Simulation Based Design with Variable Physics Modeling and Experimental Verification of a waterjet propelled Catamaran

Through a collaboration among IIHR, INSEAN-CNR and NMRI, the new developments of an advanced Simulation Based Design (SBD) design tool box, based on High Fidelity URANS simulations, has been tested to optimize a high speed, waterjet propelled catamaran. Both single- and multi-objective optimization problems have been solved by using different evolutionary type optimization algorithms. The novelty of the paper is related to the extension of the Variable Fidelity/Variable Physics (VF/VP) idea to global optimization problems. The VFVP approach aims at the reduction of the overall computational time when the computational cost of the High Fidelity (HF) objective function is too high: this is obtained by combining the use of the HF with a Low-Fidelity (LF) solver, cheaper but less accurate. Initially devoted to local optimization problems using gradient based optimization algorithms, the VF/VP has been now extended to global optimization problems, using derivative-free (evolutionary type) optimization algorithms. The VP/VF is based on the systematic use of CFDSHIP Iowa URANS solver (the HF solver), and of the WARP INSEAN-CNR linear potential solver (the LF solver). A dedicated experimental campaign has been initially carried out in two towing tanks, including self propulsion tests, to validate LF and HF solvers. The HF optimization of the inlet of the waterjet has been also performed. Final results show improvements of about 6% on the total resistance of the catamaran.