从非定常流动，计算远场声音。From unsteady flow to far-field sound.
将气动声学方法转化为可使用的软件：基于 FW-H 方程预测非定常流动产生的声辐射。Turning aeroacoustic methods into usable software, with FW-H prediction of sound radiated by unsteady flows.
了解软件与使用入口Explore the software ↗01研究RESEARCH
于流动深处，探声音之源。 Within the flow, we seek the source of sound.
围绕低空飞行器与推进噪声，我们结合高保真计算、复杂流动机理与气动声学实验，研究声音的产生、传播和控制。We study low-altitude aircraft and propulsion noise through computational methods, complex-flow mechanisms and aeroacoustic experiments, examining how sound is generated, propagates and can be controlled.
研究方向RESEARCH DIRECTION
高保真气动声学计算方法High-fidelity Computational Aeroacoustics
为更可靠地预测流动噪声，研究周期声学响应的高效求解、FW–H 声学积分的并行计算，以及声传播与散射的数值方法。方法研究同时关注计算精度、成本和适用条件；降阶与数据驱动方法作为直接求解的补充。We develop methods for more reliable flow-noise prediction, including efficient solvers for periodic acoustic response, parallel FW–H integration and numerical treatments of propagation and scattering. Accuracy, computational cost and model validity are considered together, with reduced-order and data-driven methods complementing direct simulation.
研究图像RESEARCH IMAGES
从图像观察流动与声场Flow and sound fields in images
动态可视化IN MOTION
无人机流场计算动画UAV flow simulation animation
精选论文SELECTED PUBLICATIONS
从论文了解这一方向Explore the work through selected papers
选取与本方向相关的近期成果；完整论文可在成果页面检索。Recent papers relevant to this direction. Find the full list under Publications.
- 2026He L, Jiang H. Efficient Frequency-Domain Aeroacoustic Simulation of Gust Response Using Harmonic Balance Method. AIAA Journal, 2026, 64(7): 3869-3883. ↗
基于谐波平衡法的高效频域阵风响应仿真Efficient Frequency-Domain Aeroacoustic Simulation of Gust Response Using Harmonic Balance Method
He L, Jiang H. Efficient Frequency-Domain Aeroacoustic Simulation of Gust Response Using Harmonic Balance Method. AIAA Journal, 2026, 64(7): 3869-3883.
查看原文View publication ↗采用传统时域方法求解周期非定常流动时，达到最终周期状态之前的长过渡过程常导致计算效率较低。本文通过实现对流单极子声源、频域阵风激励及无反射边界条件，将谐波平衡方法用于气动声学。测试包括均匀背景流中单极子声源、翼型俯仰振荡、阵风—翼型相互作用前缘噪声及阵风—叶栅响应。结果与解析及实验数据吻合良好，多数算例中的归一化均方根误差通常低于 10%。研究展示了谐波平衡方法在流体力学和气动声学中的广泛应用潜力。
Periodic unsteady flow problems often exhibit computational inefficiency when conventional time-domain methods are employed. This phenomenon can be attributed to the lengthy transition phase before reaching the ultimate periodic state. To address this problem, the harmonic balance method (HBM) is employed for aeroacoustic applications through the implementation of convective monopole sources, frequency-domain gust excitation, and nonreflective boundary conditions. A set of test cases is considered to evaluate this method, including monopole sources with uniform background flows, the pitching oscillations of airfoils, leading-edge noise of gust–airfoil interactions, and the response of gust–cascade interactions. The results show good agreement with analytical and experimental data, with normalized root-mean-square errors typically below 10% across most considered cases. The findings from this research demonstrate a comprehensive range of potential applications for HBM in the fields of fluid dynamics and aeroacoustics.
- 2026Wang Q, Shao M, Jiang H. Multi-DCU parallelization of the FW–H solver for aeroacoustic simulations. Computer Physics Communications, 2026, 327: 110284. ↗
多 DCU 并行 FW-H 气动声学求解器Multi-DCU parallelization of the FW–H solver for aeroacoustic simulations
Wang Q, Shao M, Jiang H. Multi-DCU parallelization of the FW–H solver for aeroacoustic simulations. Computer Physics Communications, 2026, 327: 110284.
查看原文View publication ↗FW–H 声类比广泛用于噪声预测，但大积分面和长声源时间历程会使后处理计算成本很高。本文面向配备深度计算单元（DCU，即 GPU 类加速器）的平台，提出基于 MPI 的多节点 FW–H 求解器。主机 CPU 负责数据准备、节点间通信和输入输出，主要计算量所在的 FW–H 积分卸载至 DCU。观察点、时间或频率样本及面元上的嵌套循环映射到三维线程网格，并通过分布式域分解减少各进程内存占用。典型二维和三维基准验证中，远场指向性的平均相对误差分别低于 0.15% 和 0.70%。在代表性二维频域基准中，相比经过优化的 32 核海光 C86 7185 CPU 实现，单张海光 DCU 的端到端加速最高达 21.2 倍，积分内核最高达 708 倍。该二维算例的 CPU 实现在较低至中等核数下近线性扩展，但接近 32 核时出现饱和；大型三维时域基准中，CPU 和 DCU 分别扩展至 128 核和 128 张 DCU 时仍表现良好，DCU 接近线性，部分配置达到超线性加速。分析表明，CPU 饱和主要与 MPI 开销和节点内资源竞争增加有关；DCU 超线性加速主要来自单设备任务减少后分块次数及主机—设备传输开销降低。基于额定功率的估算还表明，DCU 配置完成求解的能耗低于相应 CPU 配置。该实现为气动声学预测提供了准确、高效且可扩展的 FW–H 计算框架。
The Ffowcs Williams–Hawkings (FW–H) acoustic analogy is widely used for noise prediction, but large integration surfaces and long source-time histories can make the post-processing computationally expensive. To address this issue, we present a Message Passing Interface (MPI)-based multi-node implementation of the FW–H solver for platforms equipped with Deep Computing Units (DCUs), i.e., GPU-class accelerators. In this framework, host CPUs handle data preparation, inter-node communication, and I/O, while the computationally dominant FW–H integration is offloaded to the DCUs. The nested loops over observers, time/frequency samples, and surface elements are mapped onto a three-dimensional thread grid, and distributed domain decomposition is adopted to reduce the memory footprint on each rank. The implementation is validated on canonical two-dimensional and three-dimensional benchmarks, with mean relative errors in far-field directivity below 0.15% and 0.70%, respectively. For a representative 2D frequency-domain benchmark, a single Hygon DCU achieves an end-to-end speedup of up to 21.2 × over an optimized 32-core Hygon C86 7185 CPU implementation, and up to 708 × for the FW–H integration kernel. In this 2D case, the CPU implementation scales nearly linearly at low to moderate core counts but shows signs of saturation near 32 cores, whereas for the large 3D time-domain benchmark, both CPU and DCU implementations exhibit robust multi-node scalability up to 128 CPU cores and 128 DCUs, with the DCU path showing near-linear scaling and, in some configurations, superlinear speedup. Scalability analysis indicates that the CPU saturation trend is mainly associated with increasing MPI overhead and intra-node resource contention, whereas the superlinear DCU speedup arises primarily from reduced chunking and lower host–device transfer overhead as the per-device workload decreases. A rated-power-based estimate also suggests lower energy-to-solution for the DCU configurations than for the corresponding CPU configurations. Overall, the proposed multi-DCU implementation provides an accurate, efficient, and scalable framework for FW–H-based aeroacoustic prediction.
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.
“