Neural Network Model for Earthquake Prediction Using DMETER Data and Seismic Belt Information
Authors:
- Fangzhou XuFXFangzhou Xu
- Xianfeng SongXSXianfeng Song
- Xinhong WangXWXinhong Wang
- Juexiao SuJSJuexiao Su
- 1 December 2010
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3 (21556083) , 180-183
Abstract
The mechanism of the earthquake remains to be investigated, though some anomalies associated with earthquakes have been found by DEMETER satellite observations. The next step is to use the DEMETER data for earthquake prediction. It is a useful and feasible way to use the self-adaptive artificial neural network to construct relations between various symptom factors and earthquake occurrences. The back-propagation neural network is quite suitable to express the nonlinear relation between earthquake and various anomalies. In this paper a series of physical quantities measured by the DEMETER satellite including Electron density, Electron temperature, ions temperature and oxygen ion density, together with seismic belt information are used to form sample sets for a back-propagation neural network. The neural network model then can be used to conduct the prediction. In the end, validation tests are performed based on those important seismic events happened in 2008.Keywords
This publication has 6 references indexed in Scilit:
- Electron and ion density variations before strong earthquakes (M>6.0) using DEMETER and GPS dataNatural Hazards and Earth System Sciences, 2010
- Spacecraft observations of electromagnetic perturbations connected with seismic activityGeophysical Research Letters, 2008
- The micro-satellite DEMETERJournal of Geodynamics, 2002
- Accelerating neural network training using weight extrapolationsNeural Networks, 1999
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989
- Learning Internal Representations by Error Propagation1988