A major challenge in marine engineering simulations is to quantify the interdependence of the uncertainties in the input and output model parameters. We consider this interdependence for stochastic ocean waves in a shallowwater wave propagation model, with uncertainties in the kinematic boundary condition (BC). The BC involves 302 random variables that, together with a JONSWAP wave spectrum, define the nonlinear wave surface elevation. We discretize the wave propagation model using generalized Polynomial Chaos (gPC) and a Sparse-Grid non-intrusive Stochastic Collocation Method. Initially, the number of sparse grid points is infeasibly high, and it is therefore reduced using an active-subspace analysis (ASA) [Paul G. Constantine, Active Subspaces Emerging Ideas for Dimension Reduction in Parameter Studies, Society for Industrial and Applied Mathematics, 2015], where important general directions in the parameter space are identified. We then compute the four first statistical moments of the wave surface elevation, and use them in a moment-based Gauss transformation to analyze the up-crossing probability for the shallow-water model. We demonstrate the advantages and the effectiveness of the ASA-gPC approach against classical Monte Carlo implementations.
|Number of pages||1|
|Publication status||Published - 2018|
|Event||SIAM Conference on Uncertainty Quantification 2018 - Hyatt Regency Orange County, Garden Grove, United States|
Duration: 16 Apr 2018 → 19 Apr 2018
|Conference||SIAM Conference on Uncertainty Quantification 2018|
|Location||Hyatt Regency Orange County|
|Period||16/04/2018 → 19/04/2018|