Set-Based Approach to Design under Uncertainty and Applications to Shaping a Hydrofoil

Abstract : The paper presents a framework for set-based design under uncertainty and demonstrates its viability through a shape design case of an efficient super-cavitating hydrofoil for ultra-high speed maritime vehicles. Redirecting the effort away from optimal designs to those that safely meet relevant requirements as quantified precisely by super quantile measures of risk (s-risk), the paper addresses situations faced in practice where elimination of poor designs is more important than identifying a best one. The complexity of the design case, involving multiphysics and several design constraints, necessitates the use of surrogate models, which are built using high- and low-fidelity simulations and are here tuned for the first time to reflect s-risk. Accounting for both parameter uncertainty as well as errors in surrogate models, we obtain robust designs that satisfy requirements under a variety of manufacturing and operating conditions.