CN118151252A - Satellite observation-based seismic magnetic disturbance analysis method - Google Patents
Satellite observation-based seismic magnetic disturbance analysis method Download PDFInfo
- Publication number
- CN118151252A CN118151252A CN202410275650.0A CN202410275650A CN118151252A CN 118151252 A CN118151252 A CN 118151252A CN 202410275650 A CN202410275650 A CN 202410275650A CN 118151252 A CN118151252 A CN 118151252A
- Authority
- CN
- China
- Prior art keywords
- data
- magnetic
- magnetic field
- magnetic disturbance
- analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/40—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation specially adapted for measuring magnetic field characteristics of the earth
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/38—Processing data, e.g. for analysis, for interpretation, for correction
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a satellite observation-based seismic magnetic disturbance analysis method, which comprises the steps of obtaining magnetic field data of a research area through satellite observation, preprocessing the magnetic field data, removing trend magnetic fields of the magnetic field data to obtain first data, and removing interference magnetic fields of the first data to obtain second data; performing multidimensional parameter analysis on the second data, and identifying magnetic disturbance characteristics related to the earthquake; carrying out causal analysis on the magnetic disturbance characteristics to obtain an analysis result; and establishing a seismic magnetic disturbance analysis model according to the first data, the magnetic disturbance characteristics and the analysis result, inputting observation data into the seismic magnetic disturbance analysis model, and outputting an identification result. The method improves the accuracy and reliability of earthquake prediction through multidimensional parameter analysis, causality examination and machine learning technology, and provides support for earthquake early warning and disaster relief.
Description
Technical Field
The invention relates to the field of seismic magnetic disturbance analysis, in particular to a satellite observation-based seismic magnetic disturbance analysis method.
Background
Seismic magnetic disturbance analysis is a key technique in seismology that studies precursor signals associated with seismic activity by monitoring changes in the earth's magnetic field. Traditional ground observation methods are limited by geographic location and environmental factors, while satellite observations provide broader coverage and more accurate data. However, the processing and analysis of satellite data requires complex algorithms to identify and exclude non-seismic related disturbances such as solar wind and geomagnetic activity. In addition, the establishment of the earthquake prediction model needs to integrate the magnetic disturbance characteristics of multiple dimensions and conduct effective causal relationship analysis.
Disclosure of Invention
The invention aims to provide a satellite observation-based seismic magnetic disturbance analysis method.
In order to achieve the above purpose, the invention is implemented according to the following technical scheme:
The invention comprises the following steps:
A, acquiring magnetic field data of a research area through satellite observation, preprocessing the magnetic field data, removing trend magnetic fields of the magnetic field data to obtain first data, and removing interference magnetic fields of the first data to obtain second data;
b, carrying out multidimensional parameter analysis on the second data, and identifying magnetic disturbance characteristics related to the earthquake;
c, carrying out causal analysis on the magnetic disturbance characteristics to obtain an analysis result;
and D, establishing a seismic magnetic disturbance analysis model according to the first data, the magnetic disturbance characteristics and the analysis result, inputting observation data into the seismic magnetic disturbance analysis model, and outputting an identification result.
Further, the method for preprocessing the magnetic field data of the research area obtained through satellite observation comprises the following steps:
The time of the magnetic field data comprises pre-earthquake, earthquake and post-earthquake, and the magnetic field data comprises three components including a vertical component, an east-west component and a north-south component, and is preprocessed: three components of the magnetic field data are converted into local field-oriented coordinates.
Further, the method for removing the trend magnetic field of the magnetic field data to obtain first data, removing the interference magnetic field of the first data, and obtaining second data comprises the following steps:
Defining a trend magnetic field using a Savitzky-Golay smoothing filter, the magnetic field data subtracting the trend magnetic field to obtain the first data; and carrying out signal analysis on the first data, wherein the interference magnetic field comprises solar wind disturbance and geomagnetic disturbance, and removing the interference magnetic field in the first data to obtain the second data.
Further, the method for identifying the magnetic disturbance characteristics related to the earthquake by carrying out multidimensional parameter analysis on the second data comprises the following steps:
extracting parameters of four dimensions of spectral density, polarization parameters, magnetic field strength and wave normal angle from the second data, and using the parameters to create a parameter set comprising labels distinguishing different of the parameters:
tq＝argmaxq{ypq}
Where t q is the q-th parameter, x q is the data point in t q, y pq is the tag of x p at t q,
Calculating the weight of the parameters according to the number of the parameters, inputting the parameter set into an isolated forest algorithm, constructing an isolated tree, training four isolated forests to analyze the four parameters, and calculating an anomaly score:
Where S p is the anomaly score for the data point, T is the set of parameters, ω q is the weight of the parameters, S pq is the anomaly score for the data point in T q, h (y) is the height of the data point in the orphan tree, c (α) is the average of the path lengths,
Regarding the super parameters of the isolated forest as the positions of particles, wherein the particles represent the configuration of the super parameters, and the positions and the speeds of the particles are updated according to the formula:
Where j denotes the index of the particle, d denotes the parameter, t denotes the number of iterations, ω is the inertial weight, c 1 and c 2 are learning factors, r 1 and r 2 are random factors, ranging between 0 and 1, For a historic optimal position of the particles,As a global optimum position for the device,
Evaluating the particles, calculating the accuracy of the anomaly score, updating the global optimal position after iteration until the accuracy tends to be stable,
Training the isolated forest using the optimized hyper-parametric configuration, calculating the anomaly score,
The data points and the parameters thereof in the anomaly score greater than 0.5 are taken as the magnetic disturbance characteristics.
Further, the method for carrying out causal analysis on the magnetic disturbance characteristics to obtain an analysis result comprises the following steps:
and (3) sequencing the magnetic disturbance characteristics according to time to obtain a time sequence, and constructing a VAR model for the magnetic disturbance characteristics, wherein a characteristic equation can be expressed as follows:
Where Y a,t is the value of the a-th feature at time t, d is the total number of features, gamma a,b,e is the effect of the a-th feature on the b-th feature on the e-th lag, delta a,t is the error term, is the difference between the a-th feature value at time t and the predictive value of the VAR model,
Selecting the best hysteresis order using information criteria:
where BIC represents a bayesian information criterion, Is the variance of the estimation error of the VAR model, g is the best hysteresis order,
Performing a gland cause and effect test on the magnetic disturbance characteristics:
where F is a statistic, checking whether the independent variable has significant predictive capability on the dependent variable, RSS R is the sum of squares residuals of the constrained VAR model, RSS UR is the sum of squares residuals of the unconstrained VAR model, q is the number of variables missing in the constrained VAR model,
Cross-validating the VAR model and the gland cause and effect test, testing accuracy,
Comparing the calculated statistic with a critical value, and when the statistic is greater than or equal to the critical value, indicating that the independent variable has a grange causal relationship to the dependent variable, and taking the grange causal relationship of the magnetic disturbance characteristic as the analysis result.
Further, a seismic magnetic disturbance analysis model is established according to the first data, the magnetic disturbance characteristics and the analysis result, observation data is input into the seismic magnetic disturbance analysis model, and the method for outputting the identification result comprises the following steps:
Adjusting weights of the magnetic disturbance features in the isolated forest algorithm according to the analysis result, removing the trend magnetic field by using the Savitzky-Golay smoothing filter to obtain first data, removing the interference magnetic field to obtain second data, calculating the magnetic disturbance features by using the adjusted weights in combination with the particle swarm optimization algorithm by using the isolated forest algorithm, and training a neural network algorithm according to the first data and the magnetic disturbance features:
preprocessing and feature extraction are performed on the first data, a feature matrix is established, a target vector is a binary label, whether the magnetic disturbance feature is contained or not is indicated, a data set is established by using the feature matrix and the target vector,
Dividing the data set into a training set and a testing set, training the neural network algorithm by using the training set, and designing an input layer, a function layer and an output layer:
the node of the input layer is the same as the dimension of the feature matrix, the function layer is n layers in total, the function layer learns the training set, and the function layer loses the function:
H(y,a)＝-(y·log(a)+(1-y)·log(1-a))
Where H (y, a) represents the loss function, y is the target vector, 0 or 1, a is a prediction probability, represents a probability predicted to contain the magnetic disturbance feature,
Adding an L2 regularization term to the loss function:
Wherein L regularized is the regularized loss function, W is a weight matrix, W ij is an element in W, lambda is a regularization coefficient,
The function layer updates weight through a gradient descent method:
Wherein, beta is the learning rate, For the gradient of the loss function L regularized to the weight matrix W, X is the feature matrix, λW is the gradient of the regularization term,
The output layer is the same as the nodes of the input layer, the activation function of the function layer is a Sigmoid function, the output layer outputs the prediction probability of the feature vector, the accuracy of the prediction probability is verified by using the test set,
And setting a threshold value for the prediction probability, and when the prediction probability is larger than the threshold value, indicating that the time of the first data corresponding to the feature vector is earthquake.
In a second aspect, an embodiment of the present application further provides an electronic device, including:
A processor; and a memory arranged to store computer executable instructions which, when executed, cause the processor to perform the method steps of the first aspect.
In a third aspect, embodiments of the present application also provide a computer-readable storage medium storing one or more programs, which when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method steps of the first aspect.
The beneficial effects of the invention are as follows:
according to the method, through satellite observation, multidimensional parameter analysis, causality verification and machine learning technology, the accuracy of the identification of the characteristics of the earthquake magnetic disturbance is improved, the earthquake activity can be predicted more effectively, and a scientific basis is provided for earthquake early warning.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is pre-processed magnetic field data;
FIG. 3 shows solar wind and geomagnetic disturbances before and after an earthquake;
FIG. 4 is magnetic field data after removing the interfering magnetic field;
FIG. 5 is a plot of data points in the magnetic perturbation signature.
Detailed Description
The invention is further described by the following specific examples, which are presented to illustrate, but not to limit, the invention.
The invention discloses a satellite observation-based seismic magnetic disturbance analysis method which comprises the following steps:
as shown in fig. 1, in this embodiment, the steps include:
The invention aims to provide a satellite observation-based seismic magnetic disturbance analysis method.
In order to achieve the above purpose, the invention is implemented according to the following technical scheme:
The invention comprises the following steps:
A, acquiring magnetic field data of a research area through satellite observation, preprocessing the magnetic field data, removing trend magnetic fields of the magnetic field data to obtain first data, and removing interference magnetic fields of the first data to obtain second data;
b, carrying out multidimensional parameter analysis on the second data, and identifying magnetic disturbance characteristics related to the earthquake;
c, carrying out causal analysis on the magnetic disturbance characteristics to obtain an analysis result;
and D, establishing a seismic magnetic disturbance analysis model according to the first data, the magnetic disturbance characteristics and the analysis result, inputting observation data into the seismic magnetic disturbance analysis model, and outputting an identification result.
In this embodiment, the center of the epicenter is located at 23.05 degrees in south latitude and 171.66 degrees in east longitude, the research area is in a range of 20 latitude by 20 longitude with the center of the epicenter, and the time range is 15 days before the earthquake to 7 days after the earthquake; fig. 2 shows the magnetic field data after preprocessing, wherein the stars on the horizontal axis represent latitude in the epicenter, the vertical axis represents time difference of each orbit relative to the current day of the earthquake, and a is the radial component, b is the azimuth component, and c is the parallel component.
In this embodiment, magnetic field data of a research area is acquired through satellite observation, and the method for preprocessing the magnetic field data includes:
The time of the magnetic field data comprises pre-earthquake, earthquake and post-earthquake, and the magnetic field data comprises three components including a vertical component, an east-west component and a north-south component, and is preprocessed: three components of the magnetic field data are converted to local field-oriented coordinates, including a radial component, an azimuthal component, and a parallel component.
In this embodiment, the trend of the magnetic field is defined by a polynomial order of 2 and a frame length of a sampling frequency of 25s fs, and fig. 3 shows solar wind parameters and SYM-H indexes from 15 days before to 7 days after the earthquake, showing solar wind disturbance and geomagnetic disturbance before and after the earthquake.
In this embodiment, the method for removing the trend magnetic field of the magnetic field data to obtain first data, removing the interference magnetic field of the first data, and obtaining second data includes:
Defining a trend magnetic field using a Savitzky-Golay smoothing filter, the magnetic field data subtracting the trend magnetic field to obtain the first data; and carrying out signal analysis on the first data, wherein the interference magnetic field comprises solar wind disturbance and geomagnetic disturbance, and removing the interference magnetic field in the first data to obtain the second data.
In this embodiment, fig. 4 is the second data, and the numbers around the track represent the time differences in days.
In this embodiment, the method for performing multidimensional parameter analysis on the second data and identifying the magnetic disturbance feature related to the earthquake includes:
extracting parameters of four dimensions of spectral density, polarization parameters, magnetic field strength and wave normal angle from the second data, and using the parameters to create a parameter set comprising labels distinguishing different of the parameters:
tq＝argmaxq{ypq〕
Where t q is the q-th parameter, x q is the data point in t q, y pq is the tag of x p at t q,
Calculating the weight of the parameter according to the number of the parameters:
Where N k is the number of data points contained by the parameter, N is the number of data points contained by the parameter set,
Inputting the parameter set into an isolated forest algorithm, constructing an isolated tree, training four isolated forests to analyze the four parameters, and calculating an abnormality score:
Where S p is the anomaly score for the data point, T is the set of parameters, ω q is the weight of the parameters, S pq is the anomaly score for the data point in T q, h (y) is the height of the data point in the orphan tree, c (α) is the average of the path lengths,
Regarding the super parameters of the isolated forest as the positions of particles, wherein the particles represent the configuration of the super parameters, and the positions and the speeds of the particles are updated according to the formula:
Where j denotes the index of the particle, d denotes the parameter, t denotes the number of iterations, ω is the inertial weight, c 1 and c 2 are learning factors, r 1 and r 2 are random factors, ranging between 0 and 1, For a historic optimal position of the particles,As a global optimum position for the device,
Evaluating the particles, calculating the accuracy of the anomaly score, updating the global optimal position after iteration until the accuracy tends to be stable,
Training the isolated forest using the optimized hyper-parametric configuration, calculating the anomaly score,
The data points and the parameters thereof in the anomaly score greater than 0.5 are taken as the magnetic disturbance characteristics.
In this embodiment, FIG. 5 is a data point in the magnetic disturbance signature.
In this embodiment, the method for performing causal analysis on the magnetic disturbance feature to obtain an analysis result includes:
and (3) sequencing the magnetic disturbance characteristics according to time to obtain a time sequence, and constructing a VAR model for the magnetic disturbance characteristics, wherein a characteristic equation can be expressed as follows:
Where Y a,t is the value of the a-th feature at time t, d is the total number of features, gamma a,b,e is the effect of the a-th feature on the b-th feature on the e-th lag, delta a,t is the error term, is the difference between the a-th feature value at time t and the predictive value of the VAR model,
Selecting the best hysteresis order using information criteria:
where BIC represents a bayesian information criterion, Is the variance of the estimation error of the VAR model, g is the best hysteresis order,
Performing a gland cause and effect test on the magnetic disturbance characteristics:
where F is a statistic, checking whether the independent variable has significant predictive capability on the dependent variable, RSS R is the sum of squares residuals of the constrained VAR model, RSS UR is the sum of squares residuals of the unconstrained VAR model, q is the number of variables missing in the constrained VAR model,
Cross-validating the VAR model and the gland cause and effect test, testing accuracy,
Comparing the calculated statistic with a critical value obtained by table lookup, and when the statistic is greater than or equal to the critical value, indicating that the independent variable has the grange causal relationship to the dependent variable, and taking the grange causal relationship of the magnetic disturbance characteristic as the analysis result.
In this embodiment, a seismic magnetic disturbance analysis model is established according to the first data, the magnetic disturbance characteristics and the analysis result, the observation data is input into the seismic magnetic disturbance analysis model, and the method for outputting the identification result includes:
And adjusting the weight of the magnetic disturbance characteristic in the isolated forest algorithm according to the analysis result:
Wherein W f is the adjusted weight of the magnetic perturbation feature, I f is the feature importance, represents the number of times the magnetic perturbation feature has the Grangel causality as an independent variable, sigma I f is the sum of the feature importance,
The seismic magnetic disturbance analysis model uses the Savitzky-Golay smoothing filter to remove the trend magnetic field to obtain first data, removes the interference magnetic field to obtain second data, uses the adjusted weight to calculate the magnetic disturbance characteristic by adopting the isolated forest algorithm and the particle swarm optimization algorithm, and trains a neural network algorithm according to the first data and the magnetic disturbance characteristic:
preprocessing and feature extraction are performed on the first data, a feature matrix is established, a target vector is a binary label, whether the magnetic disturbance feature is contained or not is indicated, a data set is established by using the feature matrix and the target vector,
Dividing the data set into a training set and a testing set, training the neural network algorithm by using the training set, and designing an input layer, a function layer and an output layer:
the node of the input layer is the same as the dimension of the feature matrix, the function layer is n layers in total, the function layer learns the training set, and the function layer loses the function:
H(y,a)＝-(y·log(a)+(1-y)·log(1-a))
Where H (y, a) represents the loss function, y is the target vector, 0 or 1, a is a prediction probability, represents a probability predicted to contain the magnetic disturbance feature,
Adding an L2 regularization term to the loss function:
Wherein L regularized is the regularized loss function, W is a weight matrix, W ij is an element in W, lambda is a regularization coefficient,
The function layer updates weight through a gradient descent method:
Wherein, beta is the learning rate, For the gradient of the loss function L regularized to the weight matrix W, X is the feature matrix, λW is the gradient of the regularization term,
The output layer is the same as the nodes of the input layer, the activation function of the function layer is a Sigmoid function, the output layer outputs the prediction probability of the feature vector, the accuracy of the prediction probability is verified by using the test set,
And setting a threshold value for the prediction probability, and when the prediction probability is larger than the threshold value, indicating that the time of the first data corresponding to the feature vector is earthquake.
The processor, network interface, and memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture ) bus, a PCI (PeripheralComponent Interconnect, peripheral component interconnect standard) bus, or EISA (Extended IndustryStandard Architecture ) bus, among others. The buses may be classified as address buses, data buses, control buses, etc.
And the memory is used for storing programs. In particular, the program may include program code including computer-operating instructions. The memory may include memory and non-volatile storage and provide instructions and data to the processor.
The processor reads the corresponding computer program from the nonvolatile memory to the memory and then runs the computer program to form the seismic magnetic disturbance analysis device based on satellite observation on a logic level. And the processor is used for executing the programs stored in the memory and particularly used for executing any one of the earthquake magnetic disturbance analysis methods.
The method for analyzing the seismic magnetic disturbance based on satellite observation disclosed in the embodiment shown in fig. 1 of the present application can be applied to a processor or implemented by the processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (NetworkProcessor, NP), etc.; but may also be a digital signal Processor (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), field-Programmable gate array (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.
The electronic device may also execute the satellite observation-based seismic magnetic disturbance analysis method in fig. 1, and implement the functions of the embodiment shown in fig. 1, which is not described herein.
The embodiments of the present application also provide a computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by an electronic device comprising a plurality of application programs, perform any of the satellite observation based seismic magnetic disturbance analysis methods described above.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer 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, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transitorymedia), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (8)
1. The method for analyzing the seismic magnetic disturbance based on satellite observation is characterized by comprising the following steps of:
A, acquiring magnetic field data of a research area through satellite observation, preprocessing the magnetic field data, removing trend magnetic fields of the magnetic field data to obtain first data, and removing interference magnetic fields of the first data to obtain second data;
b, carrying out multidimensional parameter analysis on the second data, and identifying magnetic disturbance characteristics related to the earthquake;
c, carrying out causal analysis on the magnetic disturbance characteristics to obtain an analysis result;
and D, establishing a seismic magnetic disturbance analysis model according to the first data, the magnetic disturbance characteristics and the analysis result, inputting observation data into the seismic magnetic disturbance analysis model, and outputting an identification result.
2. The method for analyzing seismic magnetic disturbance based on satellite observation according to claim 1, wherein the step A of acquiring magnetic field data of a research area through satellite observation includes:
The time of the magnetic field data comprises pre-earthquake, earthquake and post-earthquake, and the magnetic field data comprises three components including a vertical component, an east-west component and a north-south component, and is preprocessed: three components of the magnetic field data are converted to local field-oriented coordinates, including a radial component, an azimuthal component, and a parallel component.
3. The method of claim 1, wherein the removing the trend magnetic field of the magnetic field data in the step B to obtain first data, and removing the interfering magnetic field of the first data to obtain second data comprises:
Defining a trend magnetic field using a Savitzky-Golay smoothing filter, the magnetic field data subtracting the trend magnetic field to obtain the first data; and carrying out signal analysis on the first data, wherein the interference magnetic field comprises solar wind disturbance and geomagnetic disturbance, and removing the interference magnetic field in the first data to obtain the second data.
4. The satellite observation based seismic magnetic disturbance analysis method according to claim 1, wherein the step C of performing a multidimensional parameter analysis on the second data includes:
extracting parameters of four dimensions of spectral density, polarization parameters, magnetic field strength and magnetic field direction change rate from the second data, and establishing a parameter set with the parameters, the parameter set comprising labels distinguishing different of the parameters:
tq＝argmaxq{ypq}
Where t q is the q-th parameter, x q is the data point in t q, y pq is the tag of x p at t q,
Calculating the weight of the parameters according to the number of the parameters, inputting the parameter set into an isolated forest algorithm, constructing an isolated tree, training four isolated forests to analyze the four parameters, and calculating an anomaly score:
Where S p is the anomaly score for the data point, T is the set of parameters, ω q is the weight of the parameters, S pq is the anomaly score for the data point in T q, h (y) is the height of the data point in the orphan tree, c (α) is the average of the path lengths,
Regarding the super parameters of the isolated forest as the positions of particles, wherein the particles represent the configuration of the super parameters, and the positions and the speeds of the particles are updated according to the formula:
Where j denotes the index of the particle, d denotes the parameter, t denotes the number of iterations, ω is the inertial weight, c 1 and c 2 are learning factors, r 1 and r 2 are random factors, ranging between 0 and 1, For the historic best position of the particle,/>As a global optimum position for the device,
Evaluating the particles, calculating the accuracy of the anomaly score, updating the global optimal position after iteration until the accuracy tends to be stable,
Training the isolated forest using the optimized hyper-parametric configuration, calculating the anomaly score,
The data points and the parameters thereof in the anomaly score greater than 0.5 are taken as the magnetic disturbance characteristics.
5. The method for analyzing seismic magnetic disturbance based on satellite observation according to claim 1, wherein the method for performing causal analysis on the magnetic disturbance characteristic in step C to obtain an analysis result comprises:
and (3) sequencing the magnetic disturbance characteristics according to time to obtain a time sequence, and constructing a VAR model for the magnetic disturbance characteristics, wherein a characteristic equation can be expressed as follows:
Where Y a,t is the value of the a-th feature at time t, d is the total number of features, gamma a,b,e is the effect of the a-th feature on the b-th feature on the e-th lag, delta a,t is the error term, is the difference between the a-th feature value at time t and the predictive value of the VAR model,
Selecting the best hysteresis order using information criteria:
where BIC represents a bayesian information criterion, Is the variance of the estimation error of the VAR model, g is the best hysteresis order,
Performing a gland cause and effect test on the magnetic disturbance characteristics:
where F is a statistic, checking whether the independent variable has significant predictive capability on the dependent variable, RSS R is the sum of squares residuals of the constrained VAR model, RSS UR is the sum of squares residuals of the unconstrained VAR model, q is the number of variables missing in the constrained VAR model,
Cross-validating the VAR model and the gland cause and effect test, testing accuracy,
Comparing the calculated statistic with a critical value, and when the statistic is greater than or equal to the critical value, indicating that the independent variable has a grange causal relationship to the dependent variable, and taking the grange causal relationship of the magnetic disturbance characteristic as the analysis result.
6. The method for analyzing seismic magnetic disturbance based on satellite observation according to claim 1, wherein in the step D, a seismic magnetic disturbance analysis model is established according to the first data, the magnetic disturbance characteristics and the analysis result, the observation data is input into the seismic magnetic disturbance analysis model, and the method for outputting the identification result comprises:
Adjusting weights of the magnetic disturbance features in the isolated forest algorithm according to the analysis result, removing the trend magnetic field by using the Savitzky-Golay smoothing filter to obtain first data, removing the interference magnetic field to obtain second data, calculating the magnetic disturbance features by using the adjusted weights in combination with the particle swarm optimization algorithm by using the isolated forest algorithm, and training a neural network algorithm according to the first data and the magnetic disturbance features:
preprocessing and feature extraction are performed on the first data, a feature matrix is established, a target vector is a binary label, whether the magnetic disturbance feature is contained or not is indicated, a data set is established by using the feature matrix and the target vector,
Dividing the data set into a training set and a testing set, training the neural network algorithm by using the training set, and designing an input layer, a function layer and an output layer:
the node of the input layer is the same as the dimension of the feature matrix, the function layer is n layers in total, the function layer learns the training set, and the function layer loses the function:
H(y,a)＝-(y·log(a)+(1-y)·log(1-a))
Where H (y, a) represents the loss function, y is the target vector, 0 or 1, a is a prediction probability, represents a probability predicted to contain the magnetic disturbance feature,
Adding an L2 regularization term to the loss function:
Wherein L regularized is the regularized loss function, W is a weight matrix, W ij is an element in W, lambda is a regularization coefficient,
The function layer updates weight through a gradient descent method:
Wherein, beta is the learning rate, For the gradient of the loss function L regularized to the weight matrix W, X is the feature matrix, λW is the gradient of the regularization term,
The output layer is the same as the nodes of the input layer, the activation function of the function layer is a Sigmoid function, the output layer outputs the prediction probability of the feature vector, the accuracy of the prediction probability is verified by using the test set,
And setting a threshold value for the prediction probability, and when the prediction probability is larger than the threshold value, indicating that the time of the first data corresponding to the feature vector is earthquake.
7. An electronic device, comprising: a processor; and
A memory arranged to store computer executable instructions which, when executed, cause the processor to perform the method of any of claims 1 to 6.
8. A computer readable storage medium storing one or more programs, which when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410275650.0A CN118151252B (en) | 2024-03-12 | 2024-03-12 | Satellite observation-based seismic magnetic disturbance analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410275650.0A CN118151252B (en) | 2024-03-12 | 2024-03-12 | Satellite observation-based seismic magnetic disturbance analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118151252A true CN118151252A (en) | 2024-06-07 |
| CN118151252B CN118151252B (en) | 2024-08-16 |
Family
ID=91298379
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410275650.0A Active CN118151252B (en) | 2024-03-12 | 2024-03-12 | Satellite observation-based seismic magnetic disturbance analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118151252B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118897326A (en) * | 2024-07-09 | 2024-11-05 | 北京自动化控制设备研究所 | Magnetic anomaly perception method and related device based on improved RLMD and support vector machine |
| CN119126200A (en) * | 2024-08-31 | 2024-12-13 | 中信建筑设计研究总院有限公司 | A method and device for selecting earthquake records in elastic time history analysis |
| CN119781007A (en) * | 2024-12-17 | 2025-04-08 | 中国地震局地震预测研究所 | Space differential disturbance extraction method based on satellite in-situ electron density data |
| CN119882089A (en) * | 2025-02-21 | 2025-04-25 | 北京市地震局 | Seismic geophysical data anomaly detection and early warning method based on big data analysis |
| CN119917985A (en) * | 2025-04-02 | 2025-05-02 | 国网四川省电力公司电力科学研究院 | Photovoltaic power distribution operation early warning method, system, device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7277797B1 (en) * | 2005-03-29 | 2007-10-02 | Kunitsyn Viatcheslav E | Prediction system and method |
| CN111814699A (en) * | 2020-07-13 | 2020-10-23 | 中国地震局地震预测研究所 | A deep learning earthquake prediction method for SWARM electromagnetic satellite data |
| CN116881798A (en) * | 2023-08-01 | 2023-10-13 | 大连理工大学 | Conditional gracile causal analysis method based on variable selection and reverse time lag feature selection for complex systems such as weather |
-
2024
- 2024-03-12 CN CN202410275650.0A patent/CN118151252B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7277797B1 (en) * | 2005-03-29 | 2007-10-02 | Kunitsyn Viatcheslav E | Prediction system and method |
| CN111814699A (en) * | 2020-07-13 | 2020-10-23 | 中国地震局地震预测研究所 | A deep learning earthquake prediction method for SWARM electromagnetic satellite data |
| CN116881798A (en) * | 2023-08-01 | 2023-10-13 | 大连理工大学 | Conditional gracile causal analysis method based on variable selection and reverse time lag feature selection for complex systems such as weather |
Non-Patent Citations (6)
| Title |
|---|
| F. ZHOU 等: "Earthquake Electromagnetic Precursor Anomaly Detection Method Based on EEMD-LSTM-IF", 2024 7TH INTERNATIONAL CONFERENCE ON ADVANCED ALGORITHMS AND CONTROL ENGINEERING (ICAACE), 3 March 2024 (2024-03-03), pages 524 - 527, XP034619849, DOI: 10.1109/ICAACE61206.2024.10549491 * |
| MIRKO STUMPO 等: "Measuring Information Coupling between the Solar Wind and the Magnetosphere–Ionosphere System", ENTROPY, vol. 22, no. 3, 28 February 2020 (2020-02-28), pages 1 - 13 * |
| YIFEI LIU 等: "Research on a matching detection method for magnetic anomaly of underwater targets", AIP ADVANCES, vol. 13, no. 2, 9 February 2023 (2023-02-09), pages 1 - 11, XP012272255, DOI: 10.1063/5.0137954 * |
| 欧阳新艳: "利用DEMETER卫星数据提取地震异常的方法初探 ————以普洱地震为例", 中国优秀硕士学位论文全文数据库 基础科学辑, no. 1, 15 January 2009 (2009-01-15), pages 012 - 10 * |
| 胡云鹏: "地震和磁暴引起的电离层电磁波动特征研究", 中国优秀硕士学位论文全文数据库 基础科学辑, no. 2, 15 February 2021 (2021-02-15), pages 012 - 90 * |
| 闫相相: "基于GPS和DEMETER卫星数据的地震电离层电子浓度异常变化研究", 中国博士学位论文全文数据库 基础科学辑, no. 5, 15 May 2014 (2014-05-15), pages 008 - 15 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118897326A (en) * | 2024-07-09 | 2024-11-05 | 北京自动化控制设备研究所 | Magnetic anomaly perception method and related device based on improved RLMD and support vector machine |
| CN119126200A (en) * | 2024-08-31 | 2024-12-13 | 中信建筑设计研究总院有限公司 | A method and device for selecting earthquake records in elastic time history analysis |
| CN119781007A (en) * | 2024-12-17 | 2025-04-08 | 中国地震局地震预测研究所 | Space differential disturbance extraction method based on satellite in-situ electron density data |
| CN119781007B (en) * | 2024-12-17 | 2025-06-17 | 中国地震局地震预测研究所 | A spatial differential disturbance extraction method based on satellite in-situ electron density data |
| CN119882089A (en) * | 2025-02-21 | 2025-04-25 | 北京市地震局 | Seismic geophysical data anomaly detection and early warning method based on big data analysis |
| CN119917985A (en) * | 2025-04-02 | 2025-05-02 | 国网四川省电力公司电力科学研究院 | Photovoltaic power distribution operation early warning method, system, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118151252B (en) | 2024-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118151252B (en) | Satellite observation-based seismic magnetic disturbance analysis method | |
| Wagner | Direct multi‐scale ordination with canonical correspondence analysis | |
| US20180157971A1 (en) | Probabilistic Matrix Factorization for Automated Machine Learning | |
| CN111767957B (en) | Log abnormality detection method and device, storage medium and electronic equipment | |
| CN109113729B (en) | Lithology identification method and device based on well logging curve | |
| CN113361853B (en) | Satellite emergency task planning scheme efficiency evaluation method and system of new consensus model | |
| CN118228585A (en) | Method for constructing earthquake prediction model based on space-time evolution characteristics | |
| Brugere et al. | Improved prediction of tree species richness and interpretability of environmental drivers using a machine learning approach | |
| US12314848B2 (en) | Computer system and model training method | |
| CN112200355A (en) | Landslide prediction method, device, equipment and storage medium | |
| CN115758876A (en) | Method, system and computer equipment for forecasting accuracy of wind speed and wind direction | |
| Haining | Specification and estimation problems in models of spatial dependence. | |
| Deb et al. | Assessment of a new quality control technique in the retrieval of atmospheric motion vectors | |
| CN119380204A (en) | Land cover change monitoring method, model training method and related device | |
| Galli et al. | Automated machine learning: a case study of genomic “image-based” prediction in maize hybrids | |
| CN119048830A (en) | Solar radio burst prediction method and system based on full-disk solar magnetic map | |
| Holm et al. | Data assimilation for ocean drift trajectories using massive ensembles and GPUs | |
| CN116152751A (en) | Image processing method, device, system and storage medium | |
| Barthel et al. | Comparative Analysis of MaxEnt and Deep Learning Approaches for Modeling Humpback Whale Distribution in North Iceland | |
| CN115144659A (en) | An electromagnetic signal analysis method, device, readable medium and electronic device | |
| Jiang et al. | Research on Argo Data Anomaly Detection Based on Improved DBSCAN Algorithm | |
| CN119781007B (en) | A spatial differential disturbance extraction method based on satellite in-situ electron density data | |
| CN112561171A (en) | Landslide prediction method, device, equipment and storage medium | |
| CN119249355B (en) | Multi-source alarm processing method, device, electronic device and storage medium | |
| CN119272162B (en) | A method for accurately identifying pollution sources in karst underground river systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |