Earthquake magnitude judgment method based on deep learning

Abstract

The invention discloses an earthquake magnitude judgment method based on deep learning, which comprises the following steps of: firstly, calculating the distance between an earthquake source and a station, screening out earthquake data of which the epicentral distance is less than 200km, then reading the data, and accurately finding P wave arrival time of the earthquake data by using a method of combining STA / LTA and AIC; then, intercepting waveform data with a certain length from the arrival time of the P wave; secondly, judging whether the waveform data with the certain length is broken or not and whether the waveform data is valid or not, and normalizing the valid data which is not broken or not; and finally, inputting the processed seismic data into a Bi-LSTM network model fused with an attention mechanism, training the network model, and storing the trained model; and inputting the preprocessed seismic waveform data into the trained network model, and outputting the seismic magnitude corresponding to the seismic data. According to the method, the magnitude corresponding to the seismic waveform can be accurately and efficiently judged, and the requirement for intelligent and automatic development of seismic monitoring is met.

Description

technical field [0001] The invention relates to the technical field of earthquake magnitude determination, in particular to a deep learning-based earthquake magnitude determination method. Background technique [0002] Earthquakes are vibrations caused by the rapid release of energy from the earth's crust, and are a common and common natural phenomenon. Earthquakes are sudden and destructive. According to statistics, more than 5 million earthquakes occur on the earth every year. Earthquakes bring huge loss of life and property to the people in the earthquake areas. The magnitude of earthquake damage is closely related to the magnitude of the earthquake. Seismic stations monitor earthquake events and give their magnitudes. Rapid and accurate magnitude identification is of great significance to earthquake prediction and subsequent research. In addition, the magnitude of the earthquake is related to the later earthquake disaster assessment and emergency response. Nowadays, th...

AI Technical Summary

Problems solved by technology

[0003] At present, the determination of earthquake magnitude is mainly based on the theoretical knowledge and experience of earthquake professionals, and the specific magnitude is calculated according to the formula. This method requires many prior parameters, and cannot guarantee the accuracy of calculation in a short time, and it is difficult to meet the automatic magnitude. Determination of needs

Images