ArticlePDF Available

Relationship between isoseismal area and magnitude of historical earthquakes in Greece by a hybrid fuzzy neural network method

Authors:
Akis Tselentis at National and Kapodistrian University of Athens
Akis Tselentis
Efthimios Sokos at University of Patras
Efthimios Sokos
Download full-text PDFDownload full-text PDF
Read full-text
Download citation

Abstract and Figures

In this paper we suggest the use of diffusion-neural-networks, (neural networks with intrinsic fuzzy logic abilities) to assess the relationship between isoseismal area and earthquake magnitude for the region of Greece. It is of particular importance to study historical earthquakes for which we often have macroseismic information in the form of isoseisms but it is statistically incomplete to assess magnitudes from an isoseismal area or to train conventional artificial neural networks for magnitude estimation. Fuzzy relationships are developed and used to train a feed forward neural network with a back propagation algorithm to obtain the final relationships. Seismic intensity data from 24 earthquakes in Greece have been used. Special attention is being paid to the incompleteness and contradictory patterns in scanty historical earthquake records. The results show that the proposed processing model is very effective, better than applying classical artificial neural networks since the magnitude macroseismic intensity target function has a strong nonlinearity and in most cases the macroseismic datasets are very small.
Figures - uploaded by Efthimios Sokos
Author content
All figure content in this area was uploaded by Efthimios Sokos
Content may be subject to copyright.
54357c270cf2dc341db2b1
06.pdf
Content uploaded by Efthimios Sokos
Author content
All content in this area was uploaded by Efthimios Sokos on Oct 08, 2014
Content may be subject to copyright.
Relationship between isoseismal area and magnitude of historical earthquakes in Greece by a hybrid fuzzy neural network meth
od.pdf
Available via license: CC BY 3.0
Content may be subject to copyright.
A preview of the PDF is not available
Agent Based Prediction of Seismic Time Series Data
Conference Paper
  • Dec 2012
In past formal statistical and evolutionary techniques have been successful in prediction of time series data. Traditional focus of these techniques encircled increase in reliability of results, consistency, accuracy and sustainability of overall system. Formal systems have been utilized to predict the earthquake structure and parameters that best fit into that structure. This research changes the orientation of traditional trends by introducing novel data quality and automation means to increase not only the accuracy but also the efficiency of prediction system. Data related to seismic activities in the geography related to Pakistan has been considered for experimentation. Moreover an intelligent methodology has been designed to achieve the desired results in an interactive manner.
View
A TOOL FOR COLLECTING, QUERYING AND MINING MACROSEISMIC DATA
Article
Full-text available
  • Jan 2004
SEISMO-SURFER is a tool for collecting, querying and mining seismic data being developed in Java programming language using Oracle database system. The objective is to combine recent research trends and results in the fields of spatial and spatio-temporal databases, data warehouses and data mining, as well as well established visualization techniques for geographical information. The database of the tool is automatically updated from remote sources while existing possibilities allow the querying on different earthquakes parameters, the analysis of the data for extraction of useful information and the graphical representation of the results via maps, charts etc. In the present work, we extend SEISMO-SURFER to include macroseismic data collected by the Geodynamic Institute and filled in a relative database. More specifically, the seismic parameters of the strong earthquakes, stored into the SEISMO-SURFER database, are linked to the macroseismic intensities observed at different sites. Administrative information for each site, local surface geology, tectonic lines, damage photographs and detailed descriptions from newspapers are also included. University of Piraeus and Geodynamic Institute are working together to continuously update and develop SEISMO-SURFER, concerning the data included, the variety of parameters stored and the mining algorithms supported for exploiting knowledge.
View
View
View
Modelling and Monitoring of Milling through Neuro-Fuzzy Techniques
Article
  • Jun 1994
Real-time nature, uncertainty handling and learning ability are essential requirements for knowledge representation and processing techniques to be applied at lower levels of intelligent manufacturing systems. The paper demonstrates and compares the applicability of neural networks and neuro fuzzy techniques for monitoring of milling tools. Learning and classification performances of back propagation (BP) networks and the neuro-fuzzy approach using different learning techniques are compared. The possible role of such a hybrid solution in an intelligent manufacturing environment is investigated.
View
View
View
View
Predicting Earthquake Ground Motion Descriptions Through Artificial Neural Networks for Testing the Constructions
Chapter
  • Dec 2001
The purpose of this chapter is to determine whether the trained with actual recorded earthquake accelerograms neural network is capable of generating reasonable looking accelerograms from design spectra. The chapter presents results from a number of investigations into the problem of the predicting ground motions using an artificial neural network for testing the constructions. The proposed method for generating spectrum compatible accelerograms uses the learning capabilities of neural networks to develop the knowledge of the inverse mapping from the response spectra to earthquake accelerogram. The chapter discusses a two-stage approach. In the first stage, a replicator neural network is used as a data compression tool to compress the vector of the discrete Fourier spectrum of the earthquake accelerogram to a vector of much smaller dimension. In the second stage, a multilayer feed forward neural network learns to relate the response spectrum to the compressed Fourier spectrum. As a result of combining these two neural networks, the artificial earthquake accelerograms are directly generated from the given response spectra. The number of attributes, which allows receiving the artificial earthquake accelerogram, can be changed in such a manner so as to receive artificial generation of the accelerogram with expected characteristics. The received results are used for testing different parts of constructions.
View
View
Spatial attenuation of intensities for central U.S. earthquakes
Article
  • Jun 1976
Attenuation of ground motion in the central United States has to be determined principally using the Modified Mercalli (MM) intensity observations because of the absence of instrumental strong ground-motion data. Nuttli's previous studies of Mississippi Valley earthquakes indicate that higher-mode surface waves produce the largest ground motion except possibly in the near-field region. Particle velocity rather than acceleration correlates directly with intensity and the coefficient of anelastic attenuation has an average value of 0.10 per degree. Using data from isoseismals of the November 9, 1968, southern Illinois and the December 16, 1811, New Madrid, Missouri earthquakes and assuming a linear relationship between log(A/T) and MM intensity, attenuation is expressed by the equation, valid for I(R) ≧IV (MM), I ( R ) = I 0 + 3.7 − 0.0011 R − 2.7 log ⁡ R ; for R ≧ 20 k m where R is the epicentral distance in kilometers. This relationship shows fairly good agreement with isoseismals of many large earthquakes in the central United States and may therefore be useful in providing realistic estimates of spatial attenuation and hence of design earthquakes for a given site. It can also be sometimes useful in estimating the epicentral intensity of an earthquake whose maximum intensity is not reliably known.
View