Structural damage detection based on PCA of vibration measurements

This paper deals with the application of statistical process control techniques based on principal component analysis to vibration-based damage diagnosis of structures.
Principal component analysis of the sensor time-responses allows to extract principal directions (i.e. features) which define a subspace that is representative of the dynamics of the instrumented structure. Any change in the response of a single sensor will affect the subspace spanned by the complete sensor response set. It follows that the subspace corresponding to the current state of the structure can be compared to the subspace of the initial state of the structure, assumed to be healthy, in order to diagnose possible damage. Principal component analysis may also be performed for every potential subset of damaged sensors in order to identify the involved sensor, and, therefore, the damaged substructure. In this paper, the problem of structural damage detection is addressed in the case of environmental vibration testing. A direct application is presented for a test item submitted to random vibration testing by means of an electro-dynamic shaker.