Identifying Flexural-Mode Shapes of Ice Floes Under Wave Actions Using Multivariate Analysis

This paper analyzes experimental data collected during the HYDRALAB+ project: Loads on Structure and Waves in Ice (LS-WICE). Several data analysis techniques, including Fast Fourier Transform, Upward Zero-Crossing, Genetic algorithm, and Dynamic Mode Decomposition (DMD) are applied. The results show that the wavelength under ice in the analyzed data does not deviate from the open water wavelength. More importantly, the authors show that ice floes bend with the same frequency as the incoming waves. In addition, the authors identify the dominant flexural-mode shapes and the associated nonlinearities. Multivariate analysis techniques, such as Proper Orthogonal Decomposition (POD) and Smooth Orthogonal Decomposition (SOD) methods are applied to the spatial and temporal responses of ice floes. The results show that modes based on the POD and SOD methods have similar spatial and temporal shapes. Applying both methods reveals that most of the motion energy is captured in a two-dimensional subspace of the phase space, where the sum of the first two modes correspond to more than 95% of the total system response. Lastly, the results of the multivariate analysis indicate that weak nonlinearity exists in the flexural motions of ice floes induced by waves.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01745205
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jul 8 2020 2:35PM