Peak ground acceleration prediction by fuzzy logic modeling for Iranian plateau

Babak Karimi Ghalehjough, Reza Mahinroosta

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)
163 Downloads (Pure)


In this study, fuzzy logic modeling is applied to a complex and nonlinear set of data to predict both horizontal and vertical peak ground accelerations in Iranian plateau. The data used for the model include an up-to-date seismic catalogue from earthquakes in Iran for prediction of both horizontal and vertical acceleration of a probable earthquake. Fuzzy logic toolbox on MATLAB program was used for modeling. Earthquake magnitude ranging from 4 to 7.4, source-to-site distance from 7 to 80 km and three different site conditions were considered: rock, stiff soil and soft soil. Results are compared with those from worldwide and regional attenuation relationships, which show the higher capability of the model in comparison with the other models. After training the model, testing of the fuzzy model with the remaining data set was performed to confirm the accuracy of the model. Changes in the peak ground accelerations in connection with changes in input parameters are studied which are in agreement with basic characteristics of earthquake input motions.
Original languageEnglish
Pages (from-to)75-89
Number of pages15
JournalActa Geophysica
Early online date16 Dec 2019
Publication statusPublished - 2020


Dive into the research topics of 'Peak ground acceleration prediction by fuzzy logic modeling for Iranian plateau'. Together they form a unique fingerprint.

Cite this