Statistical Structural Health Monitoring in the Presence of Environmental Variability and Uncertainty
Abstract
This project will develop cutting-edge statistical methods to detect damage in the face of huge data sources across broad environmental variability. To tackle the new challenges faced by Structural Health Monitoring (SHM) practitioners, this project will incorporate and bridge two communities of researchers -- statisticians and aerospace, civil and mechanical engineers. It will use Bayesian Vector Autoregressive (BVAR) models Ð a tool developed and used in econometrics, to model SHM systems It will exploit graphical models to exploit sparsity in the system to increase computational efficiency and model estimability in the face of massive data sources. it will develop new methods for specifying this sparsity, including automated methods, expert opinion, and combinations of the two in order to take advantage of knowledge of the structure under study. It will investigate how to incorporating a mechanism for switching between environmental states while maintaining computational feasibility and tractability
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 12, 2017
- Source ID
- W911NF1510172
Entities
People
- Luke Bornn
Organizations
- Army Contracting Command
- Harvard University
- United States Army