Bayesian Estimation of Wave Spectra – Proper Formulation of ABIC

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

It is possible to estimate on-site wave spectra using measured ship responses applied to Bayesian Modelling based on two prior information: the wave spectrum must be smooth both directional-wise and frequency-wise. This paper introduces two hyperparameters into Bayesian Modelling and, hence, a proper formulation of ABIC (a Bayesian Information Criterion) is given, in contrast to the improper formulation given of ABIC when only one hyperparameter is included. From a numerical example, the paper illustrates that the optimum pair of hyperparameters, determined by use of ABIC, corresponds to the best estimate of the wave spectrum. In addition, full-scale motion measurements are analysed and the estimated wave spectra are compared with estimations made by wave radar measurements and satellite measurements. The agreement is reasonable.
Original languageEnglish
Title of host publication10th International Symposium on Practical Design of Ships and Other Floating Structures
VolumeVolume I
Publication date2007
Pages284-292
Publication statusPublished - 2007
Event10th International Symposium on Practical Design of Ships and Other Floating Structures - Houston, TX, United States
Duration: 1 Oct 20075 Oct 2007
Conference number: 10

Conference

Conference10th International Symposium on Practical Design of Ships and Other Floating Structures
Number10
CountryUnited States
CityHouston, TX
Period01/10/200705/10/2007

Fingerprint Dive into the research topics of 'Bayesian Estimation of Wave Spectra – Proper Formulation of ABIC'. Together they form a unique fingerprint.

Cite this