Project:	Earthquake Time-Series Analysis and Synthesis using Machine Learning
Disciplines:	Civil Engineering, Data Science, Geophysics, Computer Science
Mentor:	Domniki Asimaki, Professor of Mechanical and Civil Engineering, (EAS), domniki@caltech.edu, Phone: 6263952742
Mentor URL:	https://www.asimaki.caltech.edu  (opens in new window)
Background:	Seismic hazard is quantified using ground motion measurements from past earthquakes. These measurements are used to develop statistical models that engineers and earth scientists then use to estimate ground motion shaking of future earthquakes. Recently, earth scientists are running simulations of future large events (i.e. The Big One) that are not included in the recorded databases. These simulations, however, are limited by the coarseness of their input parameters. In this project, we are using machine learning to fuse simulations with observations.
Description:	The student will be developing machine learning algorithms to extract high frequency features from observed ground motion signals and fusing them with simulated long period motions. Alternatively, the student will be developing machine learning algorithms to translate response spectra into Fourier spectra, intended for engineering applications of the fused ground motions in engineering practice; or developing algorithms to signal process raw observed ground motions into processed ground motions for engineering applications.
References:	https://arxiv.org/abs/2309.03447 https://openreview.net/pdf?id=j3oQF9coJd
Student Requirements:	Signal processing Machine learning Python
Programs:	This AO can be done under the following programs:   Program    Available To        SURF    Caltech students only  Click on a program name for program info and application requirements.