Seismic positioning method and equipment based on fusion of machine learning and dynamic computing
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A technology of machine learning and positioning methods, applied in the field of earthquake positioning, can solve problems such as limited historical data and inaccurate source locations
Active Publication Date: 2022-04-12
PEKING UNIV
View PDF13 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
However, due to the inaccurate location of the epicenter in the historical data and the limited amount of historical data, the positioning results obtained by the trained positioning model also have limitations.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0057] This embodiment implements an earthquake location method based on the fusion of machine learning and dynamic calculation, such as figure 1 shown, including the following steps:
[0058] S1. Acquire multi-source data, and perform regularization processing on the multi-source data;
[0059] S2. Perform short-time Fourier transform on the regularized data to obtain a time-frequency image as a sample, generate label data for each sample according to the three-dimensional Gaussian distribution according to the source position, and use the cross-validation method to generate a training set and a test set;
[0060] S3. Perform full convolutional neural network model training based on the training set, and evaluate the accuracy of the trained model based on the test set to obtain a trained model;
[0061] S4. Input the real-time monitoring data into the trained model to preliminarily estimate the source position;
[0062] S5. Correct the preliminary estimated hypocenter posit...
Embodiment 2
[0086] This embodiment implements an earthquake location method based on machine learning and dynamic calculation fusion, including the following steps:
[0087] Obtain multi-source data, and perform regularization processing on the multi-source data;
[0088] Perform short-time Fourier transform on the regularized data to obtain time-frequency images as samples, generate label data for each sample according to the three-dimensional Gaussian distribution according to the source position, and use cross-validation method to generate training set and test set;
[0089] The full convolutional neural network model is trained based on the training set, and the accuracy of the trained model is evaluated based on the test set to obtain the trained model;
[0090] Input the real-time monitoring data into the trained model to preliminarily estimate the source location;
[0091] The initial estimation of the hypocentral position is corrected by the dynamic calculation model.
[0092] T...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The present application relates to the technical field of earthquake positioning, and more specifically, the present application relates to an earthquake positioning method and equipment based on fusion of machine learning and dynamic calculation. The method includes: acquiring multi-source data, and performing regularization processing on the multi-source data; performing short-time Fourier transform on the regularized data to obtain a time-frequency image as a sample, and for each sample according to the source position Generate label data according to the three-dimensional Gaussian distribution, and use the cross-validation method to generate training sets and test sets; conduct full convolutional neural network model training based on the training set, and evaluate the accuracy of the trained model based on the test set to obtain the trained model; real-time The monitoring data is input into the trained model to preliminarily estimate the hypocenter location; the preliminarily estimated hypocenter location is corrected by using the dynamic calculation model. The method and device can quickly realize the preliminary location of the seismic source, and then use the dynamic calculation model to further determine the precise location of the seismic source.
Description
technical field [0001] The present application relates to the technical field of earthquake positioning, and more specifically, the present application relates to an earthquake positioning method and equipment based on fusion of machine learning and dynamic calculation. Background technique [0002] my country is a country with frequent earthquakes, and earthquakes are extremely destructive natural disasters, which pose a great threat to people's lives and property safety. Earthquakes caused by geological structures transmit energy outward through seismic waves, causing deformation of the ground surface and destruction of landmark building structures, and causing a series of secondary disasters. When an earthquake occurs, quickly and accurately determining the location of the earthquake source will help improve the timeliness and reliability of earthquake early warning. [0003] Due to the complexity of the earthquake scene, although great achievements have been made in ear...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more