Estimating wave spectra from the motions of dynamically positioned vessels: An assessment based on model tests

Autor(es)

Bisinotto, Gustavo A. ; de Mello, Pedro C. ; Queiroz Filho, Asdrubal N. ; Ianagui, Andre S. S. ; Simos, Alexandre N. ; Tannuri, Eduardo A.

• Tipo de documento: Artigo Científico
• Fonte: APPLIED OCEAN RESEARCH; v. 121, p. 14-pg., 2022-03-01.

Resumo

Alternatives for estimating the directional wave spectrum from the motions of a vessel have been evaluated over the past years considering different statistical approaches. This paper discusses the application of the motions of a dynamically positioned platform supply vessel to perform inferences by means of a Bayesian method. The method was formulated considering a methodology for hyperparameter calibration based on an optimization process, and the definition of an iterative estimation scheme. Small-scale seakeeping tests in irregular waves were carried out with the model of a platform supply vessel (PSV) equipped with a dynamic positioning (DP) system, which was calibrated to be consistent with a typical real system. For evaluating any possible influence the DP responses might have had on the motion-based wave estimations, the same test conditions were also performed with the model moored by means of an equivalent soft-mooring configuration. Estimation results were obtained from the motions measured in the experiments, showing that wave parameters could be computed with good precision in both arrangements, with overall average errors, in moored and DP configurations, respectively, of: 5.54% and 5.47% for significant height, 0.76% and 0.75% for peak period, and 1.25 and 1.46 for direction. A further analysis with the comparison of the estimated wave spectra attested the correspondence with the expected values and the similarities between the estimations obtained in the two different scenarios. (AU)

• Processo FAPESP: 21/00409-2 - Desenvolvimento de um sistema de monitoramento ambiental a partir de medições on-board de movimentos de embarcação com técnicas de aprendizado de máquina
• Beneficiário: Gustavo Alencar Bisinotto
• Modalidade de apoio: Bolsas no Brasil - Doutorado Direto