A Low-noise airfoil for low Reynolds applications: a multi-fidelity optimization

Authors

Nicola Casari - Department of Engineering, University of Ferrara, via Saragat 1 44122 Ferrara, Italy
Stefano Oliani - Department of Engineering, University of Ferrara, via Saragat 1 44122 Ferrara, Italy
Michele Pinelli - Department of Engineering, University of Ferrara, via Saragat 1 44122 Ferrara, Italy
Mattia Piovan* - Department of Engineering, University of Ferrara, via Saragat 1 44122 Ferrara, Italy

Abstract

Automotive fans, small wind turbines, and manned and unmanned aerial vehicles (MAVs/UAVs) are just a few of the examples in which noise generated by the flow interaction with the aerodynamic surfaces is a major concern. The always quieter vehicles (also considering the increasing share of electric cars) is pushing manufacturers to reduce the overall noise generated by the fan and other auxiliary units. Similarly, there is an interest in less noisy wind turbines: the increasing share of renewable energy in the worldwide electrical generation will require wind farms to be placed in proximity of human habitats. Lastly, UAVs (e.g. drones) often produce a significant amount of acoustic noise that interferes with their operation in urban and other inhabited areas, particularly during take-off, landing, and low-level flight. This is one of the aspects which are hindering the development of urban air mobility. For all of these applications, the flow regime in what is typically referred to as a low-to-moderate Reynolds number (< 5e5) is of great interest. In this regime, a limited number of tests and numerical data are available for the prediction of both aerodynamic performance and noise generation. In light of these considerations, the current work shows the potential of a new airfoil shape to minimize the noise generation, maintaining high efficiency in the low-Re regime (Re < 2e5). The investigation is carried out by means of a multi-fidelity approach: a low-fidelity semi-empirical model is exploited for evaluating the sound pressure level (SPL). The fast evaluation of the low-cost function enables the computation of a large range of possible profiles, and accuracy is added to the low-fidelity response surface with high-fidelity CFD data. The constraint of maintaining a pre-defined range of the lift coefficient and of efficiency ensures the possibility of using this profile in usual design procedures.