Displacement controlled nonlinear finite element analyses are performed to determine the ultimate strength of three different vessel types in vertical bending. Parametric finite element models are proposed for a bulk carrier, double hull VLCC and for a container vessel. The influence of different model parameters on the collapse behavior is shown for sagging and hogging condition. The influence of initial imperfections of the stiffeners and the attached plating due to welding is taken into account. The results are validated for all different ships against Smith's method. An inhouse code has been developed following the Common Structural Rules (CSR) proposed by the International Association of Classification Societies (IACS). The Smith's method based results of intact and damaged ships in vertical bending have been used to train a Deep Neural Network (DNN) as machine learning approach. The applied network architecture is composed of two layers with a high-level number of activation units. The applicability of DNN to predict rapidly the ultimate strength of ships in vertical bending is demonstrated exemplarily for the same bulk carrier, double hull VLCC and the container vessel. Furthermore, DNN is used to determine the shift of the neutral axis for the different vessels.

Application of Different Methods to Determine the Ultimate Strength of Ships in Bending

Di Nardo Emanuel;
2023-01-01

Abstract

Displacement controlled nonlinear finite element analyses are performed to determine the ultimate strength of three different vessel types in vertical bending. Parametric finite element models are proposed for a bulk carrier, double hull VLCC and for a container vessel. The influence of different model parameters on the collapse behavior is shown for sagging and hogging condition. The influence of initial imperfections of the stiffeners and the attached plating due to welding is taken into account. The results are validated for all different ships against Smith's method. An inhouse code has been developed following the Common Structural Rules (CSR) proposed by the International Association of Classification Societies (IACS). The Smith's method based results of intact and damaged ships in vertical bending have been used to train a Deep Neural Network (DNN) as machine learning approach. The applied network architecture is composed of two layers with a high-level number of activation units. The applicability of DNN to predict rapidly the ultimate strength of ships in vertical bending is demonstrated exemplarily for the same bulk carrier, double hull VLCC and the container vessel. Furthermore, DNN is used to determine the shift of the neutral axis for the different vessels.
2023
978-0-7918-8684-7
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11367/131218
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 0
social impact