Multivariate statistics process control for dimensionality reduction in structural assessment
(2008) Mechanical Systems and Signal Processing — Vol. 22, n° 1, p. 155-171 (2008)
Files
pdfdocument.pdf
Restricted Access - Adobe PDF
- 907.23 KB
Details
- Authors
- Abstract
- This paper presents advantages of using techniques like principal component analysis (PCA), partial least square (PLS) and some extensions called multiway PCA (MPCA) and multiway PLS (MPLS) for reducing dimensionality in damage identification problem, in particular, detecting and locating impacts in a part of a commercial aircraft wing flap. It is shown that applying MPCA and MPLS is convenient in systems which many sensors are monitoring the structures, because the reciprocal relation between signals is considered. The methodology used for detecting and locating the impact uses the philosophy of case-based reasoning, where single PCA and PLS are used also for organizing previous knowledge in memory. Sixteen approaches combining those techniques have been performed. Results from all of them are presented, compared and discussed. (c) 2007 Elsevier Ltd. All rights reserved.
- Affiliations
Citations
Mujica, L. E., Vehi, J., Ruiz, M., Verleysen, M., Staszewski, W., & Worden, K. (2008). Multivariate statistics process control for dimensionality reduction in structural assessment. Mechanical Systems and Signal Processing, 22(1), 155-171. https://doi.org/10.1016/j.ymssp.2007.05.001 (Original work published 2008)