Performance verification method and system of earthquake prediction model, medium and electronic equipment

Abstract

The invention discloses a performance verification method and system for an earthquake prediction model, a medium and electronic equipment. The method comprises the following steps: acquiring an earthquake directory of each historical earthquake in a target area of which the earthquake magnitude is greater than a preset earthquake magnitude threshold within a preset time period; calculating the seismic moment and the seismic area of each historical earthquake according to the earthquake catalog; according to the total seismic area of each historical earthquake, distributing the seismic moment of each historical earthquake to the longitude and latitude grid units belonging to the target area to obtain a seismic moment space distribution diagram of the target area; according to the seismic moment space distribution diagram, calculating a missing report rate and an abnormal space-time occupancy rate of a preset seismic prediction model under different prediction index thresholds; in a preset SASM error graph, fitting visual curves corresponding to the missing report rate and the abnormal space-time occupancy rate under different prediction index thresholds; and judging whether the performance of the preset earthquake prediction model is qualified or not according to the visual curve. Therefore, by adopting the embodiment of the invention, the accuracy of performance verification can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and particularly to a method, system, medium and electronic device for verifying the performance of an earthquake prediction model. Background Art

[0002] In seismically active regions, such as the southeastern part of the Qinghai-Tibet Plateau, earthquake prediction models are widely used to predict the likelihood of future earthquakes. The performance of these models is directly related to the effectiveness of disaster prevention measures. For example, earthquake prediction models can provide early warning information about earthquake risks to local governments and residents, so as to take corresponding preventive measures and reduce the losses that may be caused by earthquakes.

[0003] In the related art, the verification of earthquake prediction models mainly relies on a series of statistical test methods, such as the standardized test platform provided by the Collaboratory for the Study of Earthquake Predictability (CSEP) of the Southern California Earthquake Center (SCEC). These test methods include the Molchan diagram method.

[0004] Although the Molchan diagram method has played an important role in the evaluation of earthquake prediction models, it fails to comprehensively consider the earthquake impact range and ignores the earthquake energy release characteristics, such as the larger range covered by rock mass sliding, thus reducing the accuracy of performance verification. Summary of the Invention

[0005] Embodiments of the present application provide a method, system, medium and electronic device for verifying the performance of an earthquake prediction model. To provide a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary part is not a general review, nor is it intended to identify key / important constituent elements or delineate the protection scope of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the detailed description that follows.

[0006] In a first aspect, an embodiment of the present application provides a method for verifying the performance of an earthquake prediction model, the method comprising: Obtaining the earthquake catalog of each historical earthquake in a target area where the magnitude is greater than a preset magnitude threshold within a preset time period; Calculating the seismic moment and seismic area of each historical earthquake according to the earthquake catalog; Allocating the seismic moment of each historical earthquake to the longitude and latitude grid cells belonging to the target area through the total seismic area of each historical earthquake to obtain a seismic moment spatial distribution map of the target area; Calculating the false alarm rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds of a preset earthquake prediction model according to the seismic moment spatial distribution map; Fitting a visualization curve corresponding to the false alarm rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds in a preset SASM error map; Judge whether the performance of the preset earthquake prediction model is qualified according to the visualization curve.

[0007] Optionally, the preset SASM error graph includes a dotted line with Gain = 1 and and the boundary line; Gain is the probability gain, is the abnormal spatio-temporal occupancy rate, is the miss rate; Judging whether the performance of the preset earthquake prediction model is qualified according to the visualization curve includes: In the case where the visualization curve is below the dotted line, it indicates that the performance of the preset earthquake prediction model is better than random prediction to determine that the performance of the preset earthquake prediction model is qualified; or, In the case where the area enclosed by the visualization curve and the boundary line is greater than the preset area threshold, it is determined that the performance of the preset earthquake prediction model is qualified; or, Statistically calculate the proportion of seismic moment captured in the preset high-probability area, and determine that the performance of the preset earthquake prediction model is qualified when the seismic moment proportion meets the preset conditions.

[0008] Optionally, the earthquake catalog includes magnitude and epicenter location; According to the earthquake catalog, calculate the seismic moment and seismic area of each historical earthquake, including: Calculate the seismic moment of each historical earthquake according to the magnitude and the preset magnitude empirical formula; Take the epicenter location as the center of the ellipse to obtain the center of each historical earthquake; Calculate the fault length of each historical earthquake as the major axis of the ellipse according to the magnitude and the preset relationship between fault length and moment magnitude; Multiply the major axis by the preset coefficient to obtain the minor axis; Based on the center, major axis, direction of the major axis, and minor axis of each historical earthquake, construct an ellipse for each historical earthquake; the direction of the major axis is consistent with the fault direction of each historical earthquake; Calculate the area of the ellipse of each historical earthquake to obtain the total seismic area of each historical earthquake, and the seismic area is used to characterize the perceived intensity and surface damage area of each historical earthquake.

[0009] Optionally, the preset magnitude empirical formula is:

[0010] where is the seismic moment, is the magnitude; The preset relationship between fault length and moment magnitude is:

[0011] Among them, is the fault length.

[0012] Optionally, by the total seismic area of each historical earthquake, the seismic moment of each historical earthquake is distributed to the longitude-latitude grid cells belonging to the target area, and a spatial distribution map of the seismic moment of the target area is obtained, including: Mapping the ellipse of each historical earthquake to the longitude-latitude grid cells of the target area; Determining the seismic area of each historical earthquake in each longitude-latitude grid cell from the longitude-latitude grid cells of the target area according to the total seismic area of each historical earthquake; Calculating the ratio of the seismic area of each historical earthquake in each longitude-latitude grid cell to the total seismic area of each historical earthquake; Calculating the product of the seismic moment of each historical earthquake and the ratio to obtain the seismic moment assigned to each longitude-latitude grid cell; Distributing the seismic moment assigned to each longitude-latitude grid cell to each longitude-latitude grid cell to obtain a spatial distribution map of the seismic moment of the target area.

[0013] Optionally, the calculation formula for the seismic moment assigned to each longitude-latitude grid cell is:

[0014] Among them, is the seismic moment assigned to the th longitude-latitude grid cell, is the seismic moment of each historical earthquake.

[0015] Optionally, the calculation formula for the omission rate is:

[0016] The calculation formula for the abnormal spatio-temporal occupancy rate is:

[0017] Among them, is the index quantity of the preset earthquake prediction model for the th longitude-latitude grid cell, that is, in the preset earthquake prediction model, can be the earthquake prediction probability of the th longitude-latitude grid cell, is the threshold of the index quantity of the preset earthquake prediction model, is the th value of the seismic moment in the longitude-latitude grid cell, and the omission rate of the seismic moment and the value range of the abnormal spatio-temporal occupancy rate τ are both 0-1.

[0018] Second aspect, an embodiment of the present application provides a performance verification system for an earthquake prediction model, the system including: An earthquake catalog acquisition module, configured to acquire the earthquake catalog of each historical earthquake in a target area where the magnitude is greater than a preset magnitude threshold within a preset time period; An earthquake moment and earthquake area calculation module, configured to calculate the earthquake moment and earthquake area of each historical earthquake according to the earthquake catalog; An earthquake moment spatial distribution map generation module, configured to allocate the earthquake moment of each historical earthquake to the longitude and latitude grid cells belonging to the target area through the total earthquake area of each historical earthquake, so as to obtain the earthquake moment spatial distribution map of the target area; A miss rate and abnormal spatio-temporal occupancy rate calculation module, configured to calculate the miss rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds of a preset earthquake prediction model according to the earthquake moment spatial distribution map; A visualization curve fitting module, configured to fit the visualization curves corresponding to the miss rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds in a preset SASM error map; A performance determination module, configured to determine whether the performance of a preset earthquake prediction model is qualified according to the visualization curve.

[0019] Third aspect, an embodiment of the present application provides a computer storage medium, which stores multiple instructions, and the instructions are suitable for being loaded and executed by a processor to perform the above method steps.

[0020] Fourth aspect, an embodiment of the present application provides an electronic device, which may include: a processor and a memory; wherein, the memory stores a computer program, and the computer program is suitable for being loaded and executed by the processor to perform the above method steps.

[0021] The technical solution provided by the embodiment of the present application may include the following beneficial effects: In the embodiment of the present application, by accurately calculating and spatially allocating the earthquake moment and area of historical earthquakes, an earthquake moment spatial distribution map is generated, and then the performance of the earthquake prediction model is quantitatively evaluated. The earthquake moment spatial distribution map can comprehensively characterize the earthquake influence range and earthquake energy release characteristics, thereby improving the accuracy of performance verification. At the same time, by using the law of earthquake energy release and influence area, the earthquake moment reflecting the earthquake energy size is proportionally allocated to the earthquake image area to comprehensively verify the prediction ability of the model for the earthquake source area.

[0022] It should be understood that the above general description and subsequent detailed description are only exemplary and explanatory, and cannot limit the present application. Description of the Drawings

[0023] The accompanying drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0024] Figure 1 is a schematic flowchart of a method for verifying the performance of an earthquake prediction model provided by an embodiment of this application; Figure 2 is a spatial distribution map of the epicenters and seismic moments of historical earthquakes in southeastern Tibet of this application; Figure 3 is a schematic diagram of the principle of a SASM test provided by an embodiment of this application; Figure 4 is a distribution map of the earthquake areas of historical earthquakes in southeastern Tibet of this application; Figure 5 is a SASM error map of a RELM-TibetSE model provided by an embodiment of this application; Figure 6 is a schematic diagram of a retrospective test using a SASM test method provided by an embodiment of this application; Figure 7 is a schematic structural diagram of a system for verifying the performance of an earthquake prediction model provided by an embodiment of this application; Figure 8 is a schematic structural diagram of an electronic device provided by an embodiment of this application. Detailed implementation manners

[0025] The following description and drawings fully illustrate the specific implementation manners of this application, enabling those skilled in the art to practice them.

[0026] It should be clear that the described embodiments are only a part of the embodiments of this application, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application.

[0027] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with this application. On the contrary, they are only examples of systems and methods consistent with some aspects of this application as detailed in the appended claims.

[0028] In the description of the present application, it should be understood that the terms "first", "second", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances. In addition, in the description of the present application, unless otherwise specified, "a plurality of" means two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0029] The present application provides a method, a system, a medium, and an electronic device for verifying the performance of an earthquake prediction model to solve the problems existing in the above-related technical problems. In the embodiments of the present application, by accurately calculating and spatially allocating the seismic moment and area of historical earthquakes, a seismic moment spatial distribution map is generated, and then the performance of the earthquake prediction model is quantitatively evaluated. The seismic moment spatial distribution map can comprehensively characterize the earthquake influence range and the characteristics of seismic energy release, thereby improving the accuracy of performance verification. At the same time, by using the law of seismic energy release and the affected area, the seismic moment reflecting the size of seismic energy is proportionally allocated to the image area of the earthquake, realizing a comprehensive verification of the prediction ability of the model for the earthquake source area. The following will be described in detail with exemplary embodiments.

[0030] The following will be combined with the attached Figure 1 - attached Figure 6 to introduce in detail the method for verifying the performance of the earthquake prediction model provided by the embodiments of the present application. This method can be implemented depending on a computer program and can run on a performance verification system of an earthquake prediction model based on the von Neumann architecture. This computer program can be integrated into an application or run as an independent tool class application.

[0031] Please refer to Figure 1 which is a schematic flowchart of a method for verifying the performance of an earthquake prediction model provided by the embodiments of the present application. As Figure 1 shown, the method of the embodiments of the present application may include the following steps: S101, obtain the earthquake catalog of each historical earthquake in the target area where the magnitude is greater than the preset magnitude threshold within the preset time period; Among them, the preset time period refers to a time range determined in advance when conducting earthquake research and analysis, such as the past 50 years, 100 years, or specific decades. The magnitude of an earthquake is an indicator to measure the amount of energy released by the earthquake, expressed by the Richter magnitude (M). The preset magnitude threshold is a magnitude value preset in the research, used to screen earthquake events in the earthquake catalog that reach or exceed this magnitude. The target area: refers to the geographical area where earthquake research is conducted, which can be a country, a province, or a smaller area. The earthquake catalog of historical earthquakes records the detailed information of all earthquake events that occurred in the target area within the preset time period, including magnitude, epicenter, focal depth, etc.

[0032] In some embodiments of the present application, data of all earthquake events that occurred in the southeastern Tibetan Plateau in the past 100 years are collected from the seismic network, historical records, and relevant databases. Earthquake events with a magnitude greater than or equal to 6.0 are screened out from the collected data to form the earthquake catalog of each historical earthquake.

[0033] For example, earthquake data with a magnitude of 6 or above that occurred between AD 624 and 2015 on the southeastern margin of the Tibetan Plateau are collected. Earthquake data within China mainly come from the China Earthquake Networks Center (CENC), the Department of Disaster Prevention of the China Earthquake Administration (1995, 1999), and Xu and Gao (2014). As for data in places such as Myanmar and Laos, they come from the International Seismological Centre (ISC), the Global Centroid Moment Tensor (GCMT) Project, and Wang et al. (2014).

[0034] Given that there are many sources of earthquake data and the magnitude scales of each source are not unified, it is first necessary to unify the magnitude scale. More than 90% of the earthquake catalog data used here come from Chinese official records, and the most commonly used magnitude type in Chinese official documents is the surface wave magnitude (MS). In order to avoid introducing too much error in the magnitude conversion, the surface wave magnitude (MS) is used as the target magnitude scale for unification. First, for the historical earthquake catalog before 1965, the magnitude is directly processed as MS because historical magnitudes often have uncertainties. Second, for instrumental earthquakes after 1965, using the different types of magnitude conversion relationships given by predecessors (Table 1), body wave magnitude (Mb), moment magnitude (Mw), etc. are converted into surface wave magnitude (MS).

[0035] Table 1

[0036] The seismic data used for testing are divided into two groups, one is the complete earthquake catalog and the other is the declustered earthquake catalog. Among them, the complete earthquake catalog is suitable for testing the prediction ability of the prediction model for the entire earthquake sequence (including aftershocks), and the declustered earthquake catalog is more suitable for evaluating the prediction ability of the prediction model for the main shock (i.e., independent earthquake events). We use the algorithm proposed by Gardner and Knopoff (1974) to remove aftershocks, foreshocks and earthquake swarms by magnitude bins. In this method, the spatial window of aftershocks is determined by and the time window is determined by . Where the unit of distance D is km and the unit of time T is yr.

[0037] After the above processing, a total of 180 earthquakes were selected to form the "complete earthquake catalog" test data, and 124 M S ≥6 earthquakes were selected to form the "declustered earthquake catalog" test data. Figure 2 The brown circles in

[0038] Figure 2 show the spatial distribution of the epicenters in the declustered earthquake catalog.

[0039] For example, Figure 3 as shown, (a) Molchan test alarms at different thresholds during the spatial discretization of the alarm function. The stars represent the epicenters; (b) Molchan error map; (c) SASM test alarms at different thresholds in the spatial discretization of the alarm function. The ellipse represents the earthquake area; (d) SASM error map; (e) Seismic intensity map of the Ms7.4 Madoi earthquake in northeastern Qinghai-Tibet Plateau, China in 2021. The coseismic isoseismals on the horizontal plane are from the China Earthquake Administration; (f) Schematic diagram of the elliptical model of the earthquake area; (g) Linear regression analysis of the major and minor axes of the elliptical model of large earthquakes that occurred in China from 70 BC to 2021.

[0040] S102, according to the earthquake catalog, calculate the seismic moment and earthquake area of each historical earthquake; Among them, the earthquake catalog includes the magnitude and the epicenter location.

[0041] In the embodiments of the present application, the specific process of calculating the seismic moment and seismic area of each historical earthquake according to the earthquake catalog includes: calculating the seismic moment of each historical earthquake according to the magnitude and a preset magnitude empirical formula; taking the epicenter position as the center of the ellipse to obtain the center of each historical earthquake; calculating the fault length of each historical earthquake according to the magnitude and the relational expression between the preset fault length and the moment magnitude as the major axis of the ellipse; multiplying the major axis by a preset coefficient to obtain the minor axis; constructing an ellipse for each historical earthquake based on the center, major axis, direction of the major axis, and minor axis of each historical earthquake; the direction of the major axis is consistent with the fault direction of each historical earthquake; calculating the area of the ellipse of each historical earthquake to obtain the total seismic area of each historical earthquake, and the seismic area is used to characterize the perceived intensity and surface damage area of each historical earthquake.

[0042] Specifically, the preset magnitude empirical formula is:

[0043] Wherein, is the seismic moment, is the magnitude; The relational expression between the preset fault length and the moment magnitude is:

[0044] Wherein, is the fault length.

[0045] It should be noted that the direction of seismic rupture on the fault may be random. For bilateral rupture starting from the center of the fault, for simplicity, it is assumed that the major axis (D) is equal to the fault length (L). For various seismic rupture cases, the most extreme case is unilateral rupture starting from one end of the fault. In this case, the semi-major axis (R) is equal to L: Case 1 (D = L) and Case 2 (D = 2L).

[0046] Among them, the seismic intensity map reflects the distribution of seismic energy on the earth's surface and is in good agreement with the defined seismic volume area, so the seismic volume area can be estimated accordingly. The intensity distributions of different earthquakes are different. Long-term observations show that people usually use the ellipse model to describe the seismic intensity distribution (Chen and Liu, 1989). For example, the intensity map of the Madoi M7.4 earthquake in northeastern Qinghai-Tibet Plateau, China in 2021 ( Figure 3 in e) clearly shows elliptical co-seismic isoseismals on the horizontal plane. Therefore, the ellipse model is used to represent the distribution of the seismic volume area ( Figure 3In Figure f). The geometric center of the ellipse is the epicenter, and its (semi-)major axis is parallel to the fault direction (Hong and Feng, 2019). The length of the (semi-)major axis is determined by the fault length (L). The length of the (semi-)minor axis of the ellipse is determined according to the empirical relationship between the major axis and the minor axis in the seismic intensity map of the study area. The data of the major axis and minor axis of the isoseismal ellipses of 64 M≥7.0 earthquakes in China from 70 BC to 2021 AD were collected. Linear regression analysis was performed on the major axis and minor axis of the 64 earthquakes (Figure 3g), and the Pearson correlation coefficient was 0.98 and the determination coefficient was 0.96, indicating a strong positive correlation between the major axis and the minor axis. The relationship between the major axis and the minor axis is: minor axis = 0.48 * major axis. Without fault information, a circular model is used to approximate the earthquake area, and the radius of the circle is R = L / 2.

[0047] Among them, after the above processing, the seismic area distribution of historical earthquakes with a magnitude of 6 or above in the southeastern Tibetan Plateau is shown in Figure 4. The distribution of the seismic areas of historical earthquakes in the southeastern Tibetan Plateau. The red and blue circles represent the seismic areas. (a) Elliptical area model with the major axis equal to the fault length (D = L). (b) Circular area model with the major axis equal to the fault length (D = L). (c) Elliptical area model with the semi-major axis equal to the fault length (D = 2L). (d) Circular area model with the semi-major axis equal to the fault length (D = 2L).

[0048] S103, through the total seismic area of each historical earthquake, distribute the seismic moment of each historical earthquake to the longitude and latitude grid cells belonging to the target area to obtain the spatial distribution map of the seismic moment of the target area; In the embodiment of the present application, the specific process of distributing the seismic moment of each historical earthquake to the longitude and latitude grid cells belonging to the target area through the total seismic area of each historical earthquake to obtain the spatial distribution map of the seismic moment of the target area includes: mapping the ellipse of each historical earthquake to the longitude and latitude grid cells of the target area; determining the seismic area of each historical earthquake in each longitude and latitude grid cell according to the total seismic area of each historical earthquake from the longitude and latitude grid cells of the target area; calculating the ratio of the seismic area of each historical earthquake in each longitude and latitude grid cell to the total seismic area of each historical earthquake; calculating the product of the seismic moment of each historical earthquake and the ratio to obtain the seismic moment assigned to each longitude and latitude grid cell; distributing the seismic moment assigned to each longitude and latitude grid cell to each longitude and latitude grid cell to obtain the spatial distribution map of the seismic moment of the target area.

[0049] The calculation formula for the seismic moment assigned to each longitude and latitude grid cell is:

[0050] Where is the The seismic moment assigned to each latitude-longitude grid cell is the seismic moment of each historical earthquake.

[0051] Among them, when assigning the seismic moment to spatial grid cells (such as Figure 3 in c)), the assignment principle follows that the seismic moment assigned to each cell is proportional to the seismic area in each cell. Finally, the spatial distribution of the seismic moment used for retrospective testing in the SVSM test is as shown Figure 2 by the colored squares in.

[0052] S104. According to the spatial distribution map of the seismic moment, calculate the false alarm rate and the abnormal spatio-temporal occupancy rate under different prediction index thresholds of the preset earthquake prediction model; In the embodiment of the present application, under the conditions of different prediction index thresholds of the preset earthquake prediction model, calculate the proportion of the alarm area (𝜏) and the non-prediction rate of the seismic moment ( ) under each threshold.

[0053] The calculation formula for the false alarm rate is:

[0054] The calculation formula for the abnormal spatio-temporal occupancy rate is:

[0055] Among them, is the index quantity of the preset earthquake prediction model for the th latitude-longitude grid cell, that is, in the preset earthquake prediction model, can be the earthquake prediction probability of the th latitude-longitude grid cell, is the threshold of the index quantity of the preset earthquake prediction model, is the th value of the seismic moment within the th latitude-longitude grid cell. The value ranges of the seismic moment false alarm rate

[0056] and the abnormal spatio-temporal occupancy rate τ are both 0 - 1. Among them, the preset SASM error map is a coordinate map established based on the false alarm rate and the abnormal spatio-temporal occupancy rate, and fit the visualization curves corresponding to the false alarm rate and the abnormal spatio-temporal occupancy rate under different prediction index thresholds, such as Figure 3 b and d in.

[0057] S106. According to the visualization curve, judge whether the performance of the preset earthquake prediction model is qualified.

[0058] Among them, the preset SASM error graph includes a dotted line with Gain = 1 and and the boundary lines; Gain is the probability gain, is the abnormal spatio-temporal occupancy rate, is the false alarm rate.

[0059] In some embodiments of the present application, the specific process of determining whether the performance of the preset earthquake prediction model is qualified according to the visualization curve includes: when the visualization curve is below the dotted line, it indicates that the performance of the preset earthquake prediction model is better than random prediction, so as to determine that the performance of the preset earthquake prediction model is qualified; or, when the area enclosed by the visualization curve and the boundary line is greater than the preset area threshold, it is determined that the performance of the preset earthquake prediction model is qualified; or, the proportion of the seismic moment captured in the preset high-probability area is statistically analyzed, and when the seismic moment proportion meets the preset conditions, it is determined that the performance of the preset earthquake prediction model is qualified.

[0060] For example Figure 3 b and d in, the closer the (𝜏, ) curve is to the Y-axis, the better the prediction performance of its model. The area enclosed by the (𝜏, ) curve and the boundary lines 𝜏 = 1 and = 1 of the SASM graph is called the area skill score ASS. The larger the area, the more concentrated the earthquakes are in the cells with a higher earthquake probability given by the prediction model, that is, the larger the ASS, the better the model prediction performance. That is, the proportion of the seismic moment captured in the high-probability area is statistically analyzed (such as the proportion of the seismic moment in the first 30% probability grid cells, that is, in the SASM graph, when T = 0.3, the smaller the value of Vm, the better).

[0061] For example, the RELM-TibetSE model from Wei et al. (2023) was tested by the method of the present application. The RELM-TibetSE model ( Figure 5 ) is a time-independent and grid-based prediction model for M≥6.0 earthquakes in the southeastern Tibetan Plateau established using GNSS strain rate, and this model predicts the earthquake occurrence probability spatially. Therefore, in this model, the alarm function is the earthquake occurrence probability. Figure 5 The colored cells in visualize the numerical values of the alarm function. The UNI 1-3 model and the VARI 1-3 model are sub-models with different parameter variables in RELM-TibetSE.

[0062] It should be noted that in Figure 5In it, the SASM error map of the RELM-TibetSE model. (a-f) Test results using the major axis of the ellipse or the diameter of the circle equal to the fault length (D = L). (g-l) Test results using the semi-major axis or the radius equal to the fault length (D = 2L). The (𝜏, ) curve represented by the solid black (gray) line represents the results of the Molchan test, and its Y value is normalized by the number of earthquakes. The (𝜏, ) curve represented by the solid red stepped curve is the result of the SASM test, and its Y value is normalized by the elliptical seismic moment. The (𝜏, ) curve represented by the solid blue stepped curve is the result of the SASM test, and its Y value is normalized by the circular seismic moment. The dashed line represents the 5% confidence level.

[0063] Figure 6 The process of evaluating the sub-model UNI-1 of RELM-TibetSE using SVSM-test is listed. The alarm function values of the UNI-1 model are as shown in Figure 6 a. Figure 6 b-d in it show the alarm areas of three decreasing thresholds, Figure 6 and e in it is the corresponding SVSM error map. SVSM-test calculates the spatio-temporal occupancy rate of anomalies ( ) and the corresponding false alarm rate respectively by continuously reducing the probability threshold of the prediction result. The false alarm rates obtained by the two test methods are plotted as the SVSM error map as shown in Figure 5 .

[0064] In the embodiment of the present application, by accurately calculating and spatially allocating the seismic moment and area of historical earthquakes, a seismic moment spatial distribution map is generated, and then the performance of the earthquake prediction model is quantitatively evaluated. The seismic moment spatial distribution map can comprehensively characterize the earthquake influence range and the characteristics of seismic energy release, thereby improving the accuracy of performance verification. At the same time, by using the law of seismic energy release and the affected area, the seismic moment reflecting the size of seismic energy is proportionally allocated to the image area of the earthquake, realizing a comprehensive verification of the prediction ability of the model for the earthquake source area.

[0065] The following is the system embodiment of the present application, which can be used to execute the method embodiment of the present application. For the details not disclosed in the system embodiment of the present application, please refer to the method embodiment of the present application.

[0066] Please refer to Figure 7, which shows a schematic structural diagram of a performance verification system for an earthquake prediction model provided by an exemplary embodiment of the present application. The performance verification system for the earthquake prediction model can be implemented as all or part of an electronic device through software, hardware, or a combination of both. The system 1 includes an earthquake catalog acquisition module 10, an earthquake moment and earthquake area calculation module 20, an earthquake moment spatial distribution map generation module 30, a false alarm rate and abnormal spatio-temporal occupancy rate calculation module 40, a visualization curve fitting module 50, and a performance determination module 60.

[0067] The earthquake catalog acquisition module 10 is configured to acquire the earthquake catalog of each historical earthquake in a target area where the magnitude is greater than a preset magnitude threshold within a preset time period. The earthquake moment and earthquake area calculation module 20 is configured to calculate the earthquake moment and earthquake area of each historical earthquake according to the earthquake catalog. The earthquake moment spatial distribution map generation module 30 is configured to distribute the earthquake moment of each historical earthquake to the longitude and latitude grid cells belonging to the target area through the total earthquake area of each historical earthquake, so as to obtain the earthquake moment spatial distribution map of the target area. The false alarm rate and abnormal spatio-temporal occupancy rate calculation module 40 is configured to calculate the false alarm rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds of a preset earthquake prediction model according to the earthquake moment spatial distribution map. The visualization curve fitting module 50 is configured to fit the visualization curves corresponding to the false alarm rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds in a preset SASM error map. The performance determination module 60 is configured to determine whether the performance of the preset earthquake prediction model is qualified according to the visualization curve.

[0068] It should be noted that when the performance verification system for the earthquake prediction model provided in the above embodiment executes the performance verification method for the earthquake prediction model, only the above-mentioned division of each functional module is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the performance verification system for the earthquake prediction model provided in the above embodiment and the embodiment of the performance verification method for the earthquake prediction model belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0069] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0070] In the embodiments of the present application, by precisely calculating and spatially allocating the seismic moment and area of historical earthquakes, a seismic moment spatial distribution map is generated, and then the performance of the earthquake prediction model is quantitatively evaluated. The seismic moment spatial distribution map can comprehensively characterize the earthquake influence range and the characteristics of seismic energy release, thereby improving the accuracy of performance verification. At the same time, by using the law of seismic energy release and the affected area, the seismic moment reflecting the magnitude of seismic energy is proportionally allocated to the earthquake's image area to comprehensively verify the model's prediction ability for the earthquake source area.

[0071] The present application also provides a computer-readable medium, on which program instructions are stored. When the program instructions are executed by a processor, the method for verifying the performance of the earthquake prediction model provided in each of the above method embodiments is implemented.

[0072] The present application also provides a computer program product containing instructions. When it runs on a computer, the computer is caused to execute the method for verifying the performance of the earthquake prediction model provided in each of the above method embodiments.

[0073] Please refer to Figure 8 , which is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 8 shown, the electronic device 1000 may include: at least one processor 1001, at least one network interface 1004, a user interface 1003, a memory 1005, and at least one communication bus 1002.

[0074] Among them, the communication bus 1002 is used to implement connection communication between these components.

[0075] Among them, the user interface 1003 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 1003 may further include a standard wired interface and a wireless interface.

[0076] Among them, the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0077] Among them, the processor 1001 may include one or more processing cores. The processor 1001 connects various parts within the entire electronic device 1000 through various interfaces and circuits. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, and by calling the data stored in the memory 1005, it executes various functions of the electronic device 1000 and processes data. Optionally, the processor 1001 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 1001 may integrate a combination of one or several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001 and may be implemented separately by a single chip.

[0078] Among them, the memory 1005 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 1005 may include a program storage area and a data storage area. Among them, the program storage area can store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store the data involved in the above-mentioned various method embodiments. Optionally, the memory 1005 may also be at least one storage system located far from the aforementioned processor 1001. As Figure 8 shown, the memory 1005, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a performance verification application program for earthquake prediction models.

[0079] In Figure 8In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an interface for the user to input and obtain the data input by the user; and the processor 1001 can be used to call the performance verification application program of the earthquake prediction model stored in the memory 1005 and specifically perform the following operations: Obtain the earthquake catalog of each historical earthquake in the target area where the magnitude is greater than the preset magnitude threshold within the preset time period; Calculate the seismic moment and seismic area of each historical earthquake according to the earthquake catalog; Allocate the seismic moment of each historical earthquake to the longitude and latitude grid cells belonging to the target area through the total seismic area of each historical earthquake to obtain the spatial distribution map of the seismic moment in the target area; Calculate the false alarm rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds of the preset earthquake prediction model according to the spatial distribution map of the seismic moment; Fit the visualization curves corresponding to the false alarm rate and abnormal spatio-temporal occupancy rate under different prediction index thresholds in the preset SASM error map; Judge whether the performance of the preset earthquake prediction model is qualified according to the visualization curve.

[0080] In one embodiment, when the processor 1001 executes to judge whether the performance of the preset earthquake prediction model is qualified according to the visualization curve, it specifically performs the following operations: In the case that the visualization curve is below the dotted line, it indicates that the performance of the preset earthquake prediction model is better than random prediction, so as to determine that the performance of the preset earthquake prediction model is qualified; or, In the case that the area enclosed by the visualization curve and the boundary line is greater than the preset area threshold, determine that the performance of the preset earthquake prediction model is qualified; or, Statistically analyze the proportion of the seismic moment captured in the preset high-probability area, and determine that the performance of the preset earthquake prediction model is qualified when the seismic moment proportion meets the preset conditions.

[0081] In one embodiment, when the processor 1001 executes to calculate the seismic moment and seismic area of each historical earthquake according to the earthquake catalog, it specifically performs the following operations: Calculate the seismic moment of each historical earthquake according to the magnitude and the preset magnitude empirical formula; Use the epicenter position as the center of the ellipse to obtain the center of each historical earthquake; Calculate the fault length of each historical earthquake according to the magnitude and the relational formula between the preset fault length and the moment magnitude, as the major axis of the ellipse; Multiply the major axis by the preset coefficient to obtain the minor axis; Based on the center, major axis, direction of the major axis, and minor axis of each historical earthquake, construct the ellipse of each historical earthquake; the direction of the major axis is consistent with the fault direction of each historical earthquake; Calculate the area of the ellipse for each historical earthquake to obtain the total earthquake area for each historical earthquake. The earthquake area is used to characterize the perceived intensity and surface damage area of each historical earthquake.

[0082] In one embodiment, when the processor 1001 executes to distribute the seismic moment of each historical earthquake to the longitude and latitude grid cells belonging to the target area through the total earthquake area of each historical earthquake to obtain the spatial distribution map of the seismic moment in the target area, the following operations are specifically performed: Map the ellipse of each historical earthquake to the longitude and latitude grid cells of the target area; Determine the earthquake area of each historical earthquake in each longitude and latitude grid cell from the longitude and latitude grid cells of the target area according to the total earthquake area of each historical earthquake; Calculate the ratio of the earthquake area of each historical earthquake in each longitude and latitude grid cell to the total earthquake area of each historical earthquake; Calculate the product of the seismic moment of each historical earthquake and the ratio to obtain the seismic moment assigned to each longitude and latitude grid cell; Distribute the seismic moment assigned to each longitude and latitude grid cell to each longitude and latitude grid cell to obtain the spatial distribution map of the seismic moment in the target area.

[0083] In the embodiments of the present application, by accurately calculating and spatially distributing the seismic moment and area of historical earthquakes, a spatial distribution map of the seismic moment is generated, and then the performance of the earthquake prediction model is quantitatively evaluated. The spatial distribution map of the seismic moment can comprehensively characterize the earthquake influence range and the characteristics of seismic energy release, thereby improving the accuracy of performance verification. At the same time, by using the law of seismic energy release and the affected area, the seismic moment reflecting the size of seismic energy is proportionally distributed to the image area of the earthquake, realizing a comprehensive verification of the prediction ability of the model for the earthquake source area.

[0084] Those of ordinary skill in the art can understand that all or part of the processes of implementing the above method embodiments can be completed by instructing relevant hardware through a computer program. The program for verifying the performance of the earthquake prediction model can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. Among them, the storage medium of the program for verifying the performance of the earthquake prediction model can be a magnetic disk, an optical disk, a read-only memory, or a random access memory, etc.

[0085] The above-disclosed are only the preferred embodiments of the present application. Of course, the scope of rights of the present application cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A method for verifying the performance of an earthquake prediction model, characterized in that: The method comprises: Obtain an earthquake catalog of each historical earthquake in the target area with a magnitude greater than a preset magnitude threshold within a preset time period; Calculating the seismic moment and seismic area of ​​each historical earthquake according to the earthquake catalog; According to the total earthquake area of ​​each historical earthquake, the seismic moment of each historical earthquake is distributed to the latitude and longitude grid cells belonging to the target area, so as to obtain a spatial distribution map of the seismic moment of the target area; According to the seismic moment spatial distribution diagram, calculating the underreporting rate and abnormal space-time occupancy rate under different prediction index thresholds of the preset earthquake prediction model; In the preset SASM error graph, a visualization curve corresponding to the underreporting rate and abnormal spatiotemporal occupancy rate under the different prediction index thresholds is fitted; According to the visualization curve, it is judged whether the performance of the preset earthquake prediction model is qualified.

2. The method according to claim 1, characterized in that: The preset SASM error graph includes a dotted line with Gain=1 and and The boundary line of ; Gain is the probability gain, is the abnormal space-time occupancy, is the underreporting rate; The step of judging whether the performance of the preset earthquake prediction model is qualified according to the visualization curve includes: When the visualization curve is lower than the dotted line, it indicates that the performance of the preset earthquake prediction model is better than random prediction, so as to determine that the performance of the preset earthquake prediction model is qualified; or In the case where the area enclosed by the visualization curve and the boundary line is greater than a preset area threshold, determining that the performance of the preset earthquake prediction model is qualified; or, The ratio of the moments captured in the preset high probability area is statistically collected, and when the ratio of the moments meets the preset conditions, it is determined that the performance of the preset earthquake prediction model is qualified.

3. The method according to claim 1, characterized in that The earthquake catalog includes magnitude and epicenter location; According to the earthquake catalog, calculating the seismic moment and seismic area of ​​each historical earthquake, including: Calculating the seismic moment of each historical earthquake according to the magnitude and a preset empirical formula for magnitude; Taking the epicenter position as the center of the ellipse, the center of each historical earthquake is obtained; According to the magnitude and a relationship between the preset fault length and the moment magnitude, the fault length of each historical earthquake is calculated as the major axis of the ellipse; The major axis is multiplied by a preset coefficient to obtain a minor axis; Based on the center, the major axis, the direction of the major axis and the minor axis of each historical earthquake, an ellipse of each historical earthquake is constructed; the direction of the major axis is consistent with the fault direction of each historical earthquake; The area of ​​the ellipse of each historical earthquake is calculated to obtain the total earthquake area of ​​each historical earthquake, and the earthquake area is used to characterize the perceived intensity and surface damage area of ​​each historical earthquake.

4. The method according to claim 3, characterized in that The preset earthquake magnitude empirical formula is: in, is the seismic moment, is the magnitude; The relationship between the preset fault length and moment magnitude is: in, is the fault length.

5. The method according to claim 3, characterized in that: The method allocates the seismic moment of each historical earthquake to the latitude and longitude grid cells belonging to the target area through the total seismic area of ​​each historical earthquake to obtain the seismic moment spatial distribution map of the target area, including: Mapping the ellipse of each historical earthquake to the latitude and longitude grid cells of the target area; Determine, from the latitude and longitude grid cells of the target area, the seismic area of ​​each historical earthquake in each latitude and longitude grid cell according to the total seismic area of ​​each historical earthquake; Calculate the ratio of the earthquake area of ​​each historical earthquake in each latitude and longitude grid unit to the total earthquake area of ​​each historical earthquake; Calculate the product of the seismic moment of each historical earthquake and the ratio to obtain the seismic moment assigned to each longitude and latitude grid unit; The seismic moment allocated to each latitude and longitude grid unit is allocated to each latitude and longitude grid unit to obtain a spatial distribution map of the seismic moment of the target area.

6. The method according to claim 5, characterized in that The calculation formula for the seismic moment assigned to each latitude and longitude grid unit is: in, For the The seismic moment assigned to each latitude and longitude grid cell, is the seismic moment of each historical earthquake.

7. The method according to claim 1, characterized in that The calculation formula of the false negative rate is: The calculation formula of the abnormal time and space occupancy rate is: in, For the The index of the preset earthquake prediction model for the latitude and longitude grid unit, that is, in the preset earthquake prediction model, Can be the first The earthquake prediction probability of each longitude and latitude grid cell is is the threshold value of the indicator quantity of the preset earthquake prediction model, For the The value of the seismic moment within the longitude and latitude grid unit, the underreporting rate of the seismic moment The value range of and abnormal space-time occupancy τ are both 0-1.

8. A performance verification system for an earthquake prediction model, characterized in that: The system comprises: An earthquake catalog acquisition module is used to acquire an earthquake catalog of each historical earthquake in a target area with a magnitude greater than a preset magnitude threshold within a preset time period; A seismic moment and seismic area calculation module, used for calculating the seismic moment and seismic area of ​​each historical earthquake according to the earthquake catalog; A seismic moment spatial distribution map generating module, used to distribute the seismic moment of each historical earthquake to the latitude and longitude grid cells belonging to the target area according to the total seismic area of ​​each historical earthquake, so as to obtain the seismic moment spatial distribution map of the target area; A missing alarm rate and abnormal space-time occupancy rate calculation module, used to calculate the missing alarm rate and abnormal space-time occupancy rate under different prediction index thresholds of a preset earthquake prediction model according to the earthquake moment spatial distribution diagram; A visualization curve fitting module is used to fit the visualization curves corresponding to the underreporting rate and abnormal space-time occupancy rate under the different prediction index thresholds in the preset SASM error graph; The performance determination module is used to determine whether the performance of the preset earthquake prediction model is qualified according to the visualization curve.

9. A computer storage medium, characterized in that: The computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method according to any one of claims 1 to 7.

Images