Assessment of UAM and drone noise impact on the environment based on virtual flights

Abstract

Drones and urban air mobility vehicles, which are mostly intended for urban operations, have attracted intensive research interest. Their environmental noise impact should not be ignored and should be a certification consideration. Numerical simulations can provide means to efficiently assess the noise impact on the complex urban environments, thanks to the advances in both software and hardware. In this study, virtual flights of the noise sources to mimic the drones and urban air mobilities are conducted to investigate the environmental impact by using the Gaussian beam tracing method implemented in an in-house solver Environmental Acoustic Ray Tracing Code, in which the dominant acoustic processes in outdoor applications are modelled. The solver also allows for modelling of sources with generic directivity patterns, which is beneficial for investigating the drone and urban air mobility noise produced by multiple rotors. Two examples of studying noise impact on building and realistic urban environment are conducted. The effects of the surface acoustic impedance, source type and flight path are investigated. By using the efficient simulation solver, the noise impact can be quickly assessed through virtual flights, providing a promising approach to assist the low-noise flight planning and noise certification.

Introduction

Drones have been developed with a broad range of applications in the civilian sector, such as logistics, photography, engineering, and environmental monitoring [1], [2]. A straightforward extension to larger vehicles is the urban air mobility (UAM) that is attracting intensive research interest from both academic and industrial communities [3]. The noise problem brought by the growing number and application of drones and UAMs has arisen the concern of legislators on the proper flight regulations. The vehicles may fly in the proximity of buildings, especially in missions of delivery and structure inspection, community residents also care about the noise level on the building facades and their sound transmission into the buildings. The community noise impact due to the manoeuvring flights of the vehicles has attracted research interest [4], [5], [6]. A challenging in the studies is that full three-dimensional (3D) urban environment should be considered for noise assessment. Therefore, an effective and efficient tool to evaluate the noise level in the realistic complex urban environment is needed.

Field tests can provide noise measurement results directly, which, however, could still be inconvenient, inefficient, expensive and vulnerable to uncertainties. Alternatively, a part of noise assessment on could be virtually conducted through some numerical platform to allow for parametric study. However, direct numerical solutions of the sound propagation equations [7], [8], [9] are still computationally too expensive for realistic problems with large propagation distances, especially at high frequencies. In numerical simulations, Casalino et al. [10] conducted a hybrid approach to simulate the helicopter community noise in complex urban geometry. The noise scattering by the fuselage was simulated using a finite element method (FEM) and the projection to the far field is solved by Ffowcs-Williams and Hawkings analogy. Another approach is based on the boundary element methods (BEM) solving the governing Helmholtz equation need to be coupled with specialized Green function for refractive media [11]. Other wave-based methods can encounter difficulties when considering complex boundaries/terrains with varying impedance values or applied beyond the critical elevation angles, such as the fast field program (FFP) [12], [13] and parabolic equation (PE) solvers [7], [14].

Geometric acoustic methods are known for their high efficiency in handling problems with boundary surfaces. Among them, the image source method [15] could be computationally inefficient with the increasing complexity of the boundary conditions, and it is limited to the homogeneous medium. By contrast, the ray tracing method can easily incorporate the boundary geometry, material and the medium characteristic [16], [17]. However, the method may suffer from unrealistic caustics and shadow regions [18]. Also, the determination of eigen-rays in the 3D applications can lead to large computation cost. To this end, the Gaussian beam tracing method that asymptotically approximates the wave equation [19] was proposed to improve the ray tracing methods. The smoothing effect of the beams on the singularities of the ray-theoretic field greatly improves the prediction accuracy. Also, the computation is efficient as there is no need to search for the eigen-rays [20]. In the conventional Gaussian beam tracing method, the rays are emitted from an omnidirectional model, suggesting that the beam strengths are identical in all directions. For an arbitrary source, multiple point sources might be employed to synthesise the directivity, which has been known to have significant impact on the radiation property [21], [22], [23], [24]. The approach could be computationally inefficient and calls for development of Gaussian beam tracing method on account of generic directivity pattern.

This paper introduces a concept of using virtual flights, other than field testing, to assess the noise impact on the environment. Section 2 introduces the Gaussian beam tracing solver employed. Section 3 introduces two virtual flights applications, and Section 4 is a summary.