Rational Approach to Assess the Effect of Corrosion on Stiffened Panel Buckling and Ultimate Capacity

Abstract

During the annual In-Service Inspection of a spar hull, several regions of pitting corrosion on the upper portion of the north and south moon pool external wall plating were identified. The moon pool walls are constructed as typical stiffened panel structures. Visual, ultrasonic (UT), and pulsed eddy current (PEC) inspections indicated regions of corrosion with roughly 40% to 70% averaged localized wall loss. This paper discusses the analytical assessment of the structure to determine the effect of the corrosion on the structural integrity of the moon pool wall and any similar structural panel. To determine the impact of corrosion on the stiffened panel integrity, a finite element (FE) based analysis approach is used to perform a comparative assessment of the "as-built" and "corroded" configuration of the moon pool wall. The nominal plate and stiffener thicknesses are modeled in the "as-built" configuration; whereas, the measured plate thickness from the inspection is modeled in the "corroded" configuration. The structure is subjected to design loads based on the storm damaged design condition. The analysis is performed by uniformly increasing the applied loads until failure occurs, maintaining a constant ratio between the nominal loads. Two different analyses are performed as a part of the strength assessment: (1) a linear-elastic eigenvalue analysis to estimate the elastic buckling capacity and mode shapes of the structure and (2) an elastic-plastic post-buckling analysis to estimate the ultimate capacity of the structure. In addition, the results from the linear-elastic eigenvalue analysis are compared to the results from analytical buckling calculations. The analysis results indicate that the corrosion reduces the elastic plate buckling capacity significantly. However, the overall capacity of the stiffened panel is not significantly reduced. Therefore, from a global strength perspective, the stiffened panel remains acceptable in its corroded condition. The upper portion of the moon pool wall is typically fatigue insensitive in spars. Therefore, the effect of the corrosion wall loss on the fatigue performance was not assessed. Since there is limited guidance in design and assessment codes for assessing corroded stiffened panels, this approach can be used to address future stiffened panel corrosion wall loss. In addition, this method allows for inclusion of future corrosion allowance, if applicable. Determining the capacity of corroded panels using FEA-based numerical methods, like those described in this paper, allows the operators to manage their risks, repair costs, and inspection frequency by determining the actual capacity of the damaged components. This allows the operators to determine the appropriate mitigation measures based on a quantitative risk calculation.