Dipole source-based virtual three-dimensional imaging for propeller noise

The increasing attention to aerodynamic noise radiation from propellers has driven the interest in noise source identification and promoted the development of urban aerial vehicles over the last decades. Conventional acoustic imaging algorithms primarily assume monopole-type sources, but modern aeroacoustic theories have recognized that the dipole-type loading noise is dominant for propellers. To obtain physically consistent noise source distributions, we propose a moving dipole source-based beamforming method. The method is developed in the time-frequency domain to recover the time history of three-dimensional mapping for source strength and directivity, different to other techniques. The capability of the proposed method to identify noise sources is assessed by considering a two-bladed propeller benchmark. Instead of directly conducting experiments, we use results from high-fidelity numerical simulations, validated by measurements, to perform a noise source analysis. In this way, it is easier to verify the obtained source locations and directions. The results show that the noise sources are located on the trailing edge around the tip of the propeller, and the directivity varies with frequency. The proposed method exhibits competitive advantages in correctly tracing sources over conventional monopole-based methods. The proposed method can recover the three-dimensional noise sources for a high-speed rotating propeller. The consistency between the flow field and acoustic imaging results confirms the accuracy of the proposed method and suggests its potential in industrial applications.