polito.it
Politecnico di Torino (logo)

RECURSIVE ESTIMATION OF OCEAN WAVE SPECTRA

Daniel Andres Montoya Espinosa

RECURSIVE ESTIMATION OF OCEAN WAVE SPECTRA.

Rel. Giuseppe Giorgi. Politecnico di Torino, Corso di laurea magistrale in Mechatronic Engineering (Ingegneria Meccatronica), 2023

[img]
Preview
PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (9MB) | Preview
[img] Archive (ZIP) (Documenti_allegati) - Altro
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (4kB)
Abstract:

This investigation project consist in the emergence of a novel technique for continuous spectrum estimation based on a recursive approach in which the spectrum estimation will be continuously corrected implementing an Extended-Kalman Filter. In order to investigate a bond between the control theory and oceanography, enhancing and modernize the wave spectrum estimation. It was used the Extended-Kalman Filter (EKF) cyclical procedure, after had have analyzed the results. In this path was considered the stochastic behaviour of the ocean based on JONSWAP spectrum and also implementing a linearization to the model that allowed to obtain a narrow band spectra evaluating each frequency at the time on the basis of the contributing to reduce the complexity of the state-space model at the time, because it was used constant frequencies. It was opted chose the Least-Squares spectral approximation based on JONSWAP spectrum because it would give a formal solution that helped to model a State-space based of on a cascade of two second order filter and considering the basic principles of a stochastic signal processing in order to implement the robust control technique that allowed to correctly predict the JONSWAP spectrum. After setting the State-Space model for the EKF (Plan model P-M), another State-space model was prepared (Nominal Model N-M) using the same procedure as for the P-M in order to check the behaviour of the control technique when it was introduced not only similar conditions but also different. The estimation error obtained in each of the performed tests where significantly low, around of 0.094848, when analysing the P-M and N-M in equal conditions, reveling the accurate behaviour that the EKF presented. In the other hand the MSE from the analysis of P-M and N-M in different conditions exhibit a good performance. Also it was deducted that the JONSWAP shape parameter that generated a significant increase in the overall MSE, was the Hs (significant wave height). Finally in this investigation was observed that the most remarkable result was the performance of the EKF into the JONSWAP spectrum estimation,revealing that each of the obtained realization was desirable as realistic sea waves under this conditions evaluated.

Relatori: Giuseppe Giorgi
Anno accademico: 2022/23
Tipo di pubblicazione: Elettronica
Numero di pagine: 77
Soggetti:
Corso di laurea: Corso di laurea magistrale in Mechatronic Engineering (Ingegneria Meccatronica)
Classe di laurea: Nuovo ordinamento > Laurea magistrale > LM-25 - INGEGNERIA DELL'AUTOMAZIONE
Aziende collaboratrici: ET P IFP ENERGIES NOUVELLES
URI: http://webthesis.biblio.polito.it/id/eprint/26704
Modifica (riservato agli operatori) Modifica (riservato agli operatori)