Please use this identifier to cite or link to this item: http://theses.ncl.ac.uk/jspui/handle/10443/5340
Title: Progressive Collapse of Ship Structures Under Cyclic Loading
Authors: Li, Shen
Issue Date: 2021
Publisher: Newcastle University
Abstract: Following ultimate limit state philosophy, the structural safety of ships and shiptype floating structures are assessed by ensuring an acceptable margin between their maximum load-carrying capacity and the extreme design load. This ultimate limit state approach is established assuming that the structures are subjected to a monotonic load that leads to an elastoplastic buckling collapse. However, the environmental loads of most marine structures are of a cyclic nature. The evaluation procedure and analysis methodology for ship structures under extreme loads with multiple cycles is currently lacking. Within this context, the aim of this research is to assess the collapse behaviour of ship structures, including plates, stiffened panels and ship hull girders, under combinations of cyclic loads and to investigate the influence of cyclic load on the ultimate strength of ship structures. Overall, four contributions have been achieved in this thesis. A parametric nonlinear finite element study is first performed on a range of ship plates under multiple cycles of compression and tension. The outcomes of this investigation provides a new recognition, for the first time, of the buckling collapse behaviours of unstiffened plates under cyclic compression and tension. In particular the characteristic features that are relevant for ultimate limit state assessment of ship hull structures are demonstrated, such as a progressively reducing but converging compressive strength and stiffness in the reloading regime of structural members under cyclic loads as compared to those under monotonic loads. Using observed response patterns from the numerical study, a response and updating rule methodology is proposed to predict the load-shortening curve of Progressive Collapse of Ship Structures Under Cyclic Loading structural component under cyclic load by updating the critical characteristics. The comparison with equivalent nonlinear finite element results shows an acceptable correlation. This novel method provides an efficient way to represent the cyclic buckling collapse response of structural members and is in an appropriate format for implementing in a Smith-type progressive collapse analysis for estimating the hull girder response. Following the response and updating rule load-shortening curve methodology, an unique extension to the Smith method is introduced for predicting cyclic bending response. Case studies are completed out on several ship-type box girder structures under different combinations of cyclic loads. The validation with nonlinear finite element analysis shows the rationality of the proposed extension, and also demonstrates that the prediction of cyclic response is highly sensitive to structural component’s post-collapse behaviour. An uncertainty evaluation procedure is developed to analyse the effects of critical features of the load-shortening relationship on the hull girder response prediction. The influences of different load-shortening features, including elastic stiffness, ultimate compressive strength, ultimate strain and post-collapse stiffness, are quantified. It is indicated that the post-collapse stiffness of structural components have the largest influence as suggested by a sensitivity index. In addition, this procedure is not only useful for the cyclic response, but also the conventional assessment concerning monotonic load. The outcome of this research work is a validated method which has the potential to improve the safety of ships by considering cyclic load effects.
Description: Ph. D. Thesis.
URI: http://hdl.handle.net/10443/5340
Appears in Collections:School of Engineering

Files in This Item:
File Description SizeFormat 
Li Shen Final Submission.pdfThesis18.05 MBAdobe PDFView/Open
dspacelicence.pdfLicence43.82 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.