You are here: Home / Products / Earthquake Ground Motion Simulation Using Novel Machine Learning Tools
<< Return to the Knowledge Center Landing Page

Earthquake Ground Motion Simulation Using Novel Machine Learning Tools

Not available in print.

Available for free download at

By Arzhang Alimoradi. This report focuses on a novel three-phase method of model-independent probabilistic seismic hazard analysis (PSHA) and ground motion simulation, verified using previously recorded data and machine learning. Alimoradi introduces the concept of “eigenquakes,” representing characteristic earthquake records in a large database. They encapsulate the information that seismic stations collect over time. Comparison of the eigenquakes before and after events would indicate whether new events have contributed to our state of knowledge. Alimoradi emphasizes the benefits of using a model-independent method of PSHA and ground motion simulation, particularly in large urban areas where dense instrumentation is available or expected. Many convenient methods of data analysis can be borrowed from the field of machine learning. The problem of ground motion selection and scaling that has been subject of controversy in the past ten years is avoided in the proposed procedure. To exhibit its effectiveness, Alimoradi uses eight scenario examples for downtown areas of Los Angeles and San Francisco, where he shows that no dependency on specific ground motion prediction equations or processes of selection and scaling would be needed. Furthermore, principal component analysis allows systematic analysis of large databases of ground motion records that are otherwise very difficult to handle by conventional methods. Alimoradi worked under the supervision of Professor James Beck at the California Institute of Technology in Pasadena. May 2011, 179 pages.


EERI Members receive a 15% discount at checkout

Price: $0.00

Loading Updating cart...