Method for synthesizing seismic motion field by using support vector machine, principal component analysis and particle swarm optimization in machine learning

Abstract

The invention relates to a method for synthesizing a seismic motion field by applying a support vector machine, principal component analysis and a particle swarm algorithm in machine learning. The method comprises the following steps: step 1, extracting a seismic oscillation mother wave from existing domestic seismic oscillation data by applying a principal component analysis method; step 2, establishing a ground motion prediction model by applying a support vector machine algorithm, and verifying the correctness of the model; and step 3, matching the reaction spectrum Sa (T) of the synthesized seismic oscillation with the reaction spectrum Sa * (T) obtained through the prediction equation, i.e., when the value of S is minimum, obtaining a group of coefficients ki, i.e., the solution of the synthesized seismic oscillation. A machine learning theory is introduced into a synthetic seismic oscillation method, the original seismic oscillation information can be reserved, the problem that some important seismic oscillation information can be ignored when a random seismic oscillation simulation method is applied in the past is avoided, the method has good inclusivity to measured data, seismic giant disaster situation scene reproduction can be well achieved, and the method has a good application prospect.

Description

technical field [0001] The invention relates to a method for synthesizing an earthquake field using a support vector machine, a principal component analysis and a particle swarm algorithm in machine learning, and belongs to the field of earthquake risk analysis. Background technique [0002] Seismic Hazard Analysis (PSHA) has always been one of the topics of earthquake resistance research. The first step in seismic hazard analysis is the scene reconstruction of earthquake disasters. However, the lack of destructive earthquake records has always restricted the development of seismic hazard analysis. From an academic point of view and a social security point of view, the uncertainty of the source, path, and site of a large earthquake has always been the key to simulating the reproduction of a large earthquake. With the acceleration of urbanization, large-scale urban development extending to earthquake-prone areas and the aging of infrastructure are exacerbating the seismic haz...

AI Technical Summary

Problems solved by technology

Achieve more retention of the original ground motion information, and solve the problem that some important ground motion information may be ignored by random ground motion simulation methods

Images