A PCA-BASED ALGORITHM FOR STRUCTURAL HEALTH MONITORING USING FREQUENCY RESPONSE FUNCTIONS

ZHU Junhua;YU Ling;

Journal of Vibration and Shock ›› 2011, Vol. 30 ›› Issue (5) : 111-115.

PDF(1774 KB)
PDF(1774 KB)
Journal of Vibration and Shock ›› 2011, Vol. 30 ›› Issue (5) : 111-115.
论文

A PCA-BASED ALGORITHM FOR STRUCTURAL HEALTH MONITORING USING FREQUENCY RESPONSE FUNCTIONS

  • ZHU Junhua1,2; YU Ling1,2
Author information +
History +

Abstract

Abstract: Based on measured Frequency Response Functions (FRFs), an easier and more efficient method for structural health monitoring is proposed by using Principle Component Analysis (PCA) in this paper. The FRFs of the healthy and damaged structure are used as initial data. The method uses a PCA transform technique to obtain the features of intact structure, also called the Principle Components (PCs), in which an orthogonal transformation matrix packed by the first few eigenvectors of covariance matrix can be found. Further, the orthogonal transformation matrix is applied to the FRFs of damage structures so as to find the features of the corresponding damage state of structures. Both structural damage detection and structural health monitoring can be achieved by comparing the two dimensional PCs distribution charts respectively corresponding to the damaged and to the healthy state of structures. Two numerical simulation examples show that the proposed method is correct, effective and feasible. Because it is based purely on the analysis of the vibration responses of structures, which makes it quite easy to perform and demonstrates its adaptability to structural health monitoring in field.

Key words

principle component analysis / frequency response function / damage detection / structural health monitoring

Cite this article

Download Citations
ZHU Junhua;YU Ling;. A PCA-BASED ALGORITHM FOR STRUCTURAL HEALTH MONITORING USING FREQUENCY RESPONSE FUNCTIONS[J]. Journal of Vibration and Shock, 2011, 30(5): 111-115
PDF(1774 KB)

Accesses

Citation

Detail

Sections
Recommended

/