Frequency Variation and Sensor Contribution Assessment: Application to an Offshore Platform in the South China Sea

Fushun Liu,Huajun Li,Weiying Wang,Wei Li,Bin Wang
DOI: https://doi.org/10.1016/j.jsv.2014.10.041
IF: 4.7
2015-01-01
Journal of Sound and Vibration
Abstract:In this paper, a time–frequency and multiple-sensor assessment method is presented and then applied to an offshore platform in the South China Sea with the goal of providing a more suitable time duration of measured signals and evaluating each sensor׳s contribution to mode shapes of interest. By processing all measured signals simultaneously, a series of linear parameters are used to fit the measured signals. A moving window in overlapping steps along the time record of a non-stationary signal is used for time–frequency analysis while a series of amplitude matrices are obtained for all sliced segments. These segments are then used to evaluate each sensor׳s contribution to some mode(s) of interest. Compared with the short-time Fourier transform (STFT) or S-transform method, the frequencies used in this time–frequency analysis are shared by the sensors involved so that the variations in the results due to the characteristics of different sensors are avoided. In addition, the approach requires only a short-duration segment to obtain high-frequency resolution, which will improve the computing efficiency of modal analysis using measured seat est data. The proposed method could also be used to estimate damping ratios and each sensor׳s contribution to modes of interest based on the analysis of the series of amplitude matrices; this could be used to guide the installation of sensors in field tests of offshore structures. To demonstrate the proposed method with a time–frequency analysis, a numerical example of a synthesized signal with five segments, each with its owns different frequency componentsincluding a relatively weaker component, is constructed; numerical results from the analysis of this example signal indicate that the approach could yield a sharper image with a good computing efficiency. The second example simulates three signals that represent multiple measurements; this example is used to study each sensor׳s contribution to different modes based on an amplitude matrix. Finally, measurements from a four-leg jacket-type offshore platform located in the Beibu Gulf in China are used to study the proposed method. It can be concluded that sea-test measurements of offshore platforms show frequency-varying characteristics, and the proposed method could provide a more suitable time duration of signals for modal analysis. The assembled 24 sensors are shown to have different contributions to frequencies at 1.0375Hz, and the amplitude of each sensor will not be a constant when different segments of a measurement are used for analysis. However, each sensor׳s contribution to the assigned modes can be assessed based on average values. For example, the contribution of sensor 3 is estimated to be 12.5362 per cent, sensor 15׳s contribution is estimated to be 7.1803 per cent. Each sensor׳s importance to the mode of interest could then be evaluated.
What problem does this paper attempt to address?