Investigation of Corroded Ship Welds using a Finite Element Analysis Approach

Ships and offshore structures operate under harsh environmental conditions which commonly lead to the corrosion, particularly in areas which are constantly immersed. The persistence of corrosion especially at the wells between plate and at plate intersections can affect structural strength and the structural response of the asset. Due to this, it is of great interest to fully understand the phenomenon which is weld corrosion and the structural and life deteriorating effects of which it causes.

Corrosion in and around weld zones, namely the overall effect it has on steel and offshore structures is currently largely misunderstood. One of the most common types observed is pitting corrosion. 'Pitting' is categorised as a highly concentrated form of corrosion which typically appear as spherical imperfections on metals which have been exposure to the marine environment. 'Pitting' in and around weld zones has only really been explored in industry in the last ten to fifteen years, so the current state of understanding of the phenomenon and its effects is still very low. The research provided within this thesis, strives to grow upon what is currently known and provide further understanding.

Within this thesis the effect of pitting corrosion is explored through a combination of laboratory testing and finite element analyses. The laboratory testing consists of measuring the widths and depths of corrosion pits which have formed as a result of exposing a number of weld samples to the marine environment within the Port of Newcastle. The weld samples used for this investigation differed in levels of exposure, ranging from 0.5 to 3 years. The finite element analyses consisted of two independent methodologies. Both methodologies have the same goal of understanding the effect corrosion pit growth has on the distribution of stress as pits form in and around weld zones. The first methodology explored, covers a simplified cross-section model of a ballast tank which has 'pits' induced in five different location around one of the weld connections. Whereas the second methodology investigates a more complex substructural vessel model with pitting in four different locations.

Upon the completion of the laboratory testing and finite element analyses, further data processing is explored. It is shown that the relation between the depth to width ratio and exposure period can be described by a trend where there is an initial rise before peaking, decreasing and increasing once more. It is also shown the relationship between the depth to width ratio and stress increase factor (SIF) due to pitting can be described by a singular equation which incorporates both a combination of a fourth order polynomial and an exponential rise function. Through correlating the depth to width ratio and exposure trend, and SIF equation, a relationship between the exposure period and SIF is then shown.

Throughout the experimental and data processing multiple conclusions are noted, which outline key takeaways from the investigations covered within this thesis. Generally, an increased understanding of the effect pitting corrosion, near welding, has on offshore structures has been achieved, which allows for improvement to maintenance procedures allowing for the life extension of critical assets.