Optical fiber natural disaster early warning method based on machine learning and related equipment

Abstract

The invention provides an optical fiber natural disaster early warning method based on machine learning and related equipment. The method comprises the following steps: acquiring real-time optical fiber vibration data of an optical fiber pipeline deployed in a target area; performing classification processing on the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data; wherein the target classification model is obtained through training based on historical optical fiber vibration data and historical earthquake event data; and carrying out natural disaster early warning based on the earthquake information. According to the optical fiber natural disaster early warning method based on machine learning and the related equipment, the earthquake information existing in the optical fiber vibration data can be identified, then natural disaster early warning is carried out based on the earthquake information, an efficient and accurate earthquake monitoring and early warning system is established, an earthquake event can be early warned in time, and the early warning efficiency is improved. And loss caused by earthquake disasters is reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of optical fiber early warning technology, and in particular to an optical fiber natural disaster early warning method based on machine learning and related equipment. Background Art

[0002] Optical fiber communication has become the core infrastructure for the Internet, mobile communications, data centers, and television broadcasting due to its high bandwidth and long-distance transmission capabilities. However, the fragility of optical fiber networks is also obvious. Optical fibers are easily damaged by external physical factors (such as construction and natural disasters). Therefore, it is necessary to monitor and predict the operation status of optical fiber networks to detect possible fault hazards in advance.

[0003] However, at present, when it comes to natural disaster warning assisted by optical fiber, there are still problems such as one-sided predictions, difficulty in unifying time and space, and strong small regionality. Summary of the invention

[0004] In view of this, the purpose of the present disclosure is to propose a fiber optic natural disaster warning method and related equipment based on machine learning.

[0005] Based on the above objectives, the first aspect of the present disclosure provides an optical fiber natural disaster early warning method based on machine learning, comprising:

[0006] Obtain real-time fiber vibration data of fiber optic pipelines deployed in target areas;

[0007] Classify and process the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data; wherein the target classification model is trained based on historical optical fiber vibration data and historical seismic event data;

[0008] A natural disaster warning is carried out based on the earthquake information.

[0009] In some embodiments, the real-time optical fiber vibration data includes time series data formed by at least one of vibration time, vibration position, vibration phase, and vibration intensity.

[0010] In some embodiments, the classifying and processing the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data includes:

[0011] Performing feature extraction on the real-time optical fiber vibration data to obtain first features in the time domain, the frequency domain and the time-frequency domain;

[0012] Dividing the real-time optical fiber vibration data into a plurality of time windows of preset lengths, and determining a first feature vector of the real-time optical fiber vibration data in each time window based on the first feature;

[0013] The first feature vector is input into the target classification model to obtain the earthquake information.

[0014] In some embodiments, the first feature of the time domain includes at least one of the mean, standard deviation, maximum value, minimum value, kurtosis, skewness, root mean square, and zero crossing rate; the first feature of the frequency domain includes at least one of the spectral energy, main frequency component, frequency center, spectral entropy, and bandwidth; the first feature of the time-frequency domain includes at least one of the short-time Fourier transform feature and the wavelet transform feature.

[0015] In some embodiments, inputting the first feature vector into the target classification model to obtain the earthquake information includes:

[0016] Determine vibration signal characteristics at different positions based on the first characteristics, and calculate epicenter position information based on the vibration signal characteristics at different positions;

[0017] determining the amplitude, frequency component and energy distribution of the vibration signal based on the first feature, and calculating earthquake intensity information based on the amplitude, frequency component and energy distribution of the vibration signal;

[0018] The earthquake information is determined based on at least one of the epicenter location information and the earthquake intensity information.

[0019] In some embodiments, the training process of the target classification model includes:

[0020] Acquiring the historical optical fiber vibration data and the historical earthquake event data;

[0021] Using the historical earthquake event data to mark the historical optical fiber vibration data, and determining earthquake events in the historical optical fiber vibration data;

[0022] A training set and a test set are determined based on the annotated historical optical fiber vibration data, and model training is performed based on the training set and the test set to obtain the target classification model.

[0023] In some embodiments, the using the historical seismic event data to mark the historical optical fiber vibration data to determine the seismic events in the historical optical fiber vibration data includes:

[0024] Determining the position of a time window in the historical optical fiber vibration data according to the time information of the seismic information in the historical seismic event data;

[0025] Determine the marked time range according to the seismic wave propagation speed of the seismic information in the historical seismic event data and the distance from the monitoring area;

[0026] Seismic events are annotated in the historical optical fiber vibration data based on the position of the time window and the annotated time range.

[0027] In some embodiments, it also includes:

[0028] Obtain the mean, standard deviation, maximum value, minimum value, kurtosis, skewness, root mean square, zero crossing rate, spectrum energy, main frequency component, frequency center, spectrum entropy, and bandwidth through the time domain characteristics, frequency domain characteristics, and time-frequency domain characteristics of the real-time optical fiber vibration data;

[0029] The natural disaster early warning based on the earthquake information comprises:

[0030] In response to the possibility of an earthquake and / or the earthquake intensity being greater than a preset threshold, generating earthquake warning information;

[0031] Wherein, obtaining the preset threshold value includes: detecting conventional vibration information of the optical fiber pipeline other than seismic information, and determining the preset threshold value based on the intensity characteristics and time characteristics of the conventional vibration information.

[0032] A second aspect of the present disclosure provides an optical fiber natural disaster early warning device based on machine learning, comprising:

[0033] An acquisition module is configured to: acquire real-time optical fiber vibration data of an optical fiber pipeline deployed in a target area;

[0034] A classification module is configured to: classify the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data; wherein the target classification model is trained based on historical optical fiber vibration data and historical seismic event data;

[0035] The early warning module is configured to: provide natural disaster early warning based on the earthquake information.

[0036] A third aspect of the present disclosure provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the optical fiber natural disaster warning method based on machine learning as described in the first aspect is implemented.

[0037] From the above, it can be seen that the machine learning-based optical fiber natural disaster warning method and related equipment provided by the present disclosure obtain a target classification model based on historical optical fiber vibration data and historical earthquake event data training, and use the trained target classification model to classify and process the real-time optical fiber vibration data of the optical fiber pipeline deployed in the target area, thereby identifying the earthquake information present in the real-time optical fiber vibration data, and then performing natural disaster warning based on the earthquake information, thereby establishing an efficient and accurate earthquake monitoring and warning system that can timely warn of earthquake events and reduce the losses caused by earthquake disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 A flowchart of an exemplary method provided by an embodiment of the present disclosure is shown.

[0040] Figure 2 A schematic diagram showing an exemplary scenario according to an embodiment of the present disclosure is shown.

[0041] Figure 3 A schematic diagram of an exemplary device provided by an embodiment of the present disclosure is shown.

[0042] Figure 4 A schematic diagram of the hardware structure of an exemplary computer device provided in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0043] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0044] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0045] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0046] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0047] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0048] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0049] The present disclosure provides a method for early warning of natural disasters in optical fiber based on machine learning. Figure 1 As shown, the method includes:

[0050] Step S101, acquiring real-time optical fiber vibration data of an optical fiber pipeline deployed in a target area.

[0051] Among them, the optical fiber pipeline is the optical cable, which is the infrastructure of the optical fiber communication network, including the same-groove optical cable and the same-cable optical fiber. The optical cable is mainly composed of optical fiber, plastic protective sleeve and plastic sheath. It is composed of a certain number of optical fibers, such as 24, 48, 72 and other fiber cores, which are arranged in a certain way to form a cable core, covered with a sheath, and some are also covered with an outer sheath to realize a communication line for optical signal transmission. That is: a cable formed by optical fiber (optical transmission carrier) through a certain process; optical fiber refers to the physical optical fiber entity for data transmission, which is used to bear the network transmission medium and provide a transparent transmission channel.

[0052] like Figure 2 As shown, in this embodiment, optical fiber sensors (such as distributed acoustic sensing systems DAS) can be used to stably and continuously collect real-time optical fiber vibration data of part or even all optical fiber pipelines deployed in the target area, so as to use the real-time optical fiber vibration data of the entire area to detect and warn the occurrence of earthquakes in the target area. Among them, the target area can be an earthquake-prone area. By deploying optical fiber pipelines in earthquake-prone areas, the collected real-time optical fiber vibration data can include vibration information of each point of the optical fiber when an earthquake occurs.

[0053] Distributed fiber optic sensor (DAS) is a device used to measure or monitor the spatial distribution and time-varying information along the optical fiber transmission path. It uses optical fiber as a sensing sensitive element and a transmission signal medium at the same time, and adopts advanced OTDR technology and OFDR technology to detect the changes in temperature and strain at different positions along the optical fiber, realizing truly distributed measurement. Among them, the optical time domain reflectometer (OTDR) injects a high-power laser or light pulse from one end of the optical cable and receives the reflected signal through the same side. When the light pulse is transmitted through the optical cable, part of the scattering and reflection will return to the transmitting end. Optical frequency domain reflectometer (OFDR) is a special technology used to analyze the optical light path and reflection characteristics in optical fibers and optical components. Unlike the optical time domain reflectometer (OTDR), OFDR uses a continuous wave (CW) light source to work in the frequency domain, which has a higher signal-to-noise ratio.

[0054] Optical fiber vibration data is the optical fiber vibration data generated by the reflection of the optical fiber by the deformation and temperature of the optical fiber pipeline. Real-time optical fiber vibration data is the data obtained in real time for current earthquake detection and early warning.

[0055] Step S103, classifying and processing the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data; wherein the target classification model is trained based on historical optical fiber vibration data and historical seismic event data.

[0056] In this embodiment, the real-time optical fiber vibration data can be classified and processed based on a pre-trained target classification model to determine the seismic information in the real-time optical fiber vibration data, wherein the target classification model is trained based on historical optical fiber vibration data and historical earthquake event data.

[0057] Step S105: issuing a natural disaster warning based on the earthquake information.

[0058] In this embodiment, a target classification model is obtained by training based on historical optical fiber vibration data and historical earthquake event data, and the trained target classification model is used to classify and process the real-time optical fiber vibration data of the optical fiber pipeline deployed in the target area, so as to identify the earthquake information present in the real-time optical fiber vibration data, and then natural disaster warning is performed based on the earthquake information, thereby establishing an efficient and accurate earthquake monitoring and early warning system, which can timely warn of earthquake events and reduce the losses caused by earthquake disasters.

[0059] In this embodiment, real-time optical fiber vibration data of optical fiber pipelines deployed in the target area can be collected and acquired, and the real-time optical fiber vibration data can be classified and processed using a pre-trained target classification model, so as to identify earthquake information hidden in the real-time optical fiber vibration data, and perform natural disaster warning based on the earthquake information, thereby establishing an efficient and accurate earthquake monitoring and early warning system that can timely warn of earthquake events and reduce the losses caused by earthquake disasters.

[0060] In some embodiments, the real-time optical fiber vibration data may include time series data formed by at least one of vibration time, vibration position, vibration phase, and vibration intensity. In this embodiment, since the real-time optical fiber vibration data is data obtained by stably and continuously collecting the vibration of part or all optical fiber pipelines deployed in the target area, the data is a time series data.

[0061] In some embodiments, the step S103 of classifying the real-time optical fiber vibration data based on the pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data includes:

[0062] Step S201, extracting features from the real-time optical fiber vibration data to obtain first features in the time domain, frequency domain and time-frequency domain.

[0063] In some embodiments, before extracting features from the real-time optical fiber vibration data in step S201, the process further includes: preprocessing the real-time optical fiber vibration data.

[0064] In this embodiment, the real-time optical fiber vibration data is preprocessed, including: performing operations such as denoising, normalization, and signal segmentation on the collected real-time optical fiber vibration data to facilitate subsequent feature extraction and machine learning model training.

[0065] In this embodiment, feature extraction may be performed on the time series data of the real-time optical fiber vibration data to extract the first features of the real-time optical fiber vibration data in the time domain, frequency domain and time-frequency domain.

[0066] In this embodiment, the first feature in the time domain includes at least one of a mean value, a standard deviation, a maximum value, a minimum value, a kurtosis, a skewness, a root mean square, and a zero crossing rate.

[0067] Among them, the mean can be used to reflect the average amplitude of the signal;

[0068] Standard deviation, which can be used to measure the dispersion of signal amplitude;

[0069] Maximum value, which can be used for the maximum value of the signal amplitude;

[0070] Minimum value, which can be used for the minimum value of the signal amplitude;

[0071] Kurtosis, which can be used to describe the peak degree of the signal amplitude distribution;

[0072] Skewness, which can be used to describe the symmetry of the signal amplitude distribution;

[0073] Root mean square (RMS), which can be used for the root mean square value of the signal amplitude, reflecting the signal energy;

[0074] The zero-crossing rate can be used to measure the number of times a signal crosses the zero point, reflecting the frequency component of the signal.

[0075] In this embodiment, the first feature in the frequency domain includes at least one of spectrum energy, main frequency component, frequency center, spectrum entropy, and bandwidth.

[0076] Among them, spectrum energy can be used to represent the energy distribution of the signal at different frequencies;

[0077] The main frequency component can be used to represent the frequency component with the highest energy in the signal;

[0078] Frequency center, which can be used to represent the weighted average frequency of the spectrum;

[0079] Spectral entropy, which can be used to represent the randomness and complexity of the spectrum;

[0080] Bandwidth can be used to express the width of a frequency distribution.

[0081] In this embodiment, the first feature in the time-frequency domain includes at least one of a short-time Fourier transform feature and a wavelet transform feature.

[0082] Among them, the short-time Fourier transform (STFT) feature can be used to represent the frequency components in different time windows;

[0083] Wavelet transform features analyze the local characteristics of signals at different frequencies and times through wavelet functions of different scales.

[0084] In some embodiments, after obtaining the first features of the real-time optical fiber vibration data in the time domain, the frequency domain, and the time-frequency domain, feature selection and dimension reduction may be performed on the first features.

[0085] In this embodiment, methods such as PCA (Principal Component Analysis) or LDA (Linear Discriminant Analysis) can be used to perform dimensionality reduction processing on the extracted time domain and frequency domain features to reduce the feature dimension and improve the computational efficiency of the model.

[0086] Step S203: divide the real-time optical fiber vibration data into a plurality of time windows of preset lengths, and determine a first feature vector of the real-time optical fiber vibration data in each time window based on the first feature.

[0087] Step S205: input the first feature vector into the target classification model to obtain the earthquake information.

[0088] In this embodiment, the first feature vector may be input into the target classification model, and the vibration signal collected in real time may be analyzed using the trained target classification model to obtain earthquake information.

[0089] In some embodiments, the step S205 of inputting the first feature vector into the target classification model to obtain the earthquake information includes:

[0090] Step S301, determining the vibration signal characteristics of different positions based on the first characteristics, and calculating the epicenter position information based on the vibration signal characteristics of different positions.

[0091] In this embodiment, the first feature can be used to determine the vibration signal characteristics at different positions, and then the epicenter position information can be calculated based on the vibration signal characteristics at different positions.

[0092] Among them, the time difference of arrival method (TDOA) can be used to calculate the propagation path of the seismic wave and the epicenter position based on the difference in arrival time of the vibration signal at each point in the real-time optical fiber vibration data; or, through the waveform matching method, the characteristic information of the real-time optical fiber vibration data is matched with the pre-stored seismic waveform template to identify the epicenter position and earthquake intensity.

[0093] Step S303, determining the amplitude, frequency component and energy distribution of the vibration signal based on the first feature, and calculating the earthquake intensity information based on the amplitude, frequency component and energy distribution of the vibration signal.

[0094] In this embodiment, the first feature may be used to determine the amplitude, frequency component, and energy distribution of the vibration signal, and combined with an empirical formula or regression model to estimate the intensity (eg, magnitude) of the earthquake.

[0095] Step S305, determining the earthquake information based on at least one of the epicenter location information and the earthquake intensity information.

[0096] Finally, the earthquake information is determined according to at least one of the epicenter location information and the earthquake intensity information, so as to provide natural disaster warning based on the earthquake information.

[0097] In some embodiments, the training process of the target classification model includes:

[0098] Step S401, acquiring the historical optical fiber vibration data and the historical earthquake event data.

[0099] In this embodiment, the optical fiber vibration data can be stably and continuously collected by a high-performance DAS data acquisition device. In the process of collecting optical fiber vibration data, an appropriate sampling frequency is selected according to the characteristics of the seismic wave. Since the frequency of seismic waves is usually low, a sampling frequency between 100 Hz and 1 kHz can be selected to ensure that the main characteristics of the seismic wave can be captured. At the same time, sufficient storage space is configured to save the continuously collected optical fiber vibration data. A distributed storage system can be used to facilitate data management and access.

[0100] In this embodiment, historical earthquake event data can be obtained from national earthquake monitoring agencies, regional seismic networks or international earthquake monitoring organizations. The historical earthquake event data may include detailed information such as the time, location, magnitude, focal depth, and epicenter location of the earthquake. After obtaining the historical earthquake event data, the earthquake event data is uniformly converted into a CSV standard format for subsequent processing and analysis.

[0101] Step S403: using the historical earthquake event data to mark the historical optical fiber vibration data, and determining earthquake events in the historical optical fiber vibration data.

[0102] In this embodiment, before the historical earthquake event data is used to mark the historical optical fiber vibration data in step S403, the method further includes:

[0103] The historical optical fiber vibration data is preprocessed, that is, denoising, normalization, signal segmentation and other operations are performed on the historical optical fiber vibration data to facilitate subsequent feature extraction and machine learning model training.

[0104] The pre-processed historical optical fiber vibration data is subjected to feature extraction: different features of the time series data (vibration time, vibration position, vibration phase, vibration intensity, etc.) are extracted according to the features related to the seismic signal in the time domain, frequency domain and time-frequency domain: mean, standard deviation, maximum value, minimum value, kurtosis, skewness, root mean square (RMS), zero-crossing rate, spectrum energy, main frequency component, frequency center, spectrum entropy, bandwidth, short-time Fourier transform (STFT), wavelet transform, etc., to form the second feature.

[0105] The second eigenvector is normalized to ensure that each feature has a balanced impact on the model.

[0106] The time series data of historical optical fiber vibration data is divided into multiple fixed-length time windows, and the data in each window is processed as a sample. The length and step size of the sliding window need to be adjusted according to the specific application scenario.

[0107] For each time window, the second eigenvector is calculated, wherein the second eigenvector may include statistical features (such as mean, standard deviation, etc.) and frequency domain features within the window.

[0108] Afterwards, each time window of the historical optical fiber vibration data is labeled and its corresponding category (such as natural disaster type, intensity status, etc.) is marked.

[0109] In this embodiment, each time series data corresponding to the historical optical fiber vibration data is labeled based on the historical earthquake event data, so as to indicate whether an earthquake has occurred and its intensity and other information by using clear labels.

[0110] In some embodiments, before the historical optical fiber vibration data is annotated with the historical earthquake event data in step S403 to determine the earthquake events in the historical optical fiber vibration data, the method further includes: preprocessing the historical optical fiber vibration data and the historical earthquake event data. Preprocessing the historical optical fiber vibration data and the historical earthquake event data includes:

[0111] Step S501, aligning the timestamps of historical optical fiber vibration data and historical earthquake event data.

[0112] In this embodiment, the timestamps of the optical fiber vibration data and the seismic event data are ensured to be consistent. When converted into a standard format, the timestamp range can be customized to ensure that the data range of multiple optical fibers in the same area is within the specified range.

[0113] Step S503, aligning the historical optical fiber vibration data with the historical earthquake event data.

[0114] In this embodiment, the optical fiber vibration data is aligned according to the time information of the seismic event data to ensure that the vibration signal segment corresponding to each seismic event is complete and accurate.

[0115] Step S505: select a filter.

[0116] In this embodiment, an appropriate filter type is selected according to the characteristics of the optical fiber vibration signal. In the present invention, low-pass filtering is used to remove high-frequency noise and retain low-frequency signals.

[0117] Step S507, setting filtering parameters.

[0118] In this embodiment, the cutoff frequency and bandwidth of the filter are set according to the frequency distribution of the signal. The optimal parameters can be determined by spectrum analysis.

[0119] Among them, earthquakes can be divided into artificial earthquakes and natural earthquakes: the seismic waves excited by artificial earthquakes generally have a frequency range of 2-90Hz. In order to simulate the real situation, in the numerical simulation of seismic waves, the main frequency of the source sub-wave is generally in the range of 6-50Hz; the frequency of natural earthquake seismic waves is generally within 15Hz.

[0120] Step S509, net data processing.

[0121] Since the vibration intensity of earthquakes is generally higher than that of events such as vehicles and construction, it is necessary to set a higher warning value for vibration intensity. The screening of warning values ​​can be determined by the values ​​detected in the current network. One feasible method is to detect the warning values ​​of conventional vibration intensity for different optical fibers in the same channel. Conventional vibrations can be artificially created or selected near the construction site with optical cables, and recorded in the machine room using DAS. Subsequently, the reflected light signal waveform of the relevant location points is analyzed, and its intensity characteristics and time type are analyzed, and entered into the database. Finally, based on the real-time vibration data, a higher intensity value is selected to filter out false alarm reports.

[0122] Step S511, extracting time domain features, frequency domain features, and time-frequency domain features, that is, extracting features from the time series signal of historical optical fiber vibration data.

[0123] Among them, the time domain features include: peak, the maximum value of the signal; mean, the average value of the signal; variance, the degree of discreteness of the signal.

[0124] Frequency domain features: The time series signal is converted to the frequency domain through Fourier transform, and frequency domain features are extracted, such as power spectrum density and the distribution of signal energy at each frequency.

[0125] Time-frequency domain features: Through methods such as wavelet transform (WT) or short-time Fourier transform (STFT), the signal is decomposed into different time and frequency scales, and features such as wavelet coefficients are extracted.

[0126] In some embodiments, the step S203 of marking the historical optical fiber vibration data using the historical earthquake event data to determine the earthquake events in the historical optical fiber vibration data includes:

[0127] Step S601, determining the position of a time window in the historical optical fiber vibration data according to the time information of the seismic information in the historical seismic event data.

[0128] In this embodiment, a corresponding time window is determined in the optical fiber vibration data according to the time information of the earthquake event.

[0129] Among them, an appropriate time window length is selected, such as 1 second, 10 seconds or 1 minute, to ensure that each window contains sufficient information for subsequent feature extraction and analysis.

[0130] In some embodiments, a sliding window method is used to improve sample coverage and sample size. The sliding step size can be set to 10% to 50% of the window length. Alternatively, the window length can be specified. Since the earthquake vibration range is wide and the optical fiber length is limited, a window length of 10 kilometers can be selected for detection. When selecting the window length, it also has a certain significance. Since conventional vibrations are generally small in the vibration range, conventional vibrations can be cleaned out when processing data.

[0131] Step S603, determining a marked time range according to the seismic wave propagation velocity of the seismic information in the historical seismic event data and the distance from the monitoring area.

[0132] In this embodiment, the vibration signal within a certain time range before and after the earthquake event can be marked as an earthquake event, and the vibration signal in other time periods can be marked as a non-earthquake event. The marked time range can be determined according to the propagation speed of the earthquake wave and the distance of the monitoring area.

[0133] Step S605: marking earthquake events in the historical optical fiber vibration data based on the position of the time window and the marked time range.

[0134] In this embodiment, the label of each time window is determined according to the event labeling result. If the window contains an earthquake event, it is labeled as an earthquake event; otherwise, it is labeled as a non-earthquake event.

[0135] Step S405, determining a training set and a test set based on the annotated historical optical fiber vibration data, performing model training based on the training set and the test set, and obtaining the target classification model.

[0136] In this embodiment, feature statistics and analysis are performed on the historical optical fiber vibration data of the marked earthquake events, and the distribution and discrimination ability of each feature are evaluated through descriptive statistical analysis, such as mean, standard deviation, correlation coefficient, etc.; then, recursive feature elimination (RFE), L1 regularization and other algorithms are used to evaluate the importance of features and screen out the features that are most helpful for earthquake detection.

[0137] In this embodiment, a statistical distribution analysis can be performed on the historical optical fiber vibration data of the marked earthquake events, and the distribution of features in earthquake events and non-earthquake events can be analyzed using visualization tools such as histograms and box plots; a correlation analysis can also be performed on the historical optical fiber vibration data of the marked earthquake events, and the correlation coefficient between each feature and the earthquake event label can be calculated to select features with higher correlation; a feature importance analysis can also be performed on the historical optical fiber vibration data of the marked earthquake events, and a machine learning model (such as random forest) can be used to calculate the importance score of the feature.

[0138] Afterwards, the training set and the test set are divided to perform model training based on the training set and the test set to obtain the target classification model. In this embodiment, the extracted second feature vector and the corresponding label can be used as input to form a training data set. The structure of the input data is a two-dimensional array, in which each row represents a feature vector of a time window and each column represents a feature.

[0139] In some embodiments, a training set, a validation set and a test set may be determined based on the annotated historical optical fiber vibration data, model training may be performed based on the training set, model tuning may be performed based on the validation set, and model evaluation may be performed based on the test set.

[0140] In some embodiments, the target classification model may adopt an SVM model, and the training of the SVM model is achieved by optimizing the objective function to maximize the boundary interval and minimize the misclassification. The model performance is evaluated on the test set, and the main indicators include accuracy, recall rate, F1 score, and area under the ROC curve (AUC).

[0141] In this embodiment, since the collected data has no linear relationship, the SVM kernel function, radial basis function kernel (RBF), is selected according to the data characteristics, which is very suitable for processing nonlinear data and has good classification performance.

[0142] Among them, RBF (Radial Basis Function) is a scalar function that is radially symmetric, usually defined as a monotonic function of the Euclidean distance between a point in space and a certain center. The most commonly used radial basis function is the Gaussian function, which is in the form of:

[0143]

[0144] Among them, is the center vector of the σ function, C j is the width vector. The scope of the Gaussian function shows locality, that is, the function value is smaller when it is far away.

[0145] In some embodiments, the target classification model may adopt an SVM model, and the training of the SVM model is achieved by optimizing the objective function to maximize the boundary interval and minimize the misclassification. The model performance is evaluated on the test set, and the main indicators include accuracy, recall rate, F1 score, and area under the ROC curve (AUC).

[0146] Before training begins, the initial SVM parameters are set, such as the penalty parameter C and the kernel function parameter γ. The initial parameters can be determined through empirical selection or preliminary tuning. During the training process, the SVM model is trained using the training set data, and the model parameters are optimized by minimizing the training error and maximizing the classification interval. During the training process, the performance of the model on the validation set is continuously evaluated to ensure the generalization ability of the model.

[0147] In the process of model tuning, the k-fold cross-validation method is used to divide the training set into k subsets. K-1 subsets are used to train the model each time, and the remaining subset is used to verify the model. The cycle is repeated k times to evaluate the model performance, thereby realizing parameter tuning cross-validation.

[0148] Use the validation set to evaluate model performance and calculate indicators such as accuracy, recall, F1-score, ROC curve, and AUC value to evaluate model performance indicators; based on the evaluation results, further adjust model parameters or feature selection to optimize model performance.

[0149] In some embodiments, the method further includes: obtaining specific characteristics such as mean, standard deviation, maximum value, minimum value, kurtosis, skewness, root mean square, zero crossing rate, spectral energy, main frequency component, frequency center, spectral entropy, bandwidth, etc. through the time domain characteristics, frequency domain characteristics, and time-frequency domain characteristics of the real-time optical fiber vibration data.

[0150] In some embodiments, in response to the possibility of an earthquake and / or the earthquake intensity being greater than a preset threshold, earthquake warning information is generated. Wherein, obtaining the preset threshold includes: detecting conventional vibration information other than earthquake information of the optical fiber pipeline, and determining the preset threshold based on the intensity characteristics and time characteristics of the conventional vibration information. In this embodiment, warnings can be issued through various warning information such as alarm sounds, SMS notifications, and email notifications. The detected earthquake location and intensity information are transmitted to the early warning system, and the early warning information is issued in a timely manner, and corresponding emergency measures are taken.

[0151] In some embodiments, the model can be updated regularly using newly collected fiber optic vibration data and seismic event data, and the accuracy and reliability of the model can be maintained through retraining and optimization; and a continuous learning method can be used to enable the model to continuously adapt to new data and environmental changes.

[0152] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0153] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0154] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present disclosure also provides an optical fiber natural disaster warning device based on machine learning.

[0155] refer to Figure 3 , the device comprises:

[0156] The acquisition module 11 is configured to: acquire real-time optical fiber vibration data of an optical fiber pipeline deployed in a target area;

[0157] The classification module 13 is configured to: classify the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data; wherein the target classification model is trained based on historical optical fiber vibration data and historical seismic event data;

[0158] The early warning module 15 is configured to: provide a natural disaster early warning based on the earthquake information.

[0159] In some embodiments, the real-time optical fiber vibration data includes time series data formed by at least one of vibration time, vibration position, vibration phase, and vibration intensity.

[0160] In some embodiments, the classification module 13 is further configured to: perform feature extraction on the real-time optical fiber vibration data to obtain first features in the time domain, the frequency domain and the time-frequency domain;

[0161] Dividing the real-time optical fiber vibration data into a plurality of time windows of preset lengths, and determining a first feature vector of the real-time optical fiber vibration data in each time window based on the first feature;

[0162] The first feature vector is input into the target classification model to obtain the earthquake information.

[0163] In some embodiments, the first feature of the time domain includes at least one of the mean, standard deviation, maximum value, minimum value, kurtosis, skewness, root mean square, and zero crossing rate; the first feature of the frequency domain includes at least one of the spectral energy, main frequency component, frequency center, spectral entropy, and bandwidth; the first feature of the time-frequency domain includes at least one of the short-time Fourier transform feature and the wavelet transform feature.

[0164] In some embodiments, inputting the first feature vector into the target classification model to obtain the earthquake information includes:

[0165] Determine vibration signal characteristics at different positions based on the first characteristics, and calculate epicenter position information based on the vibration signal characteristics at different positions;

[0166] determining the amplitude, frequency component and energy distribution of the vibration signal based on the first feature, and calculating earthquake intensity information based on the amplitude, frequency component and energy distribution of the vibration signal;

[0167] The earthquake information is determined based on at least one of the epicenter location information and the earthquake intensity information.

[0168] In some embodiments, the training process of the target classification model includes:

[0169] Acquiring the historical optical fiber vibration data and the historical earthquake event data;

[0170] Using the historical earthquake event data to mark the historical optical fiber vibration data, and determining earthquake events in the historical optical fiber vibration data;

[0171] A training set and a test set are determined based on the annotated historical optical fiber vibration data, and model training is performed based on the training set and the test set to obtain the target classification model.

[0172] In some embodiments, the using the historical seismic event data to mark the historical optical fiber vibration data to determine the seismic events in the historical optical fiber vibration data includes:

[0173] Determining the position of a time window in the historical optical fiber vibration data according to the time information of the seismic information in the historical seismic event data;

[0174] Determine the marked time range according to the seismic wave propagation speed of the seismic information in the historical seismic event data and the distance from the monitoring area;

[0175] Seismic events are annotated in the historical optical fiber vibration data based on the position of the time window and the annotated time range.

[0176] In some embodiments, the method further includes: obtaining a mean value, a standard deviation, a maximum value, a minimum value, a kurtosis, a skewness, a root mean square, a zero crossing rate, a spectrum energy, a main frequency component, a frequency center, a spectrum entropy, and a bandwidth through the time domain characteristics, the frequency domain characteristics, and the time-frequency domain characteristics of the real-time optical fiber vibration data;

[0177] The natural disaster early warning based on the earthquake information comprises:

[0178] In response to the possibility of an earthquake and / or the earthquake intensity being greater than a preset threshold, generating earthquake warning information;

[0179] In some embodiments, obtaining the preset threshold includes: detecting conventional vibration information of the optical fiber pipeline other than seismic information, and determining the preset threshold based on intensity characteristics and time characteristics of the conventional vibration information.

[0180] For the convenience of description, the above device is described by dividing it into various modules according to its functions. Of course, when implementing the present disclosure, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0181] The device of the above embodiment is used to implement the corresponding method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0182] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any of the above embodiments is implemented.

[0183] Figure 4 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0184] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0185] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0186] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0187] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0188] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0189] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0190] The electronic device of the above embodiment is used to implement the corresponding method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0191] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method described in any of the above embodiments.

[0192] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0193] The computer instructions stored in the storage medium of the above embodiments are used to enable the computer to execute the method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0194] Those skilled in the art will appreciate that the embodiments of the present disclosure may be implemented as a system, method, or computer program product. Therefore, the present disclosure may be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." In addition, in some embodiments, the present disclosure may also be implemented in the form of a computer program product in one or more computer-readable media, which contains computer-readable program code.

[0195] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive examples) of computer-readable storage media may include, for example: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0196] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0197] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0198] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0199] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified in the boxes in the flowchart and / or block diagram.

[0200] These computer program instructions may also be stored in a computer-readable medium that enables a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable medium produce a product that includes an instruction device that implements the functions / operations specified in the blocks in the flowchart and / or block diagram.

[0201] Computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable device to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.

[0202] In addition, although the operations of the disclosed method are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted in the flow chart can be performed in a different order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

[0203] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0204] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0205] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Based on the concept of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of simplicity.

[0206] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present disclosure difficult to understand, the known power / ground connections to the integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present disclosure difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present disclosure will be implemented (that is, these details should be fully within the scope of understanding of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present disclosure, it is apparent to those skilled in the art that the embodiments of the present disclosure can be implemented without these specific details or with changes in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0207] Although the present disclosure has been described in conjunction with specific embodiments of the present disclosure, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0208] The embodiments of the present disclosure are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A fiber optic natural disaster early warning method based on machine learning, characterized in that: include: Obtain real-time fiber vibration data of fiber optic pipelines deployed in target areas; Classify and process the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data; wherein the target classification model is trained based on historical optical fiber vibration data and historical seismic event data; A natural disaster warning is carried out based on the earthquake information.

2. The method according to claim 1, characterized in that The real-time optical fiber vibration data includes time series data formed by at least one of vibration time, vibration position, vibration phase and vibration intensity.

3. The method according to claim 1, characterized in that: The classifying and processing the real-time optical fiber vibration data based on the pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data includes: Performing feature extraction on the real-time optical fiber vibration data to obtain first features in the time domain, the frequency domain and the time-frequency domain; Dividing the real-time optical fiber vibration data into a plurality of time windows of preset lengths, and determining a first feature vector of the real-time optical fiber vibration data in each time window based on the first feature; The first feature vector is input into the target classification model to obtain the earthquake information.

4. The method according to claim 3, characterized in that The first feature of the time domain includes at least one of the mean, standard deviation, maximum value, minimum value, kurtosis, skewness, root mean square, and zero crossing rate; the first feature of the frequency domain includes at least one of the spectral energy, main frequency component, frequency center, spectral entropy, and bandwidth; the first feature of the time-frequency domain includes at least one of the short-time Fourier transform feature and the wavelet transform feature.

5. The method according to claim 3, characterized in that: The step of inputting the first feature vector into the target classification model to obtain the earthquake information includes: Determine vibration signal characteristics at different positions based on the first characteristics, and calculate epicenter position information based on the vibration signal characteristics at different positions; determining the amplitude, frequency component and energy distribution of the vibration signal based on the first feature, and calculating earthquake intensity information based on the amplitude, frequency component and energy distribution of the vibration signal; The earthquake information is determined based on at least one of the epicenter location information and the earthquake intensity information.

6. The method according to claim 1, characterized in that The training process of the target classification model includes: Acquiring the historical optical fiber vibration data and the historical earthquake event data; Using the historical earthquake event data to mark the historical optical fiber vibration data, and determining earthquake events in the historical optical fiber vibration data; A training set and a test set are determined based on the annotated historical optical fiber vibration data, and model training is performed based on the training set and the test set to obtain the target classification model.

7. The method according to claim 6, characterized in that The method of using the historical earthquake event data to mark the historical optical fiber vibration data to determine the earthquake events in the historical optical fiber vibration data includes: Determining the position of a time window in the historical optical fiber vibration data according to the time information of the seismic information in the historical seismic event data; Determine the marked time range according to the seismic wave propagation speed of the seismic information in the historical seismic event data and the distance from the monitoring area; Seismic events are annotated in the historical optical fiber vibration data based on the position of the time window and the annotated time range.

8. The method according to claim 1, characterized in that Also includes: Obtain the mean, standard deviation, maximum value, minimum value, kurtosis, skewness, root mean square, zero crossing rate, spectrum energy, main frequency component, frequency center, spectrum entropy, and bandwidth through the time domain characteristics, frequency domain characteristics, and time-frequency domain characteristics of the real-time optical fiber vibration data; The natural disaster early warning based on the earthquake information comprises: In response to the possibility of an earthquake and / or the earthquake intensity being greater than a preset threshold, generating earthquake warning information; Wherein, obtaining the preset threshold value includes: detecting conventional vibration information of the optical fiber pipeline other than seismic information, and determining the preset threshold value based on the intensity characteristics and time characteristics of the conventional vibration information.

9. An optical fiber natural disaster early warning device based on machine learning, characterized in that: include: An acquisition module is configured to: acquire real-time optical fiber vibration data of an optical fiber pipeline deployed in a target area; A classification module is configured to: classify the real-time optical fiber vibration data based on a pre-trained target classification model to obtain seismic information in the real-time optical fiber vibration data; wherein the target classification model is trained based on historical optical fiber vibration data and historical seismic event data; The early warning module is configured to: provide natural disaster early warning based on the earthquake information.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the optical fiber natural disaster early warning method based on machine learning as described in any one of claims 1 to 8 is implemented.

Images