Earthquake prediction system based on neural network

Abstract

The invention provides an earthquake prediction system based on a neural network, and belongs to the field of artificial intelligence. The system comprises a data acquisition device, an earthquake prediction device and a monitoring alarm center. The data acquisition device is mainly formed by a seismometer and a network module, and is used for collecting seismic waves and transmitting the data through a network. The earthquake prediction device is generated through the following steps: to begin with, selecting a lot of training sample sets, which are mainly related to seismic wave pictures; then, constructing a convolution neural network system; and training the neural network through training data, and when training is finished, generating the earthquake prediction system. The monitoring alarm center is mainly used for analyzing the output of the neural network and the data of the data acquisition device to predicate which place is about to have an earthquake. Through the universal convolution neural network model, the earthquake prediction system overcomes the defect that a feature matching algorithm is not completely considered, has good generalization ability and improves predication accuracy.

Description

technical field [0001] The invention relates to artificial intelligence technology, in particular to an earthquake prediction system based on a neural network. Background technique [0002] According to statistics, there are more than 5 million earthquakes every year on the earth, among which there are more than 10 earthquakes that can cause serious harm to human beings. pose a serious burden. [0003] The current level of science and technology cannot accurately predict the arrival of earthquakes, so a new type of earthquake prediction device is urgently needed. Contents of the invention [0004] Nowadays, with the rise of artificial intelligence neural networks, it is possible for machines to abstract themselves and learn the characteristics of things. Therefore, the present invention proposes an earthquake prediction system based on a neural network. Machines can help us predict earthquakes by analyzing data from earthquakes over the years. [0005] The solution of ...

AI Technical Summary

Problems solved by technology

[0002] According to statistics, there are more than 5 million earthquakes every year on the earth, among which there are more than 10 earthquakes that can cause serious harm to human beings. bring a serious burden

[0003] The current level of science and technology cannot accurately predict the arrival of earthquakes, so a new type of earthquake prediction device is urgently needed

Images