Unlock instant, AI-driven research and patent intelligence for your innovation.

A Method of Seismic Phase Picking and Event Detection Based on Convolutional Neural Network

A convolutional neural network and event detection technology, which is applied in the field of seismic phase picking and event detection based on convolutional neural network, can solve the problems of misdetection of events, high error rate of seismic phase recognition of seismic events, and high rate of missed detection. Achieve the effect of improving the accuracy rate, fast and accurate calculation, and realizing accurate estimation

Active Publication Date: 2022-04-01
CTBT BEIJING NAT DATA CENT
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Although scholars have made a lot of efforts and proposed a variety of methods, there are still many problems in the detection of regional seismic events, such as high error rate of seismic phase identification, event misdetection, and high rate of missed detection.

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
  • A Method of Seismic Phase Picking and Event Detection Based on Convolutional Neural Network
  • A Method of Seismic Phase Picking and Event Detection Based on Convolutional Neural Network
  • A Method of Seismic Phase Picking and Event Detection Based on Convolutional Neural Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0056] According to the above method, take a local earthquake event recorded by Changbai (CBT), Yanbian (YNB), Kuandian (KDN) and Mudanjiang (MDJ) stations on August 5, 2018 as an example. Firstly, the trained convolutional neural network is used to process the data within the detection window (5 minutes). According to the method in this paper, the detection of the P and S seismic phases of each station is realized, the name of the seismic phase is identified, and the arrival time of the seismic phase is estimated. The seismic phase information is shown in Table 1 (assuming that the event detection window is between 1300s and 1600s). Then, in the stations with opposite P and S earthquakes, calculate α=T two by two Si -T Sj / T Pi -T Pj =v P / v S value, get α 1 = 1.67, α 2 =1.73,α 3 =1.71,α 4 =1.74,α 5 = 1.8, α 6 =1.67, the threshold δ is set to 0.5 based on historical event statistics, because α 1 -α 2 2 -α 3 1 , α 2 , α 3 ,... Take the average value to get the ...

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

No PUM Login to View More

Abstract

The invention relates to a seismic phase picking and event detection method based on a convolutional neural network, belonging to the field of seismic event detection and estimation. In order to overcome the problems of high seismic phase recognition error rate, event misdetection, and missed detection rate in seismic event detection, the present invention slides and intercepts continuous data of several stations with a set time window and step size. Each window to be detected is detected by using the pre-trained convolutional neural network model to obtain the seismic phase recognition probability sequence and the arrival time fitting value sequence, and then the seismic phase probability value and the arrival time fitting value sequence are obtained. The extreme value of the combined value determines the name of the seismic phase and the estimated value of the arrival time; the initial position and time of the event are estimated according to the arrival time difference of the near earthquake P and S, which are used as the initial value of the conventional iterative inversion positioning program to obtain the final event position and time, so as to realize the accurate identification of the seismic phase and the estimation of the arrival time, as well as the accurate location and time of the seismic event.

Description

technical field [0001] The invention belongs to the field of seismic event detection and estimation, and in particular relates to a convolutional neural network-based seismic phase picking and event detection method. Background technique [0002] Seismic event detection is the process of inversion and formation of events based on the signals and characteristics recorded by monitoring stations, generally including the detection of station signals, arrival time estimation, identification of seismic phases, correlation and positioning of multiple earthquakes, etc. The regional network is of great significance for the rapid and reliable detection of near-earthquake events in the network, for earthquake prevention and disaster reduction, and emergency response. The academic community has carried out extensive research on the problem of seismic event detection, and established a relatively mature set of methods, such as signal detection based on the short-term average and long-ter...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G01V1/30G01V1/36
CPCG01V1/30G01V1/303G01V1/364
Inventor 李健王晓明王娟邱宏茂朱国富
Owner CTBT BEIJING NAT DATA CENT