ABSTRACT
ABSTRACT: The paper is devoted on the problem of real-time earthquake prediction. Different kinds of stages of earthquake prediction are observed, and implementation of existing algorithms M8 and MSc for intermediate-term middle-range prediction for is discussed. An approach for real-time prognoses, based on neural network and vector quantization is suggesting.As input information for the neural network are given the parameters of recorded part of accelerogram, principle axis transformation and spectral characteristics of the wave. With the help of stochastic long-range dependence time series analyses and scene orientedmodel are determined the boundaries of destructive phase of strong motion acceleration. For selected diapason of transformed accelerograms is implemented one-dimensional and two-dimensional vector quantization. With self-organized map are determinedweight centers of selected classes. The prognoses are realizedwith the help of neural network, learned and trained to optimize selected target classes and determine probability density function.
