This study presents an integrated framework for the quantitative structural integrity assessment of austenitic stainless steel (SUS304L) used in cryogenic structures, such as the Cargo Containment Systems (CCS) of Liquefied Natural Gas (LNG) carriers....
This study presents an integrated framework for the quantitative structural integrity assessment of austenitic stainless steel (SUS304L) used in cryogenic structures, such as the Cargo Containment Systems (CCS) of Liquefied Natural Gas (LNG) carriers. Conventional inspection methods, including visual inspection and leak detection, have inherent limitations in quantitatively evaluating incipient damage, potentially overlooking significant structural risks.
To overcome these limitations, this research is built on two complementary pillars. The first, an engineering pillar, proposes a novel magnetic non-destructive testing (M-NDT) framework that leverages the phenomenon of Strain-Induced Martensitic Transformation (SIMT) to quantify accumulated plastic damage. The second, a scientific pillar, addresses the fundamental challenge of strain path dependency, which undermines the quantitative reliability of the M-NDT signal. Through uniaxial and equi-biaxial tension tests, it was experimentally demonstrated that the SIMT kinetics in SUS304L exhibit a pronounced dependency on the strain path. Subsequent Electron Backscatter Diffraction (EBSD) analysis revealed that the activation of multiple slip systems and an increased density of shear band intersections are the core microstructural origins of this behavior.
Building on this physical insight, this study establishes a new constitutive equation by introducing a strain path parameter to quantify the mode of deformation, which is integrated with the equivalent plastic strain representing the magnitude of deformation. When implemented in a finite element analysis to simulate a hydraulic bulge test, the model's prediction of the complex and anisotropic final distribution of martensite was validated through comparison with experimental results.
This outcome is significant, as conventional models based solely on equivalent strain fail to capture such path-dependent behavior. Furthermore, by utilizing a directly measurable strain path parameter, this approach offers a practical alternative to preceding studies that have relied on calculated, non-measurable stress state variables like stress triaxiality, thereby enhancing model robustness.
By integrating a practical NDT framework with a physics-based predictive model through material-level experiments, this research confirms the potential for magnetic NDT to be used as a quantitative damage diagnostic tool, transcending its role as a merely qualitative indicator. This work provides the foundation for subsequent research on the inverse estimation of a structure's actual strain history from measured magnetic signals and is expected to contribute to the advancement of data-driven integrity management technologies.