeated here.

[0062] Based on the signal energy calculation formula corresponding to each of the X, Y, and Z components, a judgment rule for seismic electromagnetic anomaly signals can be constructed. Specifically, for a certain component, an energy threshold can be set first. After obtaining the electromagnetic disturbance signal of the component and calculating the signal energy corresponding to the component, the calculated signal energy corresponding to the component can be compared with the set energy threshold to determine whether the signal energy exceeds the energy threshold. If the signal energy corresponding to the component exceeds the energy threshold, it can be determined that the electromagnetic disturbance signal of the component contains an electromagnetic anomaly signal, and the portion of the electromagnetic disturbance signal of the component whose signal energy exceeds the energy threshold is regarded as the electromagnetic anomaly signal, so that the relevant data at the time corresponding to the electromagnetic anomaly signal can be used to construct a comprehensive prediction model.

[0063] As a possible implementation, step S102 (i.e., extracting local electromagnetic disturbance signals of three components within a preset time period from the original time-domain electromagnetic data of the three components collected by the reference station and the target observation station within a preset time period) may include:

[0064] Step a1: filter the original time domain electromagnetic data of the three components collected by the reference station and the target observation station within a preset time period, and transform the filtered time domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station within the preset time period into frequency domain electromagnetic data of the corresponding components.

[0065] Continuing with the previous example, we can first obtain the original time domain electromagnetic data of the X, Y, and Z components collected by the reference station and a certain observation station over a period of time, and then perform bandpass filtering on the original time domain electromagnetic data of the X, Y components of the reference station and the X, Y, and Z components of the observation station respectively. After that, we use time-frequency transformation to transform the filtered time domain electromagnetic data of each component into the frequency domain electromagnetic data of the corresponding component, thereby obtaining the frequency domain electromagnetic data of the X, Y components of the reference station and the X, Y, and Z components of the observation station.

[0066] Step a2, based on the relationship between the preset time window length and the period and the frequency domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station within the preset time period, respectively calculate the coherence coefficients between the three components of the target observation station and the two horizontal components of the reference station in each period within the preset time period.

[0067] Continuing with the previous example, a corresponding time window length can be set for each cycle within the preset time period based on the relationship between the preset time window length and the cycle, and each cycle can be divided into multiple time windows according to the set time window length; for each cycle, based on the frequency domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station in each time window of the cycle, the coherence coefficients (such as wavelet coherence coefficients, etc.) between the three components of the target observation station and the two horizontal components of the reference station in each time window of the cycle are calculated respectively, and the coherence coefficients between the corresponding components in all time windows of the cycle are determined as the coherence coefficients between the three components of the target observation station and the two horizontal components of the reference station in the cycle.

[0068] Step a3, calculate the inter-station transfer function value between each of the three components of the target observation station in each period and the two horizontal components of the reference station based on the coherence coefficient between each of the three components of the target observation station in each period and the two horizontal components of the reference station and the filtered time domain electromagnetic data of the two horizontal components of the reference station in each period and the three components of the target observation station.

[0069] Continuing with the previous example, the first time window in which the coherence coefficient between the three components of the target observation station and the two horizontal components of the reference station meets the preset coherence coefficient threshold condition can be screened out from the time windows in each cycle; for each cycle, based on the filtered time domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station in each first time window of the cycle, the inter-station transfer function values ​​between the three components of the target observation station and the two horizontal components of the reference station in each first time window of the cycle are calculated, and the inter-station transfer function values ​​between the corresponding components in all first time windows of the cycle are averaged to obtain the inter-station transfer function values ​​between the three components of the target observation station and the two horizontal components of the reference station in the cycle.

[0070] Step a4, based on the frequency domain electromagnetic data of the two horizontal components of the reference station in each period and the inter-station conversion function values ​​between the three components of the target observation station in each period and the two horizontal components of the reference station, calculate the frequency domain estimated electromagnetic data of the three components of the target observation station in the preset time period, and transform the frequency domain estimated electromagnetic data of the three components of the target observation station in the preset time period into time domain estimated electromagnetic data of the corresponding components.

[0071] Step a5, based on the time domain estimated electromagnetic data and filtered time domain electromagnetic data of the three components of the target observation station within the preset time period, calculate the residual between the filtered time domain electromagnetic data and the time domain estimated electromagnetic data of the three components of the target observation station within the preset time period, and extract the residual of the three components of the target observation station within the preset time period as the local electromagnetic disturbance signal of the three components within the preset time period.

[0072] The above-mentioned local electromagnetic disturbance signal extraction method can be specifically referred to the relevant content of the authorized patent CN116661009B, which will not be repeated here.

[0073] For ease of understanding, this paper takes a specific application as an example to introduce in detail the implementation of earthquake probability prediction integrating earthquake electromagnetic anomaly precursors. Figure 2 As shown in the figure, the earthquake probability prediction process integrating earthquake electromagnetic anomaly precursors can mainly include:

[0074] Step 1: Signal extraction.

[0075] Using the local electromagnetic signal identification method disclosed in the authorized patent CN116661009B, the local electromagnetic disturbance signals of the three components X, Y, and Z of the target station (i.e., the target observation station) are extracted from the electromagnetic data of the observation station.

[0076] The local electromagnetic disturbance signal extraction process mainly includes: filtering the original time domain electromagnetic data in sequence, transforming it into frequency domain electromagnetic data and calculating the coherence coefficient, estimating the background field of the target station based on the inter-station transfer function (including calculation of the transfer function value, calculation of the frequency domain estimated electromagnetic data and transformation into time domain estimated electromagnetic data), and then extracting the residuals between the observed values ​​and the model estimated values ​​of the three components X, Y and Z of the target station as the local electromagnetic disturbance signal.

[0077] Step 2: Abnormal feature analysis.

[0078] The obtained local electromagnetic disturbance signal is subjected to frequency band energy analysis (including calculating the energy of each sub-band of the local electromagnetic disturbance signal) to extract the frequency band energy characteristics, and based on the obtained frequency band energy characteristics, the abnormal characteristics are extracted from the local electromagnetic disturbance signal as the local electromagnetic abnormality signal.

[0079] The local electromagnetic anomaly signal extraction process mainly includes: calculating the signal energy of the local electromagnetic disturbance signal of the three components X, Y, and Z. If the signal energy exceeds the preset energy threshold, it is determined that a local electromagnetic anomaly signal exists, and the part of the local electromagnetic anomaly signal whose signal energy exceeds the energy threshold is extracted as the electromagnetic anomaly signal.

[0080] Step 3: Construction of comprehensive prediction model.

[0081] The seismic self-excitation model (i.e., ETAS model) and the geomagnetic anomaly external excitation model are integrated to establish a conditional intensity function. The conditional intensity function is the following formula 1:

[0082] Formula 1:

[0083] in, is the earthquake self-excitation term (using the ETAS model), is the external excitation term of geomagnetic anomaly (using Gaussian kernel model);

[0084] The principle of the ETAS model is to judge the probability of current and future earthquakes based on earthquakes that have occurred in the past. That is, the probability of an earthquake occurring at each moment predicted by the ETAS model is determined by all earthquakes that have occurred previously. Therefore, the earthquake self-excitation term Mainly relies on existing earthquake catalogs.

[0085] The expression is the following formula 2:

[0086] Formula 2:

[0087] in, The magnitude is The average number of aftershocks triggered by an earthquake is The time difference between the parent event and its offspring The probability density function of .

[0088] The specific form is the following formula 3:

[0089] Formula 3:

[0090] The specific form is the following formula 4:

[0091] Formula 4:

[0092] The expression is the following formula 5:

[0093] Formula 5:

[0094] Laguerre polynomials are usually used. Studies have shown that there is a certain time difference between geomagnetic anomalies and earthquake occurrences. Therefore, Gaussian kernel functions can be used to describe the geomagnetic anomaly excitation model. The expression is the following formula 6:

[0095] Formula 6:

[0096] The above formulas 2 to 4 can be integrated to obtain the following expression of the earthquake self-excitation model:

[0097]

[0098] The above formulas 5 and 6 can be integrated to obtain the following expression of the geomagnetic anomaly excitation model:

[0099]

[0100] According to the above formula 1, the earthquake self-excitation model and the geomagnetic anomaly external excitation model can be integrated to obtain an earthquake probability prediction model (also known as a comprehensive prediction model) that integrates earthquake electromagnetic anomaly precursors.

[0101] Step 4: Model parameter optimization.

[0102] The maximum likelihood estimation method can be used to optimize the parameters of the comprehensive prediction model, and the optimal parameter combination of the comprehensive prediction model can be screened out in combination with AIC.

[0103] The model parameter optimization process mainly includes: establishing parameter sets , construct the log-likelihood function , establish the AIC calculation formula to maximize the likelihood function The optimization algorithm is used to perform iterative calculations to find the parameter set that minimizes the AIC value as the optimal parameter combination for the comprehensive prediction model.

[0104] Step 5: Model prediction output.

[0105] Obtain observed real-time electromagnetic data and earthquake catalog data, and calculate the signal energy of the real-time electromagnetic data; when real-time electromagnetic anomaly features appear in the real-time electromagnetic data (i.e., real-time electromagnetic anomaly data whose signal energy exceeds a preset anomaly baseline value), mark the time corresponding to the real-time electromagnetic anomaly feature as the time when the earthquake electromagnetic anomaly occurs, and input the real-time electromagnetic anomaly features marked with the time when the earthquake electromagnetic anomaly occurs and the observed earthquake catalog data into the comprehensive prediction model, which dynamically corrects the conditional intensity function to output the probability value of future earthquakes.

[0106] Compared with existing technologies, the key innovations of the above-mentioned earthquake probability prediction method integrating seismic electromagnetic anomaly precursors lie in the following aspects:

[0107] (1) Model fusion innovation: combining the ETAS self-excitation model with the Gaussian extranuclear excitation model to achieve joint modeling of earthquake triggering mechanisms and electromagnetic precursors;

[0108] (2) Dynamic optimization mechanism: Introducing the Akaike Information Criterion and real-time electromagnetic anomaly data to dynamically adjust the earthquake probability output to avoid overfitting;

[0109] (3) High-precision feature extraction: Extracting abnormal features with clear physical dimensions (i.e., signal energy) based on time-frequency analysis to reduce noise interference;

[0110] (4) Multi-source data fusion: Integrate the external excitation of electromagnetic anomalies and the self-excitation of earthquake catalogs to improve the reliability of earthquake probability prediction.

[0111] The effectiveness of the above-mentioned earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors can be verified through the following specific example data, and the effectiveness verification content is described as follows.

[0112] Figure 3 (a) shows the changes in the Z component magnetic field data of a target station (KYS) and a reference station (KAK) within 30 minutes in the early morning of a certain day. Figure 3 The three magnetic field data change curves presented in (a) from top to bottom are the magnetic field data change curves of KAK observation value, KYS observation value and KYS estimation value. There is a large difference between the actual observation values ​​of the target station and the reference station, indicating that the Z component of the target station mainly reflects the local anomaly generated near the station. Figure 3 As shown in (a), the phase of the estimated value of the target station is relatively consistent with the observed value, indicating that the phase difference between the two stations can be recovered by the inter-station transfer function.

[0113] Figure 3 (b) shows the residual change between the observed and estimated values ​​of the Z component of the target station within 30 minutes of the early morning of a certain day. It shows obvious anomalies in multiple time periods, with an amplitude of about 1 nT. This abnormal signal is the local electromagnetic signal generated near the target station (i.e., the local electromagnetic disturbance signal).

[0114] Figure 4 (a) shows the earthquake intensity and magnitude after the main shock estimated based on a self-excitation model (i.e., ETAS model) using an actual earthquake catalog. It can be seen that a series of aftershocks occurred after a large-magnitude main shock at 250 seconds, and the intensity of the aftershocks decayed exponentially, which better describes the earthquake intensity of this earthquake sample at different times.

[0115] Figure 4 (b) shows the earthquake intensity calculated by the external excitation model according to the above formula 6 under the condition of a given set of model parameters. It can be seen that the lead time after the occurrence of electromagnetic anomaly is , the time from the moment the electromagnetic anomaly occurs is The forecast window length after The intensity of the earthquake showed a Gaussian distribution, first increasing and then decreasing. The earthquake intensity is at its highest value at the moment of , and before and after this moment, the earthquake intensity shows a decreasing trend until it reaches 0. Figure 4 As can be seen from (b) in the figure, the functional form of the external excitation model conforms to the characteristics of abnormal external triggering and can describe anomalies similar to seismic electromagnetic signals.

[0116] Table 1. The 10-year observation data within 100 km of KAK station as the target observation station. The total residual energy of the earthquake catalog and the actual observed and estimated values ​​of the Z component of the KAK station is used as the characteristics of seismic electromagnetic anomalies to establish a comprehensive model (i.e., a comprehensive prediction model). The AIC obtained after model parameter optimization using the random Poisson model, the earthquake self-excitation model (i.e., the ETAS model), the anomaly external excitation model, and the comprehensive model are compared. Table 1 specifically shows the number of parameters of these models after optimization. The AIC values ​​and the changes in the AIC values ​​of the earthquake self-excitation model (i.e., ETAS model), the abnormal external excitation model, and the comprehensive model compared with the AIC value of the random Poisson model .

[0117] Table 1 Comparison of AIC values ​​of different models

[0118]

[0119] According to Table 1, both the self-excitation model and the external excitation model are better than the random model, and the comprehensive model is better than the other single models due to the smallest AIC value.

[0120] Figure 5 The curve showing the probability of an earthquake with a magnitude of 4 or above occurring within 100 km of the KAK station in mid-2001, obtained by optimizing the parameters of the comprehensive model, is presented. Figure 5 The red circle indicates the actual earthquake intensity at the time of the earthquake. The red circle indicates that the actual earthquake intensity at the time of the earthquake is greater than the earthquake intensity calculated according to the Poisson distribution ( Figure 5 (The dashed line indicates this, e.g., the total earthquake intensity in a year is divided by the number of days in a year to get the daily earthquake intensity). Figure 5 The text also distinguishes between the intensity of earthquakes triggered by anomalies and the intensity of earthquakes triggered by earthquakes. Figure 5 The actual time of the earthquake is marked with a green circle. The green circle indicates that the actual earthquake intensity is lower than the earthquake intensity calculated by Poisson distribution. Figure 5 It can be seen that the actual time of earthquake occurrence falls on both the time when the probability of earthquake occurrence increases due to abnormal triggering and the time when the probability of earthquake occurrence increases due to earthquake triggering. This shows that the comprehensive model integrating seismic electromagnetic signals is more effective, that is, the comprehensive model can predict the probability of earthquake occurrence more accurately, which is consistent with the actual occurrence of earthquakes.

[0121] The advantages of the above-mentioned earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors are mainly as follows: improving the prediction accuracy of earthquake probability through multi-source data fusion and dynamic correction; it can be extended to the comprehensive modeling of other precursor signals (such as geoelectric, geothermal, atmospheric, and ionosphere), and has strong technical compatibility.

[0122] Based on the above-mentioned earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors, the embodiment of the present invention further provides an earthquake probability prediction device integrating earthquake electromagnetic anomaly precursors, see Figure 6 As shown, the device may include the following modules:

[0123] The extraction module 602 is used to extract the local electromagnetic disturbance signals of the three components within the preset time period from the original time domain electromagnetic data of the three components collected by the reference station and the target observation station within the preset time period; wherein the three components include two horizontal components and one vertical component that are perpendicular to each other.

[0124] The determination module 604 is configured to perform signal energy analysis on the three components of the local electromagnetic disturbance signal within a preset time period to determine the local electromagnetic abnormal signal.

[0125] The construction module 606 is used to construct a comprehensive prediction model based on the local electromagnetic anomaly signal and historical earthquake catalog data.

[0126] The prediction module 608 is used to use the comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data to obtain the real-time probability of earthquake occurrence; wherein, the start time point of the time period corresponding to the real-time electromagnetic anomaly data is later than the end time point of the preset time period, and the start time point of the time period corresponding to the earthquake catalog data is later than the end time point of the time period corresponding to the historical earthquake catalog data.

[0127] The above-mentioned earthquake probability prediction device that integrates earthquake electromagnetic anomaly precursors can integrate electromagnetic anomalies and earthquake catalogs to construct a comprehensive prediction model that takes into account both earthquake self-excitation effects and external excitation factors to predict the probability of earthquake occurrence, which can significantly improve the accuracy and reliability of short-term earthquake predictions.

[0128] The above-mentioned construction module 606 can also be used to: construct an initial comprehensive prediction model based on the local electromagnetic anomaly signal and the historical earthquake catalog data; based on a preset information criterion, use the maximum likelihood estimation method to optimize the parameters of the initial comprehensive prediction model to obtain the comprehensive prediction model.

[0129] The construction module 606 can also be used to construct the following log-likelihood function with the constraints of maximizing the probability of an earthquake occurring at the time when an earthquake actually occurs and minimizing the probability of an earthquake occurring at the time when an earthquake does not actually occur: ,in, is the log-likelihood function, is the end time of the earthquake probability prediction period, is the parameter set of the initial comprehensive prediction model, For the parameter set Under the distribution conditions The intensity of the earthquake at the moment of occurrence; establishing an objective function based on a preset information criterion and the log-likelihood function; iteratively optimizing the parameters of the initial comprehensive prediction model with the goal of minimizing the objective function value until the optimization ends when a preset end condition is reached.

[0130] The preset information criterion may adopt the Akaike information criterion, and the objective function value may adopt the AIC value. Based on this, the construction module 606 may also be used to establish the following objective function based on the preset information criterion and the log-likelihood function: ,in, For the parameter set The number of parameters to be optimized.

[0131] The signal energy of the above-mentioned real-time electromagnetic anomaly data exceeds the preset anomaly threshold; based on this, the above-mentioned prediction module 608 can also be used to: before using the comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data, calculate the signal energy of the observed real-time electromagnetic data, and determine whether there is real-time electromagnetic anomaly data with signal energy exceeding the preset anomaly threshold in the real-time electromagnetic data, so as to determine the time corresponding to the real-time electromagnetic anomaly data as the time when the earthquake electromagnetic anomaly occurs when the real-time electromagnetic anomaly data exists in the real-time electromagnetic data.

[0132] The above-mentioned extraction module 602 can also be used to: filter the original time domain electromagnetic data of the three components collected by the reference station and the target observation station within the preset time period, and transform the filtered time domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station within the preset time period into frequency domain electromagnetic data of the corresponding components; based on the relationship between the preset time window length and the period and the frequency domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station within the preset time period, calculate the coherence coefficients between the three components of the target observation station and the two horizontal components of the reference station in each period within the preset time period; calculate the coherence coefficients between the three components of the target observation station and the two horizontal components of the reference station in each period and the filtered time domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station in each period The inter-station transfer function values ​​between the three components of the observation station and the two horizontal components of the reference station are obtained; based on the frequency domain electromagnetic data of the two horizontal components of the reference station in each period and the inter-station transfer function values ​​between the three components of the target observation station and the two horizontal components of the reference station in each period, the frequency domain estimated electromagnetic data of the three components of the target observation station in the preset period are calculated, and the frequency domain estimated electromagnetic data of the three components of the target observation station in the preset period are transformed into time domain estimated electromagnetic data of the corresponding components; based on the time domain estimated electromagnetic data and the filtered time domain electromagnetic data of the three components of the target observation station in the preset period, the residual between the filtered time domain electromagnetic data and the time domain estimated electromagnetic data of the three components of the target observation station in the preset period is calculated, and the residual of the three components of the target observation station in the preset period is extracted as the local electromagnetic disturbance signal of the three components in the preset period.

[0133] The above-mentioned determination module 604 can also be used to: calculate the signal energy of the local electromagnetic disturbance signal of the three components within a preset time period; determine whether the obtained signal energy exceeds the preset energy threshold, so as to determine the local electromagnetic disturbance signal whose signal energy exceeds the preset energy threshold as a local electromagnetic abnormality signal.

[0134] The earthquake probability prediction device that integrates earthquake electromagnetic anomaly precursors provided in the embodiment of the present invention has the same implementation principle and technical effects as the aforementioned earthquake probability prediction method embodiment that integrates earthquake electromagnetic anomaly precursors. For the sake of brief description, for parts not mentioned in the embodiment of the earthquake probability prediction device that integrates earthquake electromagnetic anomaly precursors, reference may be made to the corresponding content in the aforementioned earthquake probability prediction method embodiment that integrates earthquake electromagnetic anomaly precursors.

[0135] The embodiment of the present invention further provides an electronic device, such as Figure 7As shown, it is a structural diagram of the electronic device, wherein the electronic device includes a processor 71 and a memory 70, the memory 70 stores computer executable instructions that can be executed by the processor 71, and the processor 71 executes the computer executable instructions to realize the above-mentioned earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors.

[0136] exist Figure 7 In the illustrated embodiment, the electronic device further includes a bus 72 and a communication interface 73 , wherein the processor 71 , the communication interface 73 and the memory 70 are connected via the bus 72 .

[0137] Among them, the memory 70 may include high-speed random access memory (RAM), and may also include non-volatile memory (non-volatile memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 73 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 72 can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 72 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0138] The processor 71 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method may be performed by hardware integrated logic circuits or software instructions within the processor 71. The processor 71 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present invention may be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules within the decoding processor. The software module may be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or the like. The storage medium is located in the memory, and the processor 71 reads the information in the memory and, in combination with its hardware, completes the steps of the earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors of the aforementioned embodiment.

[0139] Unless otherwise specifically stated, the relative steps, numerical expressions and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0140] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0141] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0142] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A method for earthquake probability prediction integrating earthquake electromagnetic anomaly precursors, characterized in that: include: Extracting local electromagnetic disturbance signals of three components within a preset time period from the original time-domain electromagnetic data of three components collected by the reference station and the target observation station within a preset time period; wherein the three components include two mutually perpendicular horizontal components and one vertical component; Performing signal energy analysis on the three components of the local electromagnetic disturbance signal within a preset time period to determine the local electromagnetic abnormal signal; constructing a comprehensive prediction model based on the local electromagnetic anomaly signal and historical earthquake catalog data; The integrated prediction model is used to predict the observed real-time electromagnetic anomaly data and earthquake catalog data to obtain the real-time probability of earthquake occurrence; wherein the start time point of the period corresponding to the real-time electromagnetic anomaly data is later than the end time point of the preset period, and the start time point of the period corresponding to the earthquake catalog data is later than the end time point of the period corresponding to the historical earthquake catalog data; Based on the local electromagnetic anomaly signals and historical earthquake catalog data, a comprehensive prediction model is constructed, including: constructing an initial comprehensive prediction model based on the local electromagnetic anomaly signal and the historical earthquake catalog data; Based on a preset information criterion, the parameters of the initial comprehensive prediction model are optimized using the maximum likelihood estimation method to obtain the comprehensive prediction model; The comprehensive prediction model uses the following conditional intensity function: in, is the conditional intensity function, is the background earthquake occurrence rate, is the contribution of earthquake triggering to earthquake occurrence rate, The contribution of electromagnetic anomaly triggering to earthquake occurrence rate; The expression is: in, for The time of the earthquake before, is the aftershock generation rate, is the magnitude influencing factor, For the time of occurrence The magnitude of the earthquake, is the lower limit of magnitude, is the time delay, is the decay rate factor; The expression is: in, for The time when the earthquake electromagnetic anomaly observed before time appears, is the maximum intensity gain of earthquakes triggered by electromagnetic anomalies, is the lead time for electromagnetic anomalies to trigger earthquakes, The length of the time window for electromagnetic anomalies to trigger earthquakes.

2. The earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors according to claim 1 is characterized in that: Based on a preset information criterion, the maximum likelihood estimation method is used to optimize the parameters of the initial comprehensive prediction model, including: With the constraints of maximizing the probability of an earthquake occurring at the moment when an earthquake actually occurs and minimizing the probability of an earthquake occurring at the moment when an earthquake does not actually occur, the following log-likelihood function is constructed: in, is the log-likelihood function, is the end time of the earthquake probability prediction period, is the parameter set of the initial comprehensive prediction model, For the parameter set Under the conditions of distribution the intensity of earthquakes occurring at any given moment; Establishing an objective function based on a preset information criterion and the log-likelihood function; With the goal of minimizing the objective function value, the parameters of the initial comprehensive prediction model are iteratively optimized until the optimization ends when a preset end condition is reached.

3. The earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors according to claim 2 is characterized in that: The preset information criterion adopts Akaike Information Criterion, and the objective function value adopts AIC value; Based on the preset information criterion and the log-likelihood function, an objective function is established, including: Based on the preset information criterion and the log-likelihood function, the following objective function is established: in, For the parameter set The number of parameters to be optimized.

4. The earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors according to claim 1 is characterized in that: When the signal energy of the real-time electromagnetic anomaly data exceeds a preset anomaly threshold, before using the comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data, the method further includes: Calculate the signal energy of the observed real-time electromagnetic data, and determine whether there is real-time electromagnetic anomaly data in the real-time electromagnetic data whose signal energy exceeds a preset anomaly threshold, so that when the real-time electromagnetic anomaly data exists in the real-time electromagnetic data, the time corresponding to the real-time electromagnetic anomaly data is determined as the time when the seismic electromagnetic anomaly occurs.

5. The earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors according to claim 1 is characterized in that: The local electromagnetic disturbance signals of the three components within the preset time period are extracted from the original time domain electromagnetic data of the three components collected by the reference station and the target observation station within the preset time period, including: Filtering the original time-domain electromagnetic data of the three components collected by the reference station and the target observation station within a preset time period, and transforming the filtered time-domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station within the preset time period into frequency-domain electromagnetic data of the corresponding components; Based on the relationship between the preset time window length and the period and the frequency domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station in the preset period, the coherence coefficients between the three components of the target observation station and the two horizontal components of the reference station in each period in the preset period are calculated respectively; Based on the coherence coefficients between the three components of the target observation station and the two horizontal components of the reference station in each period, as well as the filtered time-domain electromagnetic data of the two horizontal components of the reference station and the three components of the target observation station in each period, the inter-station transfer function values ​​between the three components of the target observation station and the two horizontal components of the reference station in each period are calculated; Calculate the frequency domain estimated electromagnetic data of the three components of the target observation station within a preset period based on the frequency domain electromagnetic data of the two horizontal components of the reference station in each period and the inter-station transfer function values ​​between the three components of the target observation station in each period and the two horizontal components of the reference station, and transform the frequency domain estimated electromagnetic data of the three components of the target observation station within the preset period into time domain estimated electromagnetic data of the corresponding components; Based on the time domain estimated electromagnetic data and filtered time domain electromagnetic data of the three components of the target observation station within the preset time period, the residual between the filtered time domain electromagnetic data and the time domain estimated electromagnetic data of the three components of the target observation station within the preset time period is calculated, and the residual of the three components of the target observation station within the preset time period is extracted as the local electromagnetic disturbance signal of the three components within the preset time period.

6. The earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors according to claim 1 is characterized in that: Perform signal energy analysis on the three components of the local electromagnetic disturbance signal within a preset time period to determine the local electromagnetic abnormal signal, including: Calculating the signal energy of the three components of the local electromagnetic disturbance signal within a preset time period; It is determined whether the obtained signal energy exceeds a preset energy threshold, so as to determine the local electromagnetic disturbance signal whose signal energy exceeds the preset energy threshold as a local electromagnetic abnormality signal.

7. An earthquake probability prediction device integrating earthquake electromagnetic anomaly precursors, characterized in that: include: An extraction module is used to extract local electromagnetic disturbance signals of three components within a preset time period from the original time domain electromagnetic data of three components collected by the reference station and the target observation station within a preset time period; wherein the three components include two mutually perpendicular horizontal components and one vertical component; A determination module is used to perform signal energy analysis on the local electromagnetic disturbance signals of the three components within a preset time period to determine the local electromagnetic abnormality signal; A construction module, configured to construct a comprehensive prediction model based on the local electromagnetic anomaly signal and historical earthquake catalog data; A prediction module, configured to use the comprehensive prediction model to predict the observed real-time electromagnetic anomaly data and earthquake catalog data to obtain a real-time probability of earthquake occurrence; wherein the start time point of the period corresponding to the real-time electromagnetic anomaly data is later than the end time point of the preset period, and the start time point of the period corresponding to the earthquake catalog data is later than the end time point of the period corresponding to the historical earthquake catalog data; The construction module is further configured to: construct an initial comprehensive prediction model based on the local electromagnetic anomaly signal and the historical earthquake catalog data; optimize the parameters of the initial comprehensive prediction model using a maximum likelihood estimation method based on a preset information criterion to obtain the comprehensive prediction model; The comprehensive prediction model uses the following conditional intensity function: in, is the conditional intensity function, is the background earthquake occurrence rate, is the contribution of earthquake triggering to earthquake occurrence rate, The contribution of electromagnetic anomaly triggering to earthquake occurrence rate; The expression is: in, for The time of the earthquake before, is the aftershock generation rate, is the magnitude influencing factor, For the time of occurrence The magnitude of the earthquake, is the lower limit of magnitude, is the time delay, is the decay rate factor; The expression is: in, for The time when the earthquake electromagnetic anomaly observed before time appears, is the maximum intensity gain of earthquakes triggered by electromagnetic anomalies, is the lead time for electromagnetic anomalies to trigger earthquakes, The length of the time window for electromagnetic anomalies to trigger earthquakes.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the earthquake probability prediction method integrating earthquake electromagnetic anomaly precursors as described in any one of claims 1 to 6.

Images