iterative data enhancement of ambient vibration tests for the nitrate-prilling tower in shiraz petrochemical complex

In this study, the discrete wavelet packet transform (dwpt) has been used for the single-step and iterative denoising methods for enhancing data with a high level of noise to identification of modal frequencies in ambient vibration tests on a petrochemical process tower in shiraz, iran. the ambient vibration test is performed by the wind load. all mechanical systems operated during the test; hence, different noise sources exist. here, both high and low frequency ranges are decomposed effectively in the dwpt, and it provides a lot of global and localized information. the dwpt-based one-step denoising method fails to properly denoise the high-level noisy data with denoising thresholds obtained by different theoretical methods. for this reason, the so-called peeling approach achieved by an iterative denoising method is used to enhance the quality of the signal. for this iterative method, the parameters are obtained by the trial-and-error method‌. after the signal-enhancement stage, the signal processing step is performed by continuous wavelet transforms (cwts) to detect the time-frequency information in the data. furthermore, the modal frequencies are directly identified by the cross wavelet transform (xwt) and the corresponding spectral power density. finally, the estimated frequencies by xwt are compared with the natural frequencies of a damaged model simulated by the finite element (fe) method.