Epistemic uncertainties on the estimation of minimum air gap for semi-submersible platforms

Abstract

Different models are considered to estimate the minimum air gap of a semi-submersible accommodation unit in stormy conditions. The results are compared to quantify the epistemic uncertainties in the estimation of the minimum air gap. While both first- and second-order methods have been broadly considered in the literature to evaluate the air gap of semi-submersibles, the underlying hypotheses of these methods are considerably simplified, and conservative results depend on added margins of unknown degrees of confidence. Therefore, a systematic study is conducted to assess the uncertainties raised by some features of the state-of-the-art methods. First, the uncertainties resulting from the binned values of scatter diagrams and from the assumption that the highest significant wave height sea state is the critical sea state, are assessed. Then, the uncertainty implied in the finite set of control points is evaluated. The achievements are obtained considering First-Order Second-Moment methods, by performing searches over the environmental contours to find the actual critical sea states, and by performing Kriging interpolation over the whole underdeck of the platform. Moreover, the adopted uncertainty propagation tools allow characterising the minimum air gap in terms of its expected value and standard deviation. It is found that, before adding complexity to the wave-structure interaction model, air gap estimation is subject to large errors if rather few control points are assigned, or if a single critical sea state is assigned before performing any search for the actual critical sea state.