Second generation wavelet-based modal parameters identification based on through operational measurements

In many applications, the structural operating conditions may significantly differ from those applied during laboratory tests where the structure is well known, well installed and properly excited. For structures under their natural loading conditions, or excited by random forces, excitations cannot be measured and are usually non stationary. In the dynamic identification of engineering structures, the input signals are unknown and the modal parameter identification procedure is therefore based on output-only measurement signals. An operational modal analysis is a useful complement to the more traditional modal analysis approach where both input and output are measured. The wavelet transforms analysis in estimating the modal damping ratios have been investigated, but some drawbacks have not been resolved. Second generation wavelet transform ( SGWT) is a kind of time-domain transform method based on lifting scheme. SGWT is different from the classical wavelet transform not relying on Fourier transform, doing all calculation on time-domain, and constructing special property wavelet by designing predictor and updater. In another words, wavelet basis function with some special characteristic can be obtained by means of designing prediction and update coefficients, which will be useful for constructing different wavelet function for different civil engineering structures in parameters identification. The aim of this paper is to present the application of a new identification procedure, named the second generation wavelet identification technique, for structural characterization. The ambient vibration measurements are converted to free decay responses using the random decrement technique. Then the wavelet identification technique is used to identify the dynamic characteristics of the structural system in terms of natural frequencies, damping coefficients and modal shape. An experimental example is provided to prove the accuracy of this method. The results obtained agree with those previously obtained on the dynamic behaviors by other method.