Earthquake monitoring method and device based on deep learning

Abstract

The invention discloses an earthquake monitoring method and device based on deep learning. The method comprises the steps: training a neural network; intercepting a seismic waveform time window from the continuous data at a preset time interval; inputting the seismic waveform time window into the neural network; the neural network outputs corresponding Gaussian probability density distribution; judging whether the maximum value of the Gaussian probability density distribution is greater than a preset threshold value or not; if yes, judging that the seismic waveform time window contains a seismic event, and calculating a seismic source parameter solution; and if not, judging that the seismic waveform time window does not contain the seismic event. According to the earthquake monitoring method and device based on deep learning provided by the invention, the earthquake source parameters can be estimated by using a single machine, so that a generalized neural network model can be obtained,and after the neural network is trained by using global data, the network model can be applied to any region.

Description

technical field [0001] The invention belongs to the technical field of earthquake monitoring, and in particular relates to an earthquake monitoring method and device based on deep learning. Background technique [0002] The earthquake early warning and monitoring system aims to quickly generate parameters such as magnitude and location before the arrival of destructive earthquake waves, and give early warning at the first time. Earthquake early warning systems that have been built and put into practical use in the world mainly include Japan's REIS, Mexico's SAS, China's VSN, and Turkey's IERREWS. In addition, the early warning system in California is in real-time testing. At present, the research and development of earthquake early warning systems in China is also in a very active period. In addition to the earthquake early warning systems of some key projects such as nuclear power plants and high-speed railways, the systems of the Fujian Provincial Seismological Bureau, the...

AI Technical Summary

Problems solved by technology

At present, the Gaussian probability density distribution is used to mark the earthquake location, and the full convolutional neural network is used to solve the source location of the induced earthquake. The prediction accuracy can match the results of artificial processing to a certain extent. However, this method is only for earthquakes that have been detected. After the localization problem, it is impossible to detect seismic events from continuous data and estimate the magnitude and time of earthquakes, etc.

Images