Probabilistic analysis of the response of plates subjected to near-field blast loading

Accurate prediction of the response of structures subjected to close proximity blast loads is a pressing engineering concern; the landscape of global terror has shifted away from large and indiscriminate bombings towards much smaller and more targeted attacks (e.g. against critical infrastructure and/or transport). In such close-proximity blast events (in the so-called ‘nearfield’), interaction between the expanding detonation products and air shock gives rise to complex hydrodynamic features which introduce localised variations in the pressure field. The resultant loading (typically defined in terms of specific impulse since loading durations act on timescales considerably shorter than structural response) is therefore highly uncertain, and even nominally identical experiments produce loading distributions with a high degree of local variability. Current predictive approaches either grossly simplify or neglect entirely the inherent ‘fuzziness’ of nearfield blast loading, to the extent where it is currently unknown what effect this has on structural response, how sensitive plate structures are to uncertainties in loading distribution, and how this varies with plate properties and loading condition (e.g. charge mass and stand-off distance). This paper presents a numerical study aimed at answering these questions, where specific impulse distributions are probabilistically simulated with varying degrees of localised variations and mapped onto a range of different plates. This work aims to shed light on the fundamentally stochastic nature of close-proximity blast, with a view to implementing the findings in fast running engineering models for prediction of plate response under near-field blast loading.