Comparisons of Directional Wave Analysis Methodsby Taerim Kim, Univ of Florida, Gainesville, United States,
Lihwa Lin, Univ of Florida, Gainesville, United States,
Hsiang Wang, Univ of Florida, Gainesville, United States,
Abstract: Four methods analysing directional wave spectrum were testes and compared through numerical simulation. The four methods are (i) the truncated Fourier series (TFS), (ii) the Longuet-Higgins parametric model (LHM), (iii) the maximum entropy method developed based on the autoregressive process of directional waves (MEM I) and (iv) the maximum entropy method developed based on a probability density distribution of directional waves (MEM II). These methods were all tested using the first five frequency-dependent Fourier coefficients from the Fourier series expansion of a directional spectrum. The result showed that the MEM II appears to have the better overall performance. However, the MEM II also often has convergence problem when dealing with real data. For application, this convergence problem can be overcome by using a maximum entropy approximation method suggested here.
Subject Headings: Entropy methods | Numerical models | Numerical methods | Fourier analysis | Water waves | Wave spectrum | Numerical analysis | Parameters (statistics)
Services: Buy this book/Buy this article
Return to search