Stochastic Nonlinear SDOF Model and Probabilistic Fragility Assessment of Steel Plates Under Blast Loading

We present a reduced-order single-degree-of-freedom (SDOF) formulation that couples non-linear material behaviour with rate-dependent steel properties and explicit uncertainty propagation for steel plates subjected to blast over-pressure. The model is calibrated against experimental data and finite-element simulations, achieving < 5 % error on peak displacement while cutting computational time by four orders of magnitude. A Monte-Carlo framework (15,000 realizations) treats TNT mass as a log-normal variable and delivers fragility curves for slight, moderate and severe damage states. Results show that, for a 3 mm plate, the ?-parameter of the fitted sigmoid increases from 5.1 to 8.4 when explosive mass rises from 50 kg to 200 kg, indicating steeper reliability gradients. The proposed approach enables rapid performance-based assessment and is readily extendable to other impulsive loads or parameterized digital-twin environments.