CEE 200: Structural & Geotechnical Engineering Seminar
UCLA Civil & Environmental Engineering Department

Data-driven structural seismic response reconstruction framework: a machine learning perspective

Han Sun
Ph.D. Candidate
University of California, Los Angeles

Strucutral seismic resilience society has been grown rapidly in the past three decades. Extensive probabilistic techniques have been developed to address uncertainties from ground motions and building systems to reduce structural damage, economic loss and social impact of buildings subjected to seismic hazards where seismic structural responses are essential and often are retrieved through Nonlinear Response History Analysis. This process is largely limited by availability of model, ground motion time history and computational effort. An alternative data-driven framework is proposed to reconstruct structure responses through machine learning techniques from limited available sources which may potentially benefit for “real-time” interpolating monitoring data to enable rapid damage assessment and reducing computational demands for regional seismic hazard assessment. It also provides statistical insight to understand uncertainties of seismic building responses from both structural and earthquake engineering perspective.

Bio: Han is a Ph.D. candidate in the Department of Civil and Environmental Engineering working with Prof. John W. Wallace and Prof. Henry V. Burton. He is also an M.S candidate in the Department of Statistics advised by Prof. Amini A. Arash, University of California, Los Angeles. He held an M.S. in infrastructure system of civil engineering department from University of Michigan, Ann Arbor and B.S. in structural engineering from Hong Kong Polytechnic University. His research follows into seismic structural resilience category and focuses on machine learning and statistical modeling application in earthquake engineering field.

Soil-Foundation-Structure Interaction Effects on Cyclic Failure Potential of Silts and Clays

Jason Buenker
Ph.D. Candidate
University of California, Los Angeles

Earthquake-induced ground failures have generated billions of dollars in economic losses during recent earthquakes.  Ground failure can broadly be categorized as exhibiting “sand-like” (e.g. liquefaction) or “clay-like” (e.g. cyclic softening) behavior.  Many available liquefaction case histories have enabled engineers to develop procedures for predicting and mitigating liquefaction.  In contrast, mechanisms of cyclic softening are poorly understood.  Further, in liquefaction and cyclic softening evaluations, current practice solely considers free-field stress conditions and neglects the influence of soil-foundation-structure interaction (SFSI).  Shallow foundation failures in clay soils during the 1999 Chi-Chi and 1999 Kocaeli Earthquakes reflect the need for cyclic softening prediction methodologies that consider SFSI effects.  This research will seek to fill this knowledge gap by developing an understanding of the influence of SFSI on cyclic ground failure potential.

Bio: Jason is a Ph.D. candidate in the Department of Civil and Environmental Engineering working as a research assistant under Prof. Scott Brandenberg and Prof. Jon Stewart.  His research to date has consisted of design, construction, and testing of a centrifuge soil model at the Center for Geotechnical Modeling at UC-Davis.  Prior to attending UCLA, he earned BS and MS degrees at the University of Illinois Urbana-Champaign and worked in geotechnical consulting.

Where:       Haines Hall 220
When:        4 – 6 PM, Tuesday, May 1