气动声学与湍流实验室Sound and Turbulence in Aerodynamics Research Lab

01研究RESEARCH

于流动深处,探声音之源。Within the flow, we seek the source of sound.

围绕复杂流动、声音传播与高效计算,我们发展可解释、可验证并面向工程设计的气动声学方法。Across complex flows, sound propagation and high-performance computing, we develop interpretable and verifiable aeroacoustic methods for engineering design.

研究方向RESEARCH DIRECTION

气动声学计算方法与高性能计算Computational Aeroacoustics & HPC

已有工作分别研究一维线化欧拉方程的保能通量重构、周期响应的谐波平衡、边界元降阶与声传递函数插值,以及声衬阻抗反演。GPU 求解与低马赫数模型比较则分别关注效率和物理适用性,这些方法在各自论文的算例中得到评估。The work covers energy-conserving flux reconstruction for one-dimensional linearized Euler equations, harmonic balance for periodic response, reduced-order boundary elements, transfer-function interpolation and liner-impedance identification. GPU algorithms and low-Mach model comparisons address efficiency and physical validity separately, within the cases evaluated in each paper.

旋翼气动声学数值计算示意 / Rotor aeroacoustic computation · illustration
旋翼气动声学数值计算示意Rotor aeroacoustic computation · illustration

01 · 频域与时域高精度格式Frequency- and Time-domain High-order Schemes

如何减少周期声学计算的过渡成本与长期传播误差?How can periodic acoustic simulations reduce transient cost and long-time error?

研究方法Approach

针对一维线化欧拉方程构造具有能量守恒特性的通量重构格式;针对周期声学问题,将谐波平衡与频域阵风激励、非反射边界条件结合,并用声源、振荡翼型及阵风干涉算例验证。An energy-conserving flux-reconstruction scheme is developed for the one-dimensional linearized Euler equations. For periodic acoustics, harmonic balance is combined with frequency-domain gust excitation and nonreflecting boundaries, with source, pitching-airfoil and gust-interaction test cases.

研究目的Purpose

分别评估一维格式的长期精度与所测周期算例的求解效率。Assess long-time accuracy of the one-dimensional scheme and efficiency in the tested periodic cases.

对应论文 · 3 篇Related publications · 3
  1. 2026

    He L, Jiang H. Efficient Frequency-Domain Aeroacoustic Simulation of Gust Response Using Harmonic Balance Method. AIAA Journal, 2026, 64(7): 3869-3883.

    ↗
  2. 2024

    Duan, L, Jiang, H. Flux reconstruction approach for long-time calculations of the linearized Euler equation. InterNoise, 2024.

  3. 2023

    Duan L, Jiang H. Numerical analysis of underwater acoustics in the presence of complex flow, InterNoise, 2023.

02 · 边界元与声传递函数Boundary Element Methods & Acoustic Transfer Functions

怎样降低宽频散射计算的求解与存储成本?How can broadband scattering reduce solution and storage costs?

研究方法Approach

将全阶边界元问题投影到低维子空间;对预计算声传递函数采用本征正交分解与离散矩阵插值,研究降阶、频率插值及误差,并以无人机机体散射算例展示方法。Full-order boundary-element problems are projected onto a low-dimensional subspace. Proper orthogonal decomposition and discrete matrix interpolation are applied to precomputed acoustic transfer functions, with error studies and drone-fuselage scattering demonstrations.

研究目的Purpose

在相关宽频散射算例中平衡计算效率、存储需求与精度。Balance computational cost, storage and accuracy in the studied broadband scattering cases.

对应论文 · 5 篇Related publications · 5
  1. 2019

    Jiang H, Zhang X*, Huang X. Reduced-basis boundary element method for efficient broadband acoustic simulation. Journal of Sound and Vibration, 2019, 456: 374-385.

    ↗
  2. 2021

    Jiang H, Zhong S, Zhang X*, Huang X. Efficient storage and interpolation of acoustic transfer functions. Engineering Analysis with Boundary Elements, 2021, 124: 259-265.

    ↗
  3. 2020

    Jiang H, Zhong S, Zhang X. Interpolation based acoustic transfer function for drone noise simulation. Quiet Drones International e-Symposium, 2020.

  4. 2022

    Li Y, Jiang H, Ma Z, Zhong S, Zhang X. Computational aeroacoustics of an urban air mobility vehicle using the acoustic preserved artificial compressibility method. Quiet Drones International e-Symposium, 2022.

  5. 2022

    Jiang H, Zhong S, Zhang X, Zheng C. Acoustic scattering in the noise prediction of flying urban air mobility. The Delft International Conference on Urban Air Mobility, 2022.

03 · 机器学习与数据驱动方法Machine Learning & Data-driven Methods

如何将物理约束和训练数据用于声学求解与参数反演?How can physics constraints and data support acoustic solution and parameter inversion?

研究方法Approach

研究基于 Transformer 和 ConvLSTM 的物理信息神经网络,在波动方程等问题上评估时序建模;另以有限元数据训练混合神经网络,对切向流阻抗管中的声衬阻抗进行反演,并与模型及实验数据对照。Transformer- and ConvLSTM-based physics-informed networks are evaluated on problems including the wave equation. In a separate study, a mixed neural network trained with finite-element data identifies liner impedance in a grazing-flow tube and is assessed against model and experimental data.

研究目的Purpose

改善时序方程求解与声衬参数识别,分别评估模型精度及适用条件。Improve time-dependent equation solution and liner-parameter identification, assessing accuracy and applicability for each task.

对应论文 · 6 篇Related publications · 6
  1. 2024

    Zhao, Y, Zhong, S, Jiang, H. Physics-informed neural networks for efficient aeroacoustic computational based on two time sequence prediction models. InterNoise, 2024.

  2. 2026

    Zhao Y, Deng Y, Lu H, Jiang H. Deep-Learning-Based Impedance Eduction Method for Acoustic Liners with Flow. AIAA Journal, 2026, 64(3): 1399-1413.

    ↗
  3. 2022

    Chen W, Jiang H*, Huang X. Super-resolution acoustic imaging. Applied Physics Letters, 2022, 120(11): 112201.

    ↗
  4. 2018

    Jiang H, Lau A, Huang X*. Surrogate model based liner optimization for aeroengines and comparison with finite elements. Journal of Vibration and Acoustics, 2018, 140(3): 034501.

    ↗
  5. 2022

    Chen W, Jiang H*, He W. Dipole source-based virtual three-dimensional imaging for propeller noise. Aerospace Science and Technology, 2022, 124: 107562.

    ↗
  6. 2022

    Jiang H, Zhong S, Zhang X*, Huang X. Boundary element analysis of the fuselage scattering of drone noise. International Journal of Aeroacoustics, 2022, 21(1-2): 43-56.

    ↗

04 · 并行计算与低马赫数方法Parallel Computing & Low-Mach Methods

如何提高求解效率,同时保留低马赫数下的声学信息?How can faster solvers retain acoustic information at low Mach numbers?

研究方法Approach

一类工作将 Wiener–Hopf 求解重写为适于 GPU 处理的矩阵运算;另一类通过圆柱绕流、阵风–翼型和湍流–翼型算例,对比不可压缩、弱可压缩与可压缩模型,并检查 FW–H 远场预测。One study reformulates Wiener–Hopf calculations as GPU-suitable matrix operations. Separate studies compare incompressible, weakly compressible and compressible models for cylinder flow and gust– or turbulence–airfoil interactions, including FW–H far-field predictions.

研究目的Purpose

区分算法加速与模型适用性:准确的升阻力预测并不自动保证正确的声传播预测。Separate acceleration from model validity: accurate aerodynamic forces do not automatically imply accurate acoustic propagation.

对应论文 · 7 篇Related publications · 7
  1. 2026

    Wang Q, Shao M, Jiang H. Multi-DCU parallelization of the FW–H solver for aeroacoustic simulations. Computer Physics Communications, 2026, 327: 110284.

    ↗
  2. 2025

    Wang Q, Duan L, Jiang H. On the prediction of flow-induced noise at low Mach numbers: A comparative study of compressible, weakly compressible, and incompressible models [J]. Physics of Fluids, 2025, 37(12): 126138.

    ↗
  3. 2025

    Wang Q, Shao M, Duan L, et al. Effect of compressibility on flow-induced noise predictions at low Mach numbers [J]. Physics of Fluids, 2025, 37(12): 126128.

    ↗
  4. 2022

    Jiang H, Zhang X*. An acoustic-wave preserved artificial compressibility method for low-Mach-number aeroacoustic simulations. Journal of Sound and Vibration, 2022, 516: 116505.

    ↗
  5. 2017

    Jiang H, Lau A, Huang X*. An efficient algorithm of Wiener-Hopf method with graphics processing unit for duct acoustics. Journal of Vibration and Acoustics, 2017, 139(5): 054501.

    ↗
  6. 2024

    Deng, Y, Wang, Z, Zhou, T, Wu, J, and Jiang H*. Boundary-element analysis of the noise scattering for urban aerial mobility vehicles: Solver development and assessment. ASME. Journal of Vibration and Acoustics. 2024, 146(5): 051001.

    ↗
  7. 2021

    Liu Q, Liu Y, Jiang H, Yang Y, Zhou P. Effect of a poroelastic extension on acoustic scattering of a finite plate. AIAA Aviation Forum, 2021.

OPEN-SOURCE SOFTWARE

让研究方法,成为可用的工具。Research methods, ready to use.

分享实验室开发的开源计算软件,提供源码、使用文档与相关研究入口。Explore the laboratory’s open-source solvers, with source code, documentation and links to related research.

“

把复杂的流动,转化为可理解、可预测、可设计的声音。Turning complex flows into sound we can understand, predict and design.

论文摘要PUBLICATION ABSTRACT