A study of low speed flow noise and its reduction by numerical simulations

Computational Aeroacoustics (CAA) is the coupling of solving the flow field using numerical method and it predicts the sound field generated by the flow numerically using the results obtained from the flow field. This method can be classified into two categories namely, direct CAA and Hybrid CAA respectively. Direct CAA can solve the flow field and the sound field simultaneously while Hybrid CAA can solve the flow field first, and then uses the results obtained from the flow field to calculate the sound field. These two approaches are studied systematically as a means to reduce the noise generated by a flow field. Direct and Hybrid CAA were validated by comparing the results obtained computationally and experimentally on two benchmark cases. In addition, the Direct CAA was compared to Hybrid CAA to show that both approaches were feasible; results indicate that the Hybrid Approach was more efficient. Following the studies, a ventilated deflector was designed by utilizing the CAA method to reduce the wind buffeting noise by a subsonic flow over a cavity. The mechanism of buffeting noise was exploited by considering a model cavity. Extensive experiments were conducted to validate the computational scheme developed in this thesis. Experimental and computational results showed that the designed ventilated deflector allows an effective control of the buffeting noise at higher flow speeds. Another study, that uses the CAA approach is the study of flow induced noise inside a generic server rack. A hybrid CAA was used to explore the cause of noise creation in the server rack, while proposing a new design that split the flow into front and rear panels. Computational results showed that the splitter was able to suppress the tonal component that appeared in a generic server rack