Feedback

Faculté des Sciences appliquées
Faculté des Sciences appliquées
Mémoire
VIEW 50 | DOWNLOAD 146

Cruise ship's vibration behavior: simulation-measurement correlation analysis

Télécharger
Salazar Loor, Lissette Priscilla ULiège
Promoteur(s) : Lesourne, Hervé
Date de soutenance : 22-sep-2021 • URL permanente : http://hdl.handle.net/2268.2/13641
Détails
Titre : Cruise ship's vibration behavior: simulation-measurement correlation analysis
Auteur : Salazar Loor, Lissette Priscilla ULiège
Date de soutenance  : 22-sep-2021
Promoteur(s) : Lesourne, Hervé 
Membre(s) du jury : Ridley, Jonathan 
Langue : Anglais
Nombre de pages : 53
Mots-clés : [fr] Ship vibration
Discipline(s) : Ingénierie, informatique & technologie > Ingénierie mécanique
Public cible : Etudiants
Grand public
Institution(s) : Université de Liège, Liège, Belgique
Diplôme : Master : ingénieur civil mécanicien, à finalité spécialisée en "Advanced Ship Design"
Faculté : Mémoires de la Faculté des Sciences appliquées

Résumé

[fr] This work presents the analyses of different structures of a cruise ship to find the correlation between the vibrational behavior performed by finite element method simulations and measurements on-site. The structures include deck areas in the ship under construction and real scale models of passengers’ cabins. For the decks areas, there is a comparison of two different applications of the finite element method in Ansys: Classic and Workbench.
A modal analysis was performed on two decks’ sections in the area where passenger cabins will be installed. The differences between Ansys Classic and Workbench go from less than 1% to up to 20% depending on the closeness of the mesh size. Not all the mode shapes were easy to identify and not all of them visually match; the main differences between the models could be due to the different types of elements used to represent the structure’s behaviour and the distribution of those elements over the decks’ model. In the correlation with measurements from the test, there is not a clear tendency neither for Ansys Classic nor for Ansys Workbench to better fit with the measurements results, however, both show good results in terms of the range of frequency. Some parameters could be influencing the resultant frequencies and have not been considered in the models.
Then, the modal analysis of a prototype of a passenger’s cabin “Ramsess” is performed. At the time of the test, only the steel structure of the cabin was built, then, the model was formed by girders and pillars. The results of this analysis show a good correlation between the finite element model and the measurements, the maximum difference between the natural frequencies is 5%, and the mode shapes were easy to identify. It was possible to test two different support conditions to check the one which better represents the real structure, it was found that it depends on the main direction of the deformation.
Finally, the vibrational response of a cabin model is estimated by applying an external random excitation and compared with the finite element simulation. The model for finite element analysis was developed with a medium level of details, including the main steel structure, but excluding some non-structural parts. The results were compared in three points of the cabin’s balcony showing a good correlation in magnitude and a maximum difference in the peak’s frequencies of 16% for one of the points. In the other two points, the results do not correlate quite well probably due to an inaccurate representation of the model, not only by exclusion of some elements but also for the support condition.


Fichier(s)

Document(s)

File
Access M120 Master Thesis LPSL UPM-ICAM.pdf
Description:
Taille: 4.95 MB
Format: Adobe PDF

Auteur

  • Salazar Loor, Lissette Priscilla ULiège Université de Liège > Master ing. civ. méc. (EMSHIP+)

Promoteur(s)

Membre(s) du jury

  • Ridley, Jonathan
  • Nombre total de vues 50
  • Nombre total de téléchargements 146










Tous les documents disponibles sur MatheO sont protégés par le droit d'auteur et soumis aux règles habituelles de bon usage.
L'Université de Liège ne garantit pas la qualité scientifique de ces travaux d'étudiants ni l'exactitude de l'ensemble des informations qu'ils contiennent.