Research Papers

Structural Health Monitoring With Autoregressive Support Vector Machines

[+] Author and Article Information
Luke Bornn

CCS-6, Statistical Sciences Group, Los Alamos National Laboratory, MS F600, Los Alamos, NM 87545

Charles R. Farrar1

The Engineering Institute, Los Alamos National Laboratory, MS T006, Los Alamos, NM 87545farrar@lanl.gov

Gyuhae Park, Kevin Farinholt

The Engineering Institute, Los Alamos National Laboratory, MS T006, Los Alamos, NM 87545


Corresponding author.

J. Vib. Acoust 131(2), 021004 (Feb 17, 2009) (9 pages) doi:10.1115/1.3025827 History: Received May 14, 2008; Revised September 12, 2008; Published February 17, 2009

The use of statistical methods for anomaly detection has become of interest to researchers in many subject areas. Structural health monitoring in particular has benefited from the versatility of statistical damage-detection techniques. We propose modeling structural vibration sensor output data using nonlinear time-series models. We demonstrate the improved performance of these models over currently used linear models. Whereas existing methods typically use a single sensor’s output for damage detection, we create a combined sensor analysis to maximize the efficiency of damage detection. From this combined analysis we may also identify the individual sensors that are most influenced by structural damage.

Copyright © 2009 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Illustration of linear support vector regression fit

Grahic Jump Location
Figure 2

Illustration of mapping to an alternate space to induce linearity

Grahic Jump Location
Figure 3

Raw simulated data with highlighted artificial damage

Grahic Jump Location
Figure 4

Autocorrelation plot of simulated data

Grahic Jump Location
Figure 5

SVM (top) and linear AR models (bottom) fit to subset of data

Grahic Jump Location
Figure 6

Residuals from SVM (top) and linear AR models (bottom) applied to simulated data. The 99% control lines based on the residuals from the undamaged portion of the signal are shown with the horizontal lines.

Grahic Jump Location
Figure 7

Diagram of the experimental structure

Grahic Jump Location
Figure 8

Q-Q plot of residuals from SVM model

Grahic Jump Location
Figure 9

Residuals from four sensors for t=7000,…,9384. The horizontal lines are the 99% control lines.

Grahic Jump Location
Figure 10

Density estimate of combined residual (black) versus chi-squared distribution (dashed)

Grahic Jump Location
Figure 11

Combined residuals from all four sensors. The 99% control line is shown as the dashed horizontal line. Sliding window damage indicators are indicated by the boxes across the top of the plot.



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In