Cycle distributions for Gaussian waves - Exact and approximative results

Download:
Conference Proceeding/Paper | Published | English
Authors:
Lindgren, Georg ; Broberg, KB
Department:
Mathematical Statistics
Spatio-Temporal Stochastic Modelling Group
Research Group:
Spatio-Temporal Stochastic Modelling Group
Abstract:
Wave cycle distributions are notoriously difficult to calculate. Available analytic approximations, like those suggested by Longuet-Higgins (1983) and Cavanie, Arhan and Ezraty (1976) make use only of a few spectral moments, and are unreliable for moderate waves. To find distributions under general spectrum shape and parameters one has to use efficient numerical algorithms based on high-dimensional integrals. The paper presents a number of examples of wave cycle distributions computed by the software package WAFO developed at Lund Institute of Technology; http://www.maths.lth.se/matstat/wafo/. All results are compared to simulated or real data. The wave cycles studied are the max-min cycle, the crest-trough wave cycles, the rainflow cycle distribution, and the mean-separated max-min cycle.
Keywords:
wave period ; stationary Gaussian process ; density ; spectral ; random fatigue ; rainflow cycles ; crest-trough ; max-min
LUP-ID:
06912b03-149d-494c-aa91-b127cf61b0b5 | Link: https://lup.lub.lu.se/record/06912b03-149d-494c-aa91-b127cf61b0b5 | Statistics

Cite this