Support Vector Machine Method for Developing Ground Motion Models for Earthquakes in Western Part of China
Jinjun Hu, Hui Zhang
Abstract
This article develops ground-motion models (GMMs) by using a support vector regression (SVR) for western part of China. These SVR GMMs constitute mathematical equations that can predict the peak ground acceleration (PGA) and response spectral acceleration at different periods. Given the earthquake magnitude (Mw), hypocentral distance (Rhypo) and shear-wave velocity averaged over the top 30 m of soil (Vs30), predicted values can be acquired. The results demonstrate that the proposed GMMs are reasonable and can produce accurate estimations with a good generalization capability for ground motions exhibiting a range of seismic characteristics similar to those considered in the database.
Topics & Concepts
Ground motionPeak ground accelerationAccelerationRange (aeronautics)Support vector machineSpectral accelerationShear (geology)GeneralizationGeologySeismologyEarthquake engineeringComputer scienceMathematicsPhysicsEngineeringArtificial intelligenceMathematical analysisClassical mechanicsAerospace engineeringPetrologySeismic Performance and AnalysisStructural Health Monitoring TechniquesMasonry and Concrete Structural Analysis