Feature extraction and classi cation techniques for health monitoring of structures


1 Departments of Civil, Biomedical, and Electromechanical Engineering, Faculty of Engineering, Autonomous University of Queretaro, Campus San Juan del Rio, Moctezuma 249, Col. San Cayetano, 76807, San Juan del Rio, Queretaro, Mexico

2 Department of Civil, Environmental and Geodetic Engineering, The Ohio State University, 470 Hitchcock Hall, 2070 Neil Avenue, Columbus, OH 43220, U.S.A


Damage identi cation in Structural Health Monitoring (SHM) involves three main steps: signal acquisition, signal processing, and feature extraction and interpretation. Recently, the authors presented a review of recent articles on signal processing techniques for vibration-based SHM. This article presents a review of journal articles on feature extraction and classi cation techniques in order to assess the health condition of a structure in an automated manner. This review is limited to civil structures such as buildings and bridges. The methods reviewed are neural networks, wavelets, fuzzy logic, support vector machine, linear discriminant analysis, clustering algorithms, Bayesian classi ers, and hybrid methods. Further, two novel algorithms with potential for feature classi cation in SHM are suggested.