09:00
60 mins
|
Sound transmission of drones in urban wind conditions
Xin Zhang
Abstract: This keynote lecture will discuss the challenges and recent advances in evaluating sound transmission in inhomogeneous urban wind environments, with the aim of correctly and efficiently capturing the relevant acoustic physics at HKUST. First, a computational method based on a variant of Gaussian Beam Tracing (GBT) is introduced to investigate the impact of inhomogeneous mean airflow on multi-frequency sound transmission. The complex physics of broadband drone noise mapping is then described, including waveform distortion, steepening, and folding, arising from the accumulation of absorption mechanisms and dispersion in urban wind environments. Second, an acoustic scattering model based on high-order beam series is proposed to assess the impact of unsteady local airflow structures on sound transmission. Third, an efficient real-time prediction tool is developed using a machine learning (ML) approach. The model is based on a U-shaped neural network (U-Net) trained on data generated by the aforementioned GBT methods.
|
