Studies of system identification (SI)-based structural health monitoring (SHM) are being
actively conducted to ensure structural safety. Recently, many SI techniques have been
developed using the output-only SI paradigm for estimating the modal parameters. The
features of these output-only SI methods are obtained using frequency domain decomposition
(FDD) and stochastic subspace identification (SSI), both of which involve the use of
algorithms based on orthogonal decomposition, such as singular value decomposition (SVD).
However, the SVD leads to a high level of computational complexity to estimate modal
parameters. This paper proposes a technique to estimate the mode shape at lower
computational cost. This technique shows pseudo-modal responses (PMR) through a
bandpass filter and suggests a time-history modal assurance criterion (MAC). Finally, the
mode shape is estimated from the PMR and the time history MAC. Experimental tests of the
vibration measurement were performed, and the results of mode shape and computation time
between a representative SI method and the proposed method were compared.
INTRODUCTION
In a variety of fields, including civil and architectural engineering as well as mechanical
and aerospace engineering, structural health monitoring (SHM) and damage detection
techniques have been actively developed over the past several decades for evaluation of the
safety of structures[1-4]. The process of estimating structural modal parameters is referred to
as system identification (SI)[5]. Recently, SI techniques have been rapidly developed based
on the output-only SI paradigm[6-8]. The modal parameters from SI methods, such as natural
frequency, mode shape and modal damping ratio, present structural dynamic properties that
condensing the time history vibration measurement data.
The features of these SI methods are obtained using frequency domain decomposition
(FDD)[9] and stochastic subspace identification (SSI)[10]; both of these approaches involve
the use of algorithms based on orthogonal decomposition, such as singular value
decomposition (SVD), because the applications of the algorithms have been validated in
various fields[11,12]. Therefore, orthogonal decomposition algorithms are used in one part of
the SI methods to identify the modal parameters of structures from vibration measurements