Prediction of Seismic Intensity Distributions Using Neural Networks

TL;DR

This paper introduces a neural network-based hybrid model that predicts seismic intensity distributions, including abnormal cases influenced by underground structures, by treating the data as 2D images for improved accuracy.

Contribution

It presents a novel hybrid regression-classification neural network approach that effectively models complex, abnormal seismic intensity distributions as 2D data.

Findings

Accurately predicts abnormal seismic distributions

Treats seismic data as 2D images for modeling

Outperforms traditional ground motion prediction methods

Abstract

The ground motion prediction equation is commonly used to predict the seismic intensity distribution. However, it is not easy to apply this method to seismic distributions affected by underground plate structures, which are commonly known as abnormal seismic distributions. This study proposes a hybrid of regression and classification approaches using neural networks. The proposed model treats the distributions as 2-dimensional data like an image. Our method can accurately predict seismic intensity distributions, even abnormal distributions.