Wall thinning caused by corrosion changes the circumferential strain response of storage tanks, but nonlinear coupling among structural parameters, loading conditions, strain characteristics, and defect geometry makes direct inversion difficult. This paper develops a strain-based, finite element (FE)-based machine learning model to predict corrosion depth and pit diameter from externally observable strain features. A reduced-scale hydrostatic test verified the external observability of localized inner wall thinning. A parametric finite element model was then used to generate 1072 samples. TabNet was selected among five regression models optimized using Bayesian optimization (BO), and Multi-Head Attention (MHA) was incorporated to improve feature interactions. On a held-out test subset of the FE-generated dataset, the dual-output BO-MHA-TabNet achieved mean absolute errors (MAEs) of 0.226 mm for corrosion depth and 24.252 mm for pit diameter. The corresponding R² values were 0.9612 and 0.9377, respectively. Shapley additive explanations (SHAP) analysis identified the strain concentration factor and maximum circumferential strain as the most significant features, consistent with local stiffness reduction and strain concentration. This study is a numerical proof of concept supported by an observability experiment. The proposed framework provides an interpretable approach for strain-based quantitative evaluation of corrosion defects in storage tanks.

Internal wall loss in hot-fluid pipelines is difficult to monitor during operation because weak thermal perturbations are masked by global heating, axial cooling, and external heat dissipation. This study develops an external thermometry framework combining a flexible-base fiber Bragg grating (FBG) array with residual-peak morphology analysis. A closed-loop hot-water rig with five artificial wall-loss regions was tested under exposed-air and buried-soil boundaries, and a validated conjugate heat-transfer model generated 269 controlled scenarios. Experiments showed residual anomalies above measurement uncertainty, with a maximum repeatability standard deviation of approximately 0.12 °C and the clearest signals at 90–120 s after hot-water injection. Boundary conditions strongly affected observability at 115 mm and 70 °C, and the residual peak increased from about 0–1 °C in exposed air to 8–9 °C under the buried boundary. Simulations showed that defect width expanded the disturbed region from approximately 100 to 210 mm, while peak amplitude remained coupled to width and depth. Six morphology descriptors jointly estimated position, width, and depth, with mean absolute errors (MAEs) of 0.547, 1.647, and 0.245 mm, respectively. The method provides a recalibratable early-screening route for locating suspicious wall-loss regions before confirmatory inspection.