Noise reduction of Urban Air Mobility Vehicles using CFD

过量的陆上交通排放是由拥堵、公共交通不足和缺乏门到门的交通网络造成的。内燃机将在未来几十年内在英国逐步淘汰。城市空中机动车辆(UAMV)提供了利用空域极大地增加运输能力和提供近门到门服务的机会，同时大幅减少碳氢化合物燃料的排放。包括空中客车和劳斯莱斯在内的许多制造商正在开发这种飞机用于客运/货运，而亚马逊等其他制造商则在测试较小规模的送货无人机。推进系统大多基于电动垂直起降(EVTOL)概念，使用多个旋翼在人口稠密地区降落。噪音是使用地点的主要排放物，阻碍了社会对这些车辆在城市中的接受和使用。如果噪音能够降低，这一领域将会快速增长。拟议的项目旨在对包括旋翼部分、旋翼尖端和机身相互作用在内的关键飞机特征的流动和噪声进行建模。以大涡模拟(LES)为代表的高保真非定常计算流体力学(CFD)已成功地用于预测喷气、起落架和翼型噪声。这将为机翼流动控制和机身改装等降噪策略提供参考。高性能计算(HPC)将用于生成和分析大型非稳定数据集，并使用并行工具。相对于飞机，表面附近的湍流的差异尺度要求对湍流和几何表示进行多保真建模。来自这些模拟的数据将为低保真建模提供信息，以对整个飞机进行建模，并将利用机器学习/人工智能。然后，这些大型模型可用于研究(多条)飞行路径的影响、地面上的噪音足迹以及噪音缓解策略的有效性。

Excess land transport emissions result from congestion, under-occupied public transport and lack of door-to-door transport networks. Internal combustion engines are to be phased out in the UK in the coming decades. Urban air mobility vehicles (UAMVs) offer an opportunity to utilise airspace to vastly increase transport capacity and provide near door-to-door service, whilst drastically cutting emissions from hydrocarbon-based fuels. Many manufacturers including Airbus and Rolls-Royce are pursuing such craft for passenger/freight transport, whilst others such as Amazon are testing smaller scale delivery drones.Propulsion systems are mostly based on electric vertical take-off and landing (eVTOL) concepts, using multiple rotors to enable landing in densely populated areas. Noise is the dominant emission at point of use, preventing social acceptance and use of these vehicles within cities. If noise can be reduced, this sector will see rapid growth.The proposed project aims to model flow and noise of key aircraft features including rotor sections, rotor tips and airframe interactions. High fidelity unsteady Computational Fluid Dynamics (CFD) such as Large-Eddy Simulation (LES) has been successfully used to predict jet, landing gear and aerofoil noise. This will inform noise reduction strategies such as aerofoil flow control and airframe modifications. High Performance Computing (HPC) will be used to generate and analyse the large unsteady datasets with parallel tools. The disparity in turbulence flow scales near surfaces relative to the aircraft requires multi-fidelity modelling of turbulence and geometry representation. Data from these simulations will inform lower-fidelity modelling to model entire aircraft and will be exploited with machine learning/AI. These large-scale models can then be used to study impact of (multiple) flightpaths, noise foot prints on the ground and effectiveness of noise mitigation strategies.