Safety assessment for UIKKU tanker by multivariate Gaidai reliability method, utilizing self-deconvolution extrapolation scheme
摘要
This case study proposes a novel risk assessment approach for naval dynamic systems, which may be numerically modeled using MC methods or directly physically measured. Due to “curse of dimensionality”, existing risk assessment approaches are not well suited to dealing with cross-correlations between important system components or the high dimensionality (above bivariate, i.e., 2D) of large naval dynamic systems. The purpose of this case study was to demonstrate the practical efficacy of the proposed multimodal methodology by applying it to onboard measurements from a UIKKU chemical tanker. The current analysis used onboard measurements of ice loadings on the UIKKU tanker’s hull and propulsion systems during its Arctic cruise. The recorded dynamic data were examined to determine the design values of dynamic loads. The UIKKU’s hull was fully instrumented to study ice-induced stress levels and loads on the shell plating in various longitudinal and vertical orientations. The current study investigated global ship behavior when meeting thick ice ridges by measuring longitudinal bending strains at two points along the hull, as well as vertical ship accelerations to analyze dynamic loads and strains in the ice belt grillage near the vessel's bow, the Russian partners outfitted the icebreaker Capitan Dranitsyn with appropriate sensors and instruments. Throughout the Arctic expedition, the icebreaker Capitan Dranitsyn served as an escort vessel. The International Association of Classification Societies (IACS) may utilize presented reliability methodology to unify its Polar Class (PC) ship regulations, as well as to better understand and validate the ice load calculation procedures that underpin these rules. The Polar Class (PC) refers to the ice class assigned to a ship by a classification organization based on the IACS unified standards for Polar Class ships. There are seven Polar Classes, ranging from PC1 (which works in all polar waters year-round) to PC7 (which operates on thin first-year ice during the summer and fall). This case study briefly described the onboard measuring system and the results obtained during the cruise. Furthermore, a novel deconvolution extrapolation approach has been presented, providing a non-parametric design alternative to conventional parametric extrapolation methods.