Fatigue reliability of a singlestiffened ship hull panel

Fatigue reliability of a single stiffened ship hull

panel

Hussam Mahmoud a ,⇑,Guillermo Riveros b

a Department of Civil and Environmental Engineering,Colorado State University,Fort Collins,CO 80523,United States b

US Army Corps of Engineers,Engineer Research &Development Center,Vicksburg,MI 39180,United States

a r t i c l e i n f o Article history:

Received 11May 2013

Revised 20November 2013Accepted 18February 2014

Keywords:Fatigue Cracks

Finite element

Stress intensity factor Monte Carlo simulations Residual stresses Inspection

Probability of failure Reliability

a b s t r a c t

Fatigue cracks in ship structures are considered a nuisance as they require periodic inspection and repair.If left unrepaired,the crack could grow to reach a critical length and threaten the integrity of the structure.Although the cracks are typically characterized by stable propagation rate,the scatter in fatigue performance is difficult to quantify and could be on the order of thousands or even millions of cycles.The development of maintenance and management programs for ship structures should therefore account for the inherent scatter in performance through probabilistic fatigue assessment.Probabilistic assessment by testing a large number of specimens can be very costly.Therefore,the assessment can be performed through conducting large number of numerical or analytical simulations that account for the inherent statistical scatter in both load and resistance.This paper presents a framework for probabilistic assessment of the propagation rate of cracks in welded stiffened panels using finite element Monte Carlo simulations.The parameters influencing the propagation rate are treated as random variables with predefined statistical distributions.The results can be used for proposing inspection intervals for ships.In addition,very useful insight can be drawn on the most sensitive parameters affecting crack growth in the panels and the probability of failure at a given inspection period.

Ó2014Elsevier Ltd.All rights reserved.

1.Introduction

Structural elements are susceptible to fatigue cracking when subjected to fluctuating loads.The initiation of such cracks de-pends on the severity of the stress concentration and the nominal stress range.The total fatigue life of a detail is the sum of the crack initiation and crack propagation cycles.However,the initiation phase represents only small fraction of the total fatigue life of a de-tail in redundant structural systems.Therefore,the propagation stage is considered one of the primary factors affecting the safety of structures.Although the propagation is typically stable until it reaches a critical size,it is difficult to quantify due to scatter in fatigue performance.The development of maintenance and management programs for civil and marine structures should therefore account for the inherent scatter in performance through probabilistic fatigue assessment.

An example of a problematic fatigue detail is welded stiffened panels,which are used in numerous civil applications including,for example,floating bridges,ship structures,and steel bridges.Although crack growth in plates and riveted stiffened panels has been extensively studied,few of the studies were focused on crack propagation in a panel with multiple welded stiffeners.For ship structures applications,large scale experimental data exist on the propagation of cracks in welded stiffened plates [1–3].However,due to the significant cost associated with experimental testing,the data available is limited and only cover small number of parameters.Small number of finite element models were devel-oped and although compared well with experimental results,still showed observable difference.This is primarily due to the fact that the number of models developed is not large enough to account for the large scatter in the crack growth parameters [1–3].

Various inspection procedures have been proposed and verified in terms of their quality and reliability.Kim et al.[4]concluded that cracks greater than 200mm in length could be detected 70%of the time,while a study by Demsetz [5]estimated only 50%probability of detection for a crack less than 300-mm long.Luckily,ship structures have demonstrated an ability to tolerate large cracks without jeopardizing the integrity of the ship.For example,a 20m crack developed in the deck of the ship ‘‘Castor’’during a rough weather while loaded with 29,000tons of unleaded gasoline.Additional example is a 150mm long crack that was noted to propagate to 8m in a US Navy military frigate,in a single storm,without fracturing [6].Another severe storm caused a crack to propagate to a length of 15m across the deck of the 744-foot Roll-on,roll-off vehicle carrier [6].

/10.1016/j.engstruct.2014.02.0070141-0296/Ó2014Elsevier Ltd.All rights reserved.

⇑Corresponding author.Tel.:+19704916605.

E-mail address:hussam.mahmoud@ (H.Mahmoud).