Linear and Nonlinear Cointegration for the removal of environmental trends from SHM data Seminar
- Time:
- 16:00
- Date:
- 1 March 2016
- Venue:
- 13/3017
For more information regarding this seminar, please email ISVR@Soton.ac.uk .
Event details
ISVR seminar
Before structural health monitoring (SHM) technologies can be reliably implemented on structures outside laboratory conditions, the problem of environmental variability in monitored features must first be addressed. Structures that are subjected to changing environmental or operational conditions will often exhibit inherently nonstationary dynamic and static responses which can mask any changes caused by the occurrence of damage.
In this seminar we will discuss the concept of cointegration, a tool for the analysis of nonstationary time series, as a promising new approach for dealing with the problem of environmental variation in monitored features. An extension to nonlinear cointegration will also be introduced which employs Gaussian Process regression under the Engle-Granger framework.
Speaker information
Elizabeth Cross, University of Sheffield. Dr Elizabeth Cross is a Senior Lecturer in the Dynamics Research Group in the Department of Mechanical Engineering. She was awarded a first class degree in mathematics from the University of Sheffield in 2007 and an MSc (Res) in Advanced Mechanical Engineering, also from the University of Sheffield, in 2008. In 2012 she completed her PhD on Structural Health Monitoring whilst also working as a research associate in her final year in collaboration with Messier-Bugatti-Dowty. Elizabeth’s main research interests are in the field of Structural Health Monitoring (SHM), specifically vibration based SHM, which uses monitored dynamic properties of a structure for condition assessment and damage detection. SHM is still a relatively young field and so much of the research that goes on is confined to the laboratory. While it is true that research into SHM is becoming increasingly popular, it has failed, so far, to be taken up in any major way by industry, despite the obvious economic and safety benefits it could offer. Elizabeth’s current research is broadly concerned with how SHM can be made to work for the real world and encompasses the application of statistics and machine learning technology, as well as mathematics from other disciplines such as econometrics.