Data driven models for propeller cavitation noise in model scale

Model scale experiments are currently considered the stateof-the-art approach for studying cavitation noise and predict the acoustical performance of a cavitating propeller. Unfortunately, this approach requires time-consuming tests in a cavitation tunnel with a model of the propeller and the experiments are usually affected by scale effects. Among these, viscous effects on the development of vortex cavitation may alter significantly the cavitation pattern, preventing in some cases to consistently reproduce the full scale cavitation pattern in model scale. Due to the above, the availability of methods allowing for the modelling and prediction of cavitating propeller radiated noise may represent an attractive opportunity to estimate propeller noise without requiring an actual experiment. In this paper an approach based on hybrid modelling is proposed for predicting model scale cavitating propeller noise. The proposed approach exploits both the physical knowledge of the problem and the real data obtained from extensive cavitation tunnel experimental campaigns performed on different propellers in many operational conditions. Results on real data will support the validity and the effectiveness of the proposal.