The aim of this paper is shape optimization of a cryogenic flow past an isolated hydrofoil in order to reduce the cavitation. The numerical simulation of cavitating flows has been performed by way of the commercially available code Fluent (release 6.3), implementing a cavitation model by using external routines. The model is based on a simplified Rayleigh-Plesset equation, and takes into account both nucleation and thermal effects. This study has been divided in two parts. Firstly the cavitation model has been validated by comparison with experimental data, in particular, water cavitation on a NACA0015 airfoil and hydrogen cavitating flow over an external profile. Secondly, Fluent has been coupled with a multi-objective genetic algorithm (MOGA). Genetic algorithms have proved their interest with respect to gradient-based methods because of their high flexibility, and also because of their ability to find global optima of multi-modal problem. The representation of the design space has been previously investigated through a Design of Experiment (DOE) procedure. A shape optimization of an hydrofoil has been computed in order to minimize the vapor volume in different operating conditions.
Shape Optimization For Cryogenic Cavitating Flows Past An Isolated Hydrofoil
DE GIORGI, Maria Grazia;RODIO, MARIA GIOVANNA;CONGEDO, PIETRO MARCO;FICARELLA, Antonio
2008-01-01
Abstract
The aim of this paper is shape optimization of a cryogenic flow past an isolated hydrofoil in order to reduce the cavitation. The numerical simulation of cavitating flows has been performed by way of the commercially available code Fluent (release 6.3), implementing a cavitation model by using external routines. The model is based on a simplified Rayleigh-Plesset equation, and takes into account both nucleation and thermal effects. This study has been divided in two parts. Firstly the cavitation model has been validated by comparison with experimental data, in particular, water cavitation on a NACA0015 airfoil and hydrogen cavitating flow over an external profile. Secondly, Fluent has been coupled with a multi-objective genetic algorithm (MOGA). Genetic algorithms have proved their interest with respect to gradient-based methods because of their high flexibility, and also because of their ability to find global optima of multi-modal problem. The representation of the design space has been previously investigated through a Design of Experiment (DOE) procedure. A shape optimization of an hydrofoil has been computed in order to minimize the vapor volume in different operating conditions.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
