Earthquake type identification method and device

Abstract

The application provides a kind of earthquake type identification method and device, it is related to the technical field of earthquake classification, after obtaining the earthquake data to be identified, first, the target feature of the earthquake data is extracted;Wherein, the signal characteristics that target feature is the difference of natural earthquake and non-natural earthquake exists, then the target Gaussian naive Bayes network model is used to process target feature, to obtain the estimated probability that the earthquake data belongs to natural earthquake, finally, the target calibration function is used to calibrate the estimated probability, to obtain the target probability that the earthquake data belongs to natural earthquake, and based on target probability, determine the earthquake type to which the earthquake data belongs.In view of the target calibration function is the function that is fitted out with the least deviation square sum of the estimated probability mean of multiple target feature sample groups and the real earthquake probability as the target, therefore, it can be guaranteed that the earthquake type determined by the method of the application has high accuracy and confidence.

Description

Earthquake type identification method and device Technical Field

[0001] This invention relates to the technical field of earthquake classification, and in particular to a method and apparatus for earthquake type identification. Background Technology

[0002] With the establishment of dense global and national seismic monitoring networks, various non-natural vibration events will be recorded. These records will be mixed with natural earthquake records, and if they are not removed in time, they will obscure the earthquake catalog and affect subsequent seismic hazard analysis. Therefore, resolving the earthquake classification problem in regional and local seismic records is an important task in earthquake monitoring in the future.

[0003] Theoretically, the focal mechanisms of natural earthquakes and man-made explosions differ significantly. However, due to the large spacing between stations in fixed seismic networks, solving for the focal parameters of low-magnitude earthquakes (M < 3.0) and regional high-frequency attenuation models using local and regional seismic records remains a challenge. Therefore, artificial intelligence technologies such as machine learning have been widely applied in earthquake and explosion identification. However, in machine learning scenarios, especially in classification tasks, existing technologies output probabilities that the model predicts based on the training set and model framework, indicating whether a sample belongs to a positive or negative class. Furthermore, the model's predictions may differ significantly from the actual type of the event. In summary, existing earthquake type identification methods suffer from low accuracy and poor reliability in their predictions. Summary of the Invention

[0004] The purpose of this invention is to provide an earthquake type identification method and apparatus to alleviate the technical problems of low accuracy and poor reliability of prediction results in existing earthquake type identification methods.

[0005] In a first aspect, the present invention provides an earthquake type identification method, comprising: acquiring earthquake data to be identified; extracting target features from the earthquake data; wherein the target features are signal features that differ between natural and non-natural earthquakes; processing the target features using a target Gaussian Naive Bayes network model to obtain an estimated probability that the earthquake data belongs to a natural earthquake; calibrating the estimated probability using a target calibration function to obtain a target probability that the earthquake data belongs to a natural earthquake, thereby determining the earthquake type to which the earthquake data belongs based on the target probability; wherein the target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability.

[0006] Optionally, the target features include at least one of the following: P / S amplitude ratio of a specified frequency band, P-wave high-low frequency energy ratio, S-wave high-low frequency energy ratio, P-wave corner frequency based on Brune source spectrum model, S-wave corner frequency based on Brune source spectrum model, P-wave corner frequency based on Jimenez source spectrum model, S-wave corner frequency based on Jimenez source spectrum model, waveform duration, P-wave waveform complexity, S-wave waveform complexity, P-wave zero-crossing rate, S-wave zero-crossing rate, P-wave cepstral complexity, S-wave cepstral complexity, P-wave instantaneous frequency complexity, and S-wave instantaneous frequency complexity.

[0007] Optionally, before processing the target features using the target Gaussian Naive Bayes network model, the method further includes: acquiring a set of earthquake data samples; wherein each earthquake data sample in the set has a real earthquake type label; if the earthquake data sample belongs to a natural earthquake, the real earthquake type label is 0, otherwise it is 1; extracting the target features of each earthquake data sample in the set to obtain a target feature set; wherein each target feature in the target feature set corresponds to a real earthquake type label; and iteratively training the initial Gaussian Naive Bayes network model using the target feature set until a preset termination condition is reached to obtain the target Gaussian Naive Bayes network model.

[0008] Optionally, before calibrating the estimated probability using the target calibration function, the method further includes: obtaining a target feature sample set; wherein each target feature sample in the target feature sample set has a true earthquake type label; processing each target feature sample in the target feature sample set using a target Gaussian Naive Bayes network model to determine the estimated probability that each target feature sample belongs to a natural earthquake; dividing the target feature sample set into a specified number of target feature sample groups according to the value of the estimated probability, and calculating the mean of the estimated probability and the true earthquake probability of each target feature sample group to obtain multiple sets of probability calibration samples; wherein, The estimated probabilities corresponding to each target feature sample group are within the same value range; the true earthquake probability represents the proportion of the true earthquake type labeled as a natural earthquake in the target feature sample group; each probability calibration sample group includes: the mean of the estimated probability and the true earthquake probability; the multiple probability calibration samples are sorted in ascending order based on the magnitude of the mean of the estimated probability to obtain the sorted probability calibration sample group; the sorted probability calibration sample group is fitted using a monotonic regression algorithm to obtain the target calibration function that minimizes the objective function; wherein, the objective function is the sum of squared deviations of multiple mean estimated probabilities and the corresponding true earthquake probabilities.

[0009] Optionally, the objective function can be expressed as: Among them, w i Let y represent the weights, i = 1, 2, ..., n, where n represents the number of probability calibration sample groups. iLet x represent the true earthquake probability in the i-th probability calibration sample. i Let represent the mean of the predicted probabilities in the i-th probability calibration sample.

[0010] Secondly, the present invention provides an earthquake type identification device, comprising: an acquisition module for acquiring earthquake data to be identified; an extraction module for extracting target features from the earthquake data; wherein the target features are signal features that differ between natural and non-natural earthquakes; a processing module for processing the target features using a target Gaussian Naive Bayes network model to obtain an estimated probability that the earthquake data belongs to a natural earthquake; and a calibration module for calibrating the estimated probability using a target calibration function to obtain a target probability that the earthquake data belongs to a natural earthquake, so as to determine the earthquake type to which the earthquake data belongs based on the target probability; wherein the target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability.

[0011] Optionally, the target features include at least one of the following: P / S amplitude ratio of a specified frequency band, P-wave high-low frequency energy ratio, S-wave high-low frequency energy ratio, P-wave corner frequency based on Brune source spectrum model, S-wave corner frequency based on Brune source spectrum model, P-wave corner frequency based on Jimenez source spectrum model, S-wave corner frequency based on Jimenez source spectrum model, waveform duration, P-wave waveform complexity, S-wave waveform complexity, P-wave zero-crossing rate, S-wave zero-crossing rate, P-wave cepstral complexity, S-wave cepstral complexity, P-wave instantaneous frequency complexity, and S-wave instantaneous frequency complexity.

[0012] Optionally, the earthquake type identification device is also used to: acquire an earthquake data sample set; wherein each earthquake data sample in the earthquake data sample set has a real earthquake type label; if the earthquake data sample belongs to a natural earthquake, the real earthquake type label is 0, otherwise it is 1; extract the target features of each earthquake data sample in the earthquake data sample set to obtain a target feature set; wherein each target feature in the target feature set corresponds to a real earthquake type label; use the target feature set to iteratively train the initial Gaussian Naive Bayes network model until a preset termination condition is reached to obtain the target Gaussian Naive Bayes network model.

[0013] Thirdly, the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the earthquake type identification method of any of the foregoing embodiments.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the earthquake type identification method of any of the foregoing embodiments.

[0015] This invention provides a method for earthquake type identification. After acquiring the earthquake data to be identified, the method first extracts target features from the earthquake data. These target features are signal characteristics that differ between natural and non-natural earthquakes. Then, a target Gaussian Naive Bayes network model is used to process the target features to obtain an estimated probability that the earthquake data belongs to a natural earthquake. Finally, a target calibration function is used to calibrate the estimated probability to obtain the target probability that the earthquake data belongs to a natural earthquake, and the earthquake type is determined based on the target probability. Since the target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability, the earthquake type determined by the method of this invention can be guaranteed to have high accuracy and confidence. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 is a flowchart of an earthquake type identification method provided by an embodiment of the present invention;

[0018] Figure 2 is a schematic diagram showing the distribution of 36 characteristics of natural and non-natural earthquakes according to an embodiment of the present invention;

[0019] Figure 3 is a schematic diagram of the ROC curve predicted by the target Gaussian Naive Bayes network model;

[0020] Figure 4 is a schematic diagram of the prediction confusion matrix results of the target Gaussian Naive Bayes network model;

[0021] Figure 5 is a comparison of the earthquake classification effects of the method of the present invention and two comparative methods;

[0022] Figure 6 is a functional block diagram of an earthquake type identification device provided in an embodiment of the present invention;

[0023] Figure 7 is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0025] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0026] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0027] Example 1

[0028] Figure 1 is a flowchart of an earthquake type identification method provided by an embodiment of the present invention. As shown in Figure 1, the method specifically includes the following steps:

[0029] Step S102: Obtain the earthquake data to be identified.

[0030] Step S104: Extract target features from seismic data.

[0031] Among them, the target features are the signal characteristics that differ between natural earthquakes and non-natural earthquakes.

[0032] Specifically, the earthquake data to be identified is the earthquake record data recorded by earthquake monitoring stations. After acquiring the earthquake data, digital signal processing methods and seismological theory are used to process the earthquake data, including removing instrument response, detrending, filtering, removing abnormal records and waveform data with low signal-to-noise ratio, and then extracting target features. In this embodiment of the invention, to assist in identifying natural and non-natural earthquake events, the target features are selected based on signal characteristics that differ between natural and non-natural earthquakes to highlight the true earthquake type. The target features include time-domain features and frequency-domain features. This embodiment of the invention does not specifically limit the type and number of target features; users can set them according to their actual needs.

[0033] Step S106: The target features are processed using the target Gaussian Naive Bayes network model to obtain the estimated probability that the earthquake data belongs to a natural earthquake.

[0034] Specifically, an initial Gaussian Naive Bayes network model can be constructed using the sklearn library. This initial model is then trained, and its parameters are continuously optimized until the preset performance requirements are met, thus obtaining the target Gaussian Naive Bayes network model. The input data for the target Gaussian Naive Bayes network model is the target feature, and the output data is the estimated probability that the earthquake data corresponding to the target feature belongs to a natural earthquake. The sum of the estimated probabilities of natural earthquakes and non-natural earthquakes is 1. If the estimated probability of natural earthquakes is greater than the estimated probability of non-natural earthquakes, it indicates that the target Gaussian Naive Bayes network model has preliminarily determined that the earthquake data belongs to the natural earthquake type. However, given the low accuracy and reliability of the target Gaussian Naive Bayes network model, this embodiment of the invention requires further calibration of the estimated probabilities output by the target Gaussian Naive Bayes network model.

[0035] Step S108: The estimated probability is calibrated using the target calibration function to obtain the target probability that the earthquake data belongs to a natural earthquake, so as to determine the earthquake type of the earthquake data based on the target probability.

[0036] The target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the predicted probabilities of multiple target feature sample groups and the actual earthquake probabilities.

[0037] In other words, the input data for the target calibration function is the output data of the target Gaussian Naive Bayes network model. The target calibration function processes the estimated probabilities, and its output probability is the target probability that the earthquake data belongs to a natural earthquake. Since the target calibration function is a function fitted to minimize the sum of squared deviations between the mean of a large number of estimated probabilities and the actual earthquake probability, the probability calibrated by the target calibration function has high confidence and accuracy. In this embodiment of the invention, after processing the earthquake data to be identified, the final earthquake type identification result includes both the earthquake type to which the earthquake data belongs and the uncertainty (i.e., probability) of belonging to that earthquake type. For example, if the target probability of the earthquake data belonging to a natural earthquake is determined to be 80%, the earthquake type identification result is: natural earthquake (80%); if the target probability of the earthquake data belonging to a natural earthquake is determined to be 30%, the earthquake type identification result is: non-natural earthquake (70%).

[0038] This invention provides a method for earthquake type identification. After acquiring the earthquake data to be identified, the method first extracts target features from the earthquake data. These target features are signal characteristics that differ between natural and non-natural earthquakes. Then, a target Gaussian Naive Bayes network model is used to process the target features to obtain an estimated probability that the earthquake data belongs to a natural earthquake. Finally, a target calibration function is used to calibrate the estimated probability to obtain the target probability that the earthquake data belongs to a natural earthquake, and the earthquake type is determined based on the target probability. Since the target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability, the earthquake type determined by the method of this invention has high confidence and accuracy.

[0039] In one alternative implementation, the target features include at least one of the following: P / S amplitude ratio of a specified frequency band, P-wave high-low frequency energy ratio, S-wave high-low frequency energy ratio, P-wave corner frequency based on Brune source spectrum model, S-wave corner frequency based on Brune source spectrum model, P-wave corner frequency based on Jimenez source spectrum model, S-wave corner frequency based on Jimenez source spectrum model, waveform duration, P-wave waveform complexity, S-wave waveform complexity, P-wave zero-crossing rate, S-wave zero-crossing rate, P-wave cepstral complexity, S-wave cepstral complexity, P-wave instantaneous frequency complexity, and S-wave instantaneous frequency complexity.

[0040] The more target features there are, the more stable the earthquake type identification results will be. To highlight the differences between different types of earthquake events, this embodiment of the invention proposes quantifying the above eight types of earthquake features. Users can calculate the P / S amplitude ratio of multiple frequency bands according to their actual needs. If the number of frequency bands is 21, Figure 2 shows a schematic diagram of the distribution of 36 features of natural and non-natural earthquakes provided by this embodiment of the invention. In Figure 2, artificial explosions represent non-natural earthquakes. The extraction methods for corner frequencies and waveform complexity are described below.

[0041] The corner frequency is known to be the intersection of the high-frequency and low-frequency asymptotic trends of the amplitude spectrum, and is also a physical quantity reflecting the size of the earthquake source. The seismic wave amplitude spectrum is a description of the seismic waves radiated by the source in the frequency domain; it is closely related to the mechanical parameters of the source, and different source models radiate different theoretical seismic wave amplitude spectra. For small to medium-sized earthquakes, the theoretical source spectrum conforms to the Brune model. Ignoring inelastic attenuation, the Brune model source spectrum Ω(f) is expressed as: Ω(f) = Ω0 / (1+(f / f)) c ) 2 The source spectrum Ω(f) of the Jimenez model is expressed as:

[0042] In the two expressions above, Ω0 represents the zero-frequency limit of the displacement spectrum, and f represents the frequency. c The corner frequency is represented by f. In this embodiment of the invention, the difference between the observed amplitude spectrum and the theoretical spectrum is used as the objective function, and the corner frequency f is calculated at the point where the deviation between the theoretical and actual values ​​is minimized. c Referring to the methods described above, the corner frequencies of P-waves and S-waves can be calculated using the Brune model and the Jimenez model, respectively.

[0043] It is generally believed that the P-waves radiated from an explosive source are simpler than those radiated from a natural earthquake source. This is because the energy of an explosive event is mainly concentrated in the early part of the time period, with the tail end suppressed, making the overall waveform simpler compared to a natural earthquake. In contrast, natural earthquakes contain various phases, with energy distribution throughout the early and late parts of the time period, resulting in a more complex overall waveform than an explosive event. The waveform complexity formula defined in this invention is as follows: Where C represents waveform complexity; if calculating the complexity of a P-wave, then t1 and t2 represent the start and end times of the P-wave, respectively; if calculating the complexity of an S-wave, then t1 and t2 represent the start and end times of the S-wave, respectively; t p t represents the arrival time of the P wave. coda The time it takes for the waveform to decay to the noise level is represented by y(t), and the amplitude of the waveform at time t is represented by y(t).

[0044] In an optional implementation, before processing the target features using a target Gaussian Naive Bayes network model, the method of the present invention further includes the following steps:

[0045] Step S201: Obtain the earthquake data sample set; wherein, each earthquake data sample in the earthquake data sample set has a real earthquake type label; if the earthquake data sample belongs to a natural earthquake, the real earthquake type label is 0, otherwise it is 1.

[0046] Step S202: Extract the target features of each earthquake data sample in the earthquake data sample set to obtain a target feature set; wherein, each target feature in the target feature set corresponds to a real earthquake type label.

[0047] Step S203: Iteratively train the initial Gaussian Naive Bayes network model using the target feature set until the preset termination condition is met, and obtain the target Gaussian Naive Bayes network model.

[0048] Specifically, before model training, this embodiment of the invention sets the earthquake data samples of natural earthquakes with a true earthquake type label of 0, and the earthquake data samples of non-natural earthquakes with a true earthquake type label of 1. After obtaining the earthquake data sample set, features are extracted from each sample, and then the target features of each sample are processed using an initial Gaussian Naive Bayes network model to predict the probability that it belongs to a natural earthquake. The loss function value is then calculated in combination with the true earthquake type label of the sample, and the model is trained using the loss function value. Model training can also be done in batches. When the model training reaches a preset number of rounds, or when the accuracy of the model prediction results reaches a specified threshold, training can be stopped, and the target Gaussian Naive Bayes network model is obtained.

[0049] Figure 3 shows the ROC curve of the target Gaussian Naive Bayes network model, with an AUC of 0.9763. Figure 4 shows the prediction confusion matrix results of the target Gaussian Naive Bayes network model, in which 574 earthquakes were accurately identified in class 0 (natural earthquakes) and 425 earthquakes were accurately identified in class 1 (non-natural earthquakes).

[0050] In an optional implementation, before calibrating the estimated probability using the target calibration function, the method of the present invention further includes the following steps:

[0051] Step S301: Obtain the target feature sample set; wherein each target feature sample in the target feature sample set has a real earthquake type label.

[0052] Step S302: Process each target feature sample in the target feature sample set using the target Gaussian Naive Bayes network model to determine the estimated probability that each target feature sample belongs to a natural earthquake.

[0053] Step S303: Divide the target feature sample set into a specified number of target feature sample groups according to the value of the estimated probability, and calculate the mean of the estimated probability and the actual earthquake probability for each target feature sample group to obtain multiple sets of probability calibration samples; wherein, the estimated probability corresponding to each target feature sample group is within the same value range; the actual earthquake probability represents the proportion of the actual earthquake type label in the target feature sample group that is a natural earthquake; each set of probability calibration samples includes: the mean of the estimated probability and the actual earthquake probability.

[0054] Step S304: Sort multiple groups of probability calibration samples in ascending order based on the magnitude of the estimated probability mean to obtain the sorted probability calibration sample group.

[0055] Step S305: The sorted probability calibration sample group is fitted using a monotonic regression algorithm to obtain the target calibration function that minimizes the objective function; wherein, the objective function is the sum of squared deviations of multiple estimated probability means and corresponding true earthquake probabilities.

[0056] Steps S301-S305 above constitute the method flow for determining the target calibration function. Specifically, after obtaining the target Gaussian Naive Bayes network model through training, the model is used to predict the earthquake type for each sample feature (i.e., the target feature sample) in the target feature sample set, thereby obtaining the estimated probability that the sample feature belongs to a natural earthquake.

[0057] Next, based on the condition that the estimated probabilities corresponding to each target feature sample group are within the same value range, the target feature sample set is divided into a specified number of target feature sample groups according to the value of the estimated probability. For example, the estimated probabilities are taken at equal intervals. For instance, the target feature sample set is divided into 10 target feature sample groups with an interval of 0.1. That is, the estimated probabilities (0, 0.1) are in one group, (0.1, 0.2) are in another group, and so on. According to the value of the estimated probability, a total of 10 target feature sample groups are divided.

[0058] Next, the mean of the predicted probability and the true earthquake probability are calculated for each target feature sample group, resulting in multiple probability calibration samples. If a target feature sample group contains 100 target feature samples, and 15 of these samples have the true earthquake type label as natural earthquake, then the true earthquake probability for that target feature sample group is 0.15. The mean of the predicted probability is the average of the predicted probabilities of the aforementioned 100 target feature samples.

[0059] Then, the multiple probability calibration samples are sorted in ascending order of the mean of the predicted probability. After sorting, the monotonic regression algorithm is used for fitting. The target calibration function is obtained when the objective function is minimized. The target calibration function is also the expression of the correction curve of the predicted probability.

[0060] In one alternative implementation, the objective function is expressed as: Among them, w i Let y represent the weights, i = 1, 2, ..., n, where n represents the number of probability calibration sample groups. i Let x represent the true earthquake probability in the i-th probability calibration sample. i Let represent the mean of the predicted probabilities in the i-th probability calibration sample.

[0061] To verify the performance of the method provided in this embodiment of the invention, the method provided in this embodiment of the invention is compared with the following two earthquake type identification methods: Gaussian Naive Bayes network model and a method of calibrating the output probability of Gaussian Naive Bayes network model using Pratt scaling method (hereinafter referred to as Pratt scaling calibration). For each method, the earthquake classification effect curve is plotted with reference to the following process.

[0062] ① Calculate the predicted probability values ​​of all target feature samples using earthquake type identification methods;

[0063] ② Divide the predicted probability values ​​into several groups;

[0064] ③ Calculate the mean of each group of predicted probability values, i.e., an estimated probability mean;

[0065] ④ Calculate the proportion of each positive category to obtain the true earthquake probability;

[0066] ⑤ Plot a scatter plot with the mean of each group's predicted probability on the x-axis and the actual earthquake probability of each group on the y-axis, and connect the points.

[0067] Figure 5 shows a comparison of the earthquake classification performance of the method in this embodiment and two comparative methods. The dashed line in the earthquake classification performance curve in Figure 5 is a diagonal line. The predicted value of each point on the diagonal line is equal to the true value, that is, the mean predicted probability of each group is equal to the true earthquake probability of that group. In other words, the diagonal line is the curve of completely correct classification. The closer the classification performance curve of the earthquake type identification method is to the diagonal line, the better the calibration effect.

[0068] As shown in Figure 5, the uncalibrated Gaussian Naive Bayes model exhibits large curve oscillations and poor prediction probability performance. This is because the extracted features violate the assumption of Bayesian independence distribution, leading to overconfidence in the classifier. Similarly, the Pratt scaling calibration curve also shows significant oscillations, with only minor improvement compared to the uncalibrated Gaussian Naive Bayes model. Furthermore, as indicated in the bar chart of Figure 5, its extracted features violate the assumption of Bayesian independence distribution, resulting in poor classification performance. Compared to the two methods mentioned above, the method provided in this embodiment of the invention exhibits smaller curve oscillations and approximates the diagonal line, clearly demonstrating that the method of this invention significantly improves the calibration results.

[0069] Furthermore, multiple comprehensive quantitative indicators were used to analyze the merits of the three earthquake classification methods. These indicators included Brier score, log loss, precision, recall, F1 score, ROC, and AUC. Table 1 below shows the comparison results of the comprehensive evaluation indicators for the three methods.

[0070] Table 1

[0071]

[0072] As can be seen from Table 1, the three earthquake type identification methods are similar in terms of precision, recall, and F1 score. Among the log loss and Brier score indices, the uncalibrated Gaussian Naive Bayes model has the highest score, indicating that the model's predicted probability differs the most from the actual result. The Pratt scaling model's predicted probability was improved after calibration. The method provided by the embodiment of this invention has the lowest score, indicating that this method is the best and its predicted probability matches the actual result the most closely, and the calibration effect has achieved the expected result.

[0073] Example 2

[0074] This invention also provides an earthquake type identification device, which is mainly used to execute the earthquake type identification method provided in Embodiment 1 above. The earthquake type identification device provided in this invention will be described in detail below.

[0075] Figure 6 is a functional block diagram of an earthquake type identification device provided in an embodiment of the present invention. As shown in Figure 6, the device mainly includes: an acquisition module 10, an extraction module 20, a processing module 30, and a calibration module 40, wherein:

[0076] The acquisition module 10 is used to acquire the seismic data to be identified.

[0077] Extraction module 20 is used to extract target features from seismic data; wherein, the target features are signal features that differ between natural earthquakes and non-natural earthquakes.

[0078] The processing module 30 is used to process the target features using the target Gaussian Naive Bayes network model to obtain the estimated probability that the earthquake data belongs to a natural earthquake.

[0079] The calibration module 40 is used to calibrate the estimated probability using a target calibration function to obtain the target probability that the earthquake data belongs to a natural earthquake, so as to determine the earthquake type of the earthquake data based on the target probability; wherein, the target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability.

[0080] This invention provides an earthquake type identification device. After acquiring earthquake data to be identified, it first extracts target features from the earthquake data. These target features are signal characteristics that differ between natural and non-natural earthquakes. Then, a target Gaussian Naive Bayes network model is used to process the target features to obtain an estimated probability that the earthquake data belongs to a natural earthquake. Finally, a target calibration function is used to calibrate the estimated probability to obtain the target probability that the earthquake data belongs to a natural earthquake, and the earthquake type is determined based on the target probability. Since the target calibration function is a function fitted to minimize the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability, it ensures that the earthquake type determined using this invention has high accuracy and confidence.

[0081] Optionally, the target features include at least one of the following: P / S amplitude ratio of a specified frequency band, P-wave high-low frequency energy ratio, S-wave high-low frequency energy ratio, P-wave corner frequency based on Brune source spectrum model, S-wave corner frequency based on Brune source spectrum model, P-wave corner frequency based on Jimenez source spectrum model, S-wave corner frequency based on Jimenez source spectrum model, waveform duration, P-wave waveform complexity, S-wave waveform complexity, P-wave zero-crossing rate, S-wave zero-crossing rate, P-wave cepstral complexity, S-wave cepstral complexity, P-wave instantaneous frequency complexity, and S-wave instantaneous frequency complexity.

[0082] Optionally, the earthquake type identification device is also used for:

[0083] Obtain a set of earthquake data samples; each earthquake data sample in the set has a label for the actual earthquake type; if the earthquake data sample is a natural earthquake, the label for the actual earthquake type is 0, otherwise it is 1.

[0084] The target features of each earthquake data sample in the earthquake data sample set are extracted to obtain the target feature set; each target feature in the target feature set corresponds to a real earthquake type label.

[0085] The initial Gaussian Naive Bayes network model is iteratively trained using the target feature set until the preset termination condition is met, thus obtaining the target Gaussian Naive Bayes network model.

[0086] Optionally, the earthquake type identification device is also used for:

[0087] Obtain the target feature sample set; where each target feature sample in the target feature sample set has a real earthquake type label.

[0088] The target Gaussian Naive Bayes network model is used to process each target feature sample in the target feature sample set to determine the estimated probability that each target feature sample belongs to a natural earthquake.

[0089] The target feature sample set is divided into a specified number of target feature sample groups based on the value of the estimated probability. The mean of the estimated probability and the actual earthquake probability are calculated for each target feature sample group to obtain multiple sets of probability calibration samples. The estimated probability of each target feature sample group is within the same value range. The actual earthquake probability represents the proportion of the actual earthquake type labeled as natural earthquake in the target feature sample group. Each set of probability calibration samples includes: the mean of the estimated probability and the actual earthquake probability.

[0090] The multiple groups of probability calibration samples are sorted in ascending order based on the magnitude of the estimated probability mean to obtain the sorted probability calibration sample groups.

[0091] The monotonic regression algorithm is used to fit the sorted probability calibration sample group to obtain the target calibration function that minimizes the objective function; where the objective function is the sum of squared deviations of multiple estimated probability means and corresponding true earthquake probabilities.

[0092] Alternatively, the objective function can be expressed as: Among them, w i Let y represent the weights, i = 1, 2, ..., n, where n represents the number of probability calibration sample groups. i Let x represent the true earthquake probability in the i-th probability calibration sample. i Let represent the mean of the predicted probabilities in the i-th probability calibration sample.

[0093] Example 3

[0094] Referring to Figure 7, an embodiment of the present invention provides an electronic device, which includes: a processor 60, a memory 61, a bus 62 and a communication interface 63. The processor 60, the communication interface 63 and the memory 61 are connected through the bus 62. The processor 60 is used to execute executable modules, such as computer programs, stored in the memory 61.

[0095] The memory 61 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 63 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0096] Bus 62 can be an ISA bus, PCI bus, or EISA bus, etc. Buses can be divided into address buses, data buses, control buses, etc. For ease of illustration, only a single bidirectional arrow is used in Figure 7, but this does not mean that there is only one bus or one type of bus.

[0097] The memory 61 is used to store the program. After receiving the execution instruction, the processor 60 executes the program. The method executed by the apparatus defined by the process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 60 or implemented by the processor 60.

[0098] Processor 60 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 60 or by instructions in software form. Processor 60 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can 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. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 61. Processor 60 reads the information in memory 61 and, in conjunction with its hardware, completes the steps of the above method.

[0099] The computer program product of the earthquake type identification method and apparatus provided in this embodiment of the invention includes a computer-readable storage medium storing non-volatile program code executable by a processor. The instructions included in the program code can be used to execute the methods in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0100] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0101] If the functionality is implemented as a software functional unit and sold or used as an independent product, it can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part 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 several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0102] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0103] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. In addition, the terms "first," "second," "third," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0104] Furthermore, terms such as "horizontal," "vertical," and "sag" do not imply that components must be absolutely horizontal or suspended, but rather that they can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0105] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for earthquake type identification, characterized in that, include: The process involves: acquiring earthquake data to be identified; extracting target features from the earthquake data; wherein the target features are signal features that differ between natural and non-natural earthquakes; processing the target features using a target Gaussian Naive Bayes network model to obtain an estimated probability that the earthquake data belongs to a natural earthquake; calibrating the estimated probability using a target calibration function to obtain a target probability that the earthquake data belongs to a natural earthquake, and determining the earthquake type based on the target probability; wherein the target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability.

2. The earthquake type identification method according to claim 1, characterized in that, The target features include at least one of the following: P / S amplitude ratio in a specified frequency band, P-wave high-low frequency energy ratio, S-wave high-low frequency energy ratio, P-wave corner frequency based on Brune source spectrum model, S-wave corner frequency based on Brune source spectrum model, P-wave corner frequency based on Jimenez source spectrum model, S-wave corner frequency based on Jimenez source spectrum model, waveform duration, P-wave waveform complexity, S-wave waveform complexity, P-wave zero-crossing rate, S-wave zero-crossing rate, P-wave cepstral complexity, S-wave cepstral complexity, P-wave instantaneous frequency complexity, and S-wave instantaneous frequency complexity.

3. The earthquake type identification method according to claim 1, characterized in that, Before processing the target features using the target Gaussian Naive Bayes network model, the method further includes: acquiring a set of earthquake data samples; wherein each earthquake data sample in the set has a real earthquake type label; if the earthquake data sample is a natural earthquake, the real earthquake type label is 0, otherwise it is 1; extracting the target features of each earthquake data sample in the set to obtain a target feature set; wherein each target feature in the target feature set corresponds to a real earthquake type label; and iteratively training the initial Gaussian Naive Bayes network model using the target feature set until a preset termination condition is reached to obtain the target Gaussian Naive Bayes network model.

4. The earthquake type identification method according to claim 1, characterized in that, Before calibrating the estimated probability using the target calibration function, the method further includes: acquiring a target feature sample set; wherein each target feature sample in the target feature sample set has a real earthquake type label; processing each target feature sample in the target feature sample set using the target Gaussian Naive Bayes network model to determine the estimated probability that each target feature sample belongs to a natural earthquake; dividing the target feature sample set into a specified number of target feature sample groups according to the value of the estimated probability, and calculating the mean of the estimated probability and the real earthquake probability of each target feature sample group to obtain multiple sets of probability calibration samples; wherein... The estimated probabilities corresponding to each target feature sample group are within the same value range; the actual earthquake probability represents the proportion of the actual earthquake type labeled as a natural earthquake in the target feature sample group; each probability calibration sample group includes: the mean of the estimated probability and the actual earthquake probability; the multiple probability calibration samples are sorted in ascending order based on the magnitude of the mean of the estimated probability to obtain the sorted probability calibration sample group; the sorted probability calibration sample group is fitted using a monotonic regression algorithm to obtain the target calibration function that minimizes the objective function; wherein, the objective function is the sum of squared deviations of multiple mean estimated probabilities and the corresponding actual earthquake probabilities.

5. The earthquake type identification method according to claim 4, characterized in that, The objective function is expressed as: Among them, w i Let y represent the weights, i = 1, 2, ..., n, where n represents the number of the probability calibration sample groups. i Let x represent the true earthquake probability in the i-th probability calibration sample. i Let represent the mean of the predicted probabilities in the i-th probability calibration sample.

6. An earthquake type identification device, characterized in that, include: The system comprises: an acquisition module for acquiring earthquake data to be identified; an extraction module for extracting target features from the earthquake data, wherein the target features are signal features that differ between natural and non-natural earthquakes; a processing module for processing the target features using a target Gaussian Naive Bayes network model to obtain an estimated probability that the earthquake data belongs to a natural earthquake; and a calibration module for calibrating the estimated probability using a target calibration function to obtain a target probability that the earthquake data belongs to a natural earthquake, thereby determining the earthquake type of the earthquake data based on the target probability; wherein the target calibration function is a function fitted with the objective of minimizing the sum of squared deviations between the mean of the estimated probabilities of multiple target feature sample groups and the actual earthquake probability.

7. The earthquake type identification device according to claim 6, characterized in that, The target features include at least one of the following: P / S amplitude ratio in a specified frequency band, P-wave high-low frequency energy ratio, S-wave high-low frequency energy ratio, P-wave corner frequency based on Brune source spectrum model, S-wave corner frequency based on Brune source spectrum model, P-wave corner frequency based on Jimenez source spectrum model, S-wave corner frequency based on Jimenez source spectrum model, waveform duration, P-wave waveform complexity, S-wave waveform complexity, P-wave zero-crossing rate, S-wave zero-crossing rate, P-wave cepstral complexity, S-wave cepstral complexity, P-wave instantaneous frequency complexity, and S-wave instantaneous frequency complexity.

8. The earthquake type identification device according to claim 6, characterized in that, The earthquake type identification device is further configured to: acquire an earthquake data sample set; wherein each earthquake data sample in the earthquake data sample set has a real earthquake type label; if the earthquake data sample belongs to a natural earthquake, the real earthquake type label is 0, otherwise it is 1; extract target features from each earthquake data sample in the earthquake data sample set to obtain a target feature set; wherein each target feature in the target feature set corresponds to a real earthquake type label; and use the target feature set to iteratively train an initial Gaussian Naive Bayes network model until a preset termination condition is reached to obtain the target Gaussian Naive Bayes network model.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the earthquake type identification method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the earthquake type identification method according to any one of claims 1 to 5.

Images