26 Unsupervised Neural Networks for Site Parameters Characterization of Recent Iran Earthquakes Available to Purchase

Artificial neural networks have been used for determination of site parameters in the region lacking site investigation reports. This method benefits advantages of unsupervised neural network in classification problems. Using this method the ranges of shear wave velocity for determination of each soil type may be estimated for a large data base. This site classification has been performed using linear response spectra of the earthquake records. The local site conditions are an important factor in the recorded waveform of earthquake ground motions. This also becomes important in ground motion analysis and in earthquake resistant designs. Different site conditions can induce amplifications of different period ranges in the response spectra. The method has also been applied to 2003 Bam Earthquake in south of Iran. The effects of earthquake magnitude and duration on the classification results have also been discussed.