Main Article Content

Authors

Ambient vibration studies have proven to be effective for system identification in buildings. Traditionally, the fast Fourier transform has been used to create spectrograms and analyze the time-frequency distribution of a signal, but new processing tools have advantages compared with this method. The spectrogram (based on the fast Fourier transform), the Choi-Williams transform, and the pseudo-frequency distribution of the wavelets transform are compared by applying them to the acceleration response of a building subject to ambient excitation. The results for the three methods are similar, and their application is considered valid. Notable differences are observed in the resolution of the images and the processing time. The use of wavelets and the Choi-Williams transform yields results that have better resolution, making it easier to identify closely spaced vibration modes. The Fourier transform is much faster than the other two methods, this feature being desirable when processing large quantities of data.

Helene Tischer

escuela de ingenieria civil y geomatica,universidad del valle ,cali colombia

Peter Thomson

escuela de ingenieria civil y geomatica,universidad del valle ,cali colombia

Johannio Marulanda A.

escuela de ingenieria civil y geomatica,universidad del valle ,cali colombia
1.
Tischer H, Thomson P, Marulanda A. J. comparision of there transforms for time-frequency distributions from their application to ambient vibration records. inycomp [Internet]. 2007 Jun. 8 [cited 2024 Nov. 22];9(2):21-32. Available from: https://revistaingenieria.univalle.edu.co/index.php/ingenieria_y_competitividad/article/view/2484
1 2 > >>