Incident component extraction from disturbed waves around large fixed cylindrical structures

Abstract

The parameters of incident waves are critical for real-time wave load estimation of structures in service. Nonetheless, it is challenging to characterize incident waves accurately using the measured wave surface elevation around large fixed cylindrical structures due to the interaction with the structure in the wave field. To provide a better understanding of incident waves, which are usually buried in directly measured waves, a new time-domain method for the extraction of first-order and second-order incident waves around large fixed cylindrical structures is proposed. In contrast to most existing separation methods that are suitable for structures with equal reflection coefficients, the amplitude and phase changes of near-field waves around cylindrical structures can be determined by considering the significant diffraction effect, and then the time-frequency characteristic of the wavelet transform is employed, which enables the extraction of incident waves in the time domain. The accuracy of the proposed method is studied using several examples with known incident waves which are generated with the OpenFOAM. The numerical results show that the deviations between the exact and extracted incident waves change from 6.16% to 16.77% for different wave conditions. To further investigate the performance of the proposed method, an experimental study on waves around a mono-pile offshore wind turbine (OWT) is conducted in the laboratory of the Ocean University of China. The predicted results basically agree well with the target waves in terms of amplitude and phase. The deviations between predicted waves using the proposed method and target waves are 110% smaller than those between directly experimental measured waves and target waves for all tested conditions. Finally, 48 h of measured wave data were obtained during calm and typhoon periods around a mono-pile OWT located near Rudong County, Jiangsu Province, in the Yellow Sea of China. There are almost 150% and 30% differences between the extraction results and measured data in the time series and statistical wave heights, respectively, which means that employing disturbed wave data as the input for calculating real-time wave loads leads to deviations that cannot be ignored.