A method for predicting the propagation of urban noise for a rotorcraft

Abstract

The application discloses a rotorcraft urban noise propagation prediction method and rotorcraft aerodynamic acoustics and environmental noise evaluation and prediction technology, and relates to the technical field of rotorcraft urban noise propagation prediction, and comprises the following steps: acquiring sound pressure level spectrum data at a preset observation point in the near field of a rotorcraft; reconstructing an equivalent point sound source model containing multipole characteristics based on the sound pressure level spectrum data; establishing a macro acoustic propagation model containing the geometry of urban buildings, and configuring the equivalent point sound source model at a preset target position in the macro acoustic propagation model; constructing a finite element calculation region around the equivalent point sound source at the target position, and taking the outer surface of the finite element calculation region as a coupling surface; calculating the near-field acoustic response inside the finite element calculation region, and solving the far-field sound pressure distribution and spectrum containing the reflection condition of the building through the coupling surface by combining the boundary element method. The method can strictly consider the reflection and diffraction of buildings while reducing the calculation cost under the premise of ensuring accurate extraction of aerodynamic sound sources during noise prediction.

Description

Technical Field

[0001] This invention relates to the field of aeroacoustics and environmental noise assessment and prediction technology for aircraft, and more specifically to a method for predicting urban noise propagation of rotorcraft. Background Technology

[0002] With the rapid development of Urban Air Mobility (UAM), urban rotorcraft such as electric vertical takeoff and landing (eVTOL) aircraft and drones have been widely used in emergency rescue, logistics delivery, and other fields due to their advantages of vertical takeoff and landing and high maneuverability. However, the resulting rotational noise problems, such as thickness noise and load noise, are becoming increasingly prominent. Unlike the open high-altitude flight environment, the dense building complexes in urban environments cause strong reflections, diffractions, and multipath propagation effects during sound wave propagation. Therefore, noise generated by rotorcraft has a significant negative impact on the surrounding ecological environment and public health. To mitigate noise pollution and promote industrial development, it is essential to have the technical means to efficiently and accurately assess the noise distribution of a large number of rotorcraft in complex urban building complexes. This is not only an urgent engineering need but also an indispensable technical foundation for developing noise control strategies. Therefore, how to accurately predict rotor noise, reasonably consider the impact of urban buildings, and strike a balance between prediction accuracy and computational cost has become a pressing engineering challenge.

[0003] Currently, the main methods for predicting aerodynamic noise of aircraft include the following, but all of them have significant limitations in application to complex urban scenarios:

[0004] First, directly solving the compressible Navier-Stokes equations (CAA direct noise method) offers ideal physical consistency, naturally incorporating nonlinear and near-field coupling effects. However, this method requires sound source generation and sound propagation to occur within the same computational domain, placing extremely stringent demands on the acoustic mesh and time step. When dealing with rotors and macroscopic urban-scale spaces, the computational load is extremely high, making it difficult to cover large-scale building clusters and far-field calculations in practical engineering.

[0005] Secondly, the Field-Warrant-Hill (FW-H) analogy method, which extracts rotor noise sources by solving the FW-H equations, is a relatively mature and computationally low-cost prediction method. However, traditional fluid simulation software, such as ANSYS Fluent's FW-H integral acoustic model, can typically only accurately calculate aerodynamic noise in free-field sound or under simple reflecting surfaces. The core problem lies in the fact that the integral solution process is highly dependent on the exact Green's function. Since there is no exact Green's function for complex urban building clusters, this method cannot rigorously and accurately handle the complex physical phenomena such as multiple reflections, diffraction, scattering, and absorption of sound waves in densely populated urban spaces.

[0006] Furthermore, the control surface acoustic source combined with the finite element / boundary element method (FEM / BEM) constructs an equivalent acoustic boundary by extracting unsteady flow field data at the rotor rotation domain boundary, which can rigorously consider building reflections and diffractions. However, its difficulty lies in the fact that the selection of the control surface location is extremely sensitive, and under high-frequency broadband noise prediction, the mesh size and computational cost of the boundary element (BEM) are enormous.

[0007] Furthermore, existing urban environmental noise prediction tools (such as CadnaA) are mostly based on semi-empirical formulas for point or line sound sources, which cannot accurately reproduce the complex three-dimensional transient dipole sound source generated by the high-speed rotation of UAV propellers, i.e., they cannot effectively reproduce the coupling of thickness noise and load noise. At the same time, there is a lack of an efficient and high-fidelity data interface in the conversion process from transient time-domain flow field to steady-state frequency-domain sound field.

[0008] Therefore, in noise prediction, how to balance the three major requirements of accurately extracting aerodynamic sound sources, rigorously calculating the physical diffraction of complex urban environments, and controlling engineering calculation costs is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0009] In view of the above problems, the present invention proposes a method for predicting urban noise propagation of rotary-wing aircraft, which aims to overcome or at least partially solve the above problems.

[0010] To achieve the above objectives, the present invention adopts the following technical solution: This invention provides a method for predicting urban noise propagation from rotary-wing aircraft, comprising: Acquire the sound pressure level spectrum data at preset observation points in the near-field region of the rotorcraft; Based on the sound pressure level spectrum data, an equivalent point source model containing multipole characteristics is reconstructed, and the sound power spectrum of the equivalent point source is determined. A macroscopic acoustic propagation model incorporating the geometry of urban buildings is established, and the equivalent point source model is positioned at a preset target location within the macroscopic acoustic propagation model. A finite element calculation region is constructed around the equivalent point sound source at the target location, and the outer surface of the finite element calculation region is used as the coupling surface with the boundary element calculation region. Using the finite element method, the near-field acoustic response within the finite element calculation region is calculated based on the sound power spectrum. Combined with the boundary element method, the far-field sound pressure distribution and spectrum, including building reflection conditions, are solved through the coupling surface.

[0011] Furthermore, the acquisition of sound pressure level spectrum data at a preset observation point within the near-field region of the rotorcraft specifically includes: A local fluid computational domain for a rotorcraft is established, and the transient flow field of the local fluid computational domain is solved using the large eddy simulation method. During the solution process, multiple observation points at different angles are set in the three-dimensional space of the rotorcraft within the local fluid computational domain. Based on the FW-H acoustic analogy equation, the sound pressure level spectrum at each observation point is extracted; the number of observation points is no less than four. The frequencies corresponding to the sound pressure level spectrum at each observation point are subjected to cubic spline interpolation based on the logarithmic frequency axis to unify the frequency axis of different observation points.

[0012] Furthermore, the reconstructing of an equivalent point source model incorporating multipole characteristics based on the sound pressure level spectrum data specifically includes: The equivalent point sound source is set as an axisymmetric structure, and its sound field is composed of monopoles, dipoles and quadrupoles. For each frequency point in the sound pressure level spectrum data, the equivalent point source at that frequency is set at any observation polar angle. The square of the total sound pressure level in the direction and the sound field at that frequency satisfy the following preset conditions:

[0013] in, To observe the polar angle The squared value of the total sound pressure level measured in the direction; It is a monopole directionality function, and ; Monopole strength; It is the dipole directionality function, and ; The strength of the dipole; It is the quadrupole directionality function, and ; The quadrupole strength.

[0014] Furthermore, determining the sound power spectrum of the equivalent point sound source specifically includes: For each frequency point in the sound pressure level spectrum data, the total squared sound pressure value at that frequency point under at least four observation polar angles is substituted into the formula of the preset condition to establish a linear equation system, and the monopole intensity, dipole intensity and quadrupole intensity at that frequency point are analytically solved. Based on the source intensity results at each frequency point obtained by the solution, and combined with the conversion relationship between sound intensity and sound power, the sound power spectrum of the equivalent point sound source as a function of frequency is calculated.

[0015] Furthermore, the sound pressure level spectrum data is derived from a scaled-down rotor model. Before establishing a macroscopic acoustic propagation model that includes the geometry of urban buildings, the process further includes: determining the amplification ratio based on the characteristic lengths of the rotorcraft and the scaled-down model; determining the scaling correction coefficient based on the rotor aeroacoustic similarity criterion of the blade tip velocity under the same rotational speed conditions; and amplifying the sound power spectrum of the equivalent point sound source by the sixth power of the amplification ratio.

[0016] Furthermore, the finite element calculation region is a spherical spatial region that encloses the equivalent point sound source.

[0017] Furthermore, in the finite element calculation region, the acoustic power spectrum is used as the input excitation of the equivalent point sound source to calculate the near-field acoustic response.

[0018] Furthermore, in the boundary element calculation region, the surfaces of urban buildings are set as rigid reflective boundary conditions or symmetric boundary conditions to calculate the reflection and diffraction effects of sound waves between buildings.

[0019] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method for predicting urban noise propagation of rotary-wing aircraft, which has the following beneficial effects: This invention decouples the extraction of microscopic high-fidelity aerodynamic sound sources from macroscopic large-scale urban sound wave propagation at the physical scale. It employs a workflow of "FW-H calculation of the near field + equivalent sound source reconstruction + FEM / BEM surface coupling to solve the far field," effectively avoiding the direct generation of high-density fluid meshes in large-scale spaces. Compared to methods that directly solve compressible fluid dynamics (CAA), this invention significantly reduces computational costs in urban-level noise assessment. Simultaneously, by using a multipole equivalent sound source, it accurately preserves the directivity and spectral characteristics of the sound source, ensuring prediction accuracy.

[0020] Traditional FW-H acoustic analogy methods rely on precise Green's functions, which are difficult to apply rigorously in real urban environments with complex building clusters and reflections and diffractions. This invention introduces the boundary element method (BEM) to process urban buildings, setting the building surfaces as rigid reflective or symmetrical boundary conditions. This enables accurate calculation of the reflection and diffraction effects of sound waves between building clusters, accurately reproducing the complex physical phenomena of urban sound propagation, and filling the technical gap of traditional methods in complex geometric environments.

[0021] This invention uses CFD and FW-H as front-end data extraction tools to reconstruct complex blade rotation boundaries into equivalent multipole point sound sources (including monopoles, dipoles, and quadrupoles) with high-fidelity directivity characteristics through rigorous analytical equations. By extracting far-field spectral point data from specific azimuths and working in conjunction with the reconstructed equivalent point sound source model, analytical dimensionality reduction from complex three-dimensional rotating sound sources to equivalent point sound sources is achieved, significantly reducing the model complexity of subsequent sound propagation calculations.

[0022] This invention establishes a specific frequency interpolation mapping mechanism to unify the frequency axis processing of sound pressure level spectra from different observation points, effectively solving the data distortion problem caused by the different scales of the fluid computational domain grid and the acoustic propagation domain grid, and providing a reliable data interface for the transfer of high-fidelity sound source data to the acoustic model. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the process for predicting urban noise propagation of rotary-wing aircraft provided in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the local fluid computational domain and geometric model of a rotorcraft provided in an embodiment of the present invention; Figure 3 This is a comparison diagram of the original sound pressure level spectrum extracted based on the FW-H equation and the reference frequency axis spectrum after cubic spline interpolation based on the logarithmic frequency axis, provided in an embodiment of the present invention. Figure 4 This is a comparison and verification diagram of the original aerodynamic spectrum and the equivalent acoustic power spectrum reconstructed using multipole provided in the embodiments of the present invention; Figure 5 This is a schematic diagram of the macroscopic acoustic calculation model provided in the embodiments of the present invention; Figure 6 This is a spatial distribution cloud map of urban sound pressure and a schematic diagram of the total sound pressure level spectrum at a specific observation point, provided in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] This invention discloses a method for predicting urban noise propagation from rotary-wing aircraft, such as... Figure 1 As shown, it includes the following steps: S1. Obtain the sound pressure level spectrum data at a preset observation point in the near-field region of the rotorcraft; S2. Based on the sound pressure level spectrum data, reconstruct an equivalent point source model containing multipole characteristics, and determine the sound power spectrum of the equivalent point source. S3. Establish a macroscopic acoustic propagation model that includes the geometry of urban buildings, and place the equivalent point source model at the preset target position in the macroscopic acoustic propagation model; S4. Construct a finite element calculation region around the equivalent point sound source at the target location, and use the outer surface of the finite element calculation region as the coupling surface with the boundary element calculation region. S5. Using the finite element method, the near-field acoustic response within the finite element calculation region is calculated based on the sound power spectrum. Combined with the boundary element method, the far-field sound pressure distribution and spectrum including building reflection conditions are solved through the coupling surface.

[0028] The overall model of this invention is easily extended to large-scale urban-level noise assessment and prediction, and can support rapid scanning analysis of multiple operating conditions and multiple flight parameters. It has strong engineering feasibility and is suitable for noise prediction and flight path planning of rotorcraft such as UAVs and eVTOLs in urban environments.

[0029] Next, each of the above steps will be explained in detail.

[0030] In step S1 above, the sound pressure level spectrum data at a preset observation point in the near field region of the rotorcraft is obtained. This sound pressure level spectrum data is used to characterize the aerodynamic noise characteristics of the rotorcraft. In this embodiment, a fluid dynamics solver (such as ANSYS Fluent) is used to calculate the aerodynamic noise source. Figure 2 As shown, a local fluid computational domain and geometric model of a rotating machine (such as a rotorcraft propeller) are established. A Phantom 9450 propeller model is used, with a diameter of... With pitch The computational domain should be set to ensure the full development of the wake flow, for example, 3.5D upstream, 6.7D downstream, and 3D laterally.

[0031] The computational domain is spatially discretized and meshed: the rotating domain uses a Polyhexcore mesh, and the stationary domain uses a hexahedral structured mesh. Boundary conditions are set as follows: the inlet is a pressure inlet. Exports are pressure exports The sidewalls are designed with slip walls to simulate free flow conditions.

[0032] After setting the boundary conditions in the fluid dynamics solver, steady-state and transient flow field calculations are performed. The calculation process includes the following two stages: (1) Steady-state flow field calculation: The rotating domain containing the propeller is set as a multiple reference frame (MRF), and the stationary domain maintains a fixed boundary. The Reynolds-averaged Navier-Stokes (RANS) equation is used in combination with the SIMPLE pressure-velocity coupling algorithm for steady-state iteration until the residuals of each variable in the flow field drop to a set threshold (e.g., below 1e-4), thus obtaining a converged initial flow field, which provides a stable initial field for transient calculation.

[0033] (2) Transient acoustic calculation: To accurately capture the small-scale eddy structure that generates broadband load noise in the blade wake, the turbulence model was switched to Large Eddy Simulation (LES), and the WALE subgrid scale model was selected; the unsteady PISO algorithm was used for pressure-velocity decoupling solution. In this stage, to meet the analytical requirements of high-frequency aerodynamic noise and follow the Nyquist sampling theorem, the physical time step was strictly set according to the highest target analysis frequency and rotor speed to ensure that the transient flow field characteristics were not distorted. At the same time, the FW-H acoustic integration module was opened to collect noise data: In the transient calculation process, the blade surface other than the shaft was selected as the source surface; and multiple acoustic virtual observation points with specific orientations were preset in three-dimensional space, with no less than four observation points; for example, the distance from the source radius place axial direction and angle with the axis of rotation horn Acoustic observation points were set up at specific locations to extract unsteady surface pressures that vary over time. For the time-domain signal obtained from the FW-H method, a Fourier transform was first performed. Equivalent point sources were established at each discrete frequency point, and the Helmholtz equations for the corresponding frequencies were solved. Since the frequency-domain complex sound pressure level contains both amplitude and phase information, phase interference, reflection, diffraction, and multipath superposition generated during sound wave propagation are naturally reflected in the complex sound pressure level solution. Finally, the broadband sound field and total sound pressure level distribution were obtained based on the principle of spectral superposition.

[0034] Because subtle differences in the relative distance between the sound source and the observation point during the integral calculation of the FW-H method can cause frequency shifts, cubic spline interpolation based on the logarithmic frequency axis is performed on the extracted sound pressure level spectrum at each observation point to unify the common frequency axis. A comparison of the original sound pressure level spectrum extracted based on the FW-H equation and the reference frequency axis spectrum after cubic spline interpolation is shown in the figure below. Figure 3 As shown, this figure presents an interpolated sound pressure level (SPL) response spectrum based on a common logarithmic frequency axis, aiming to quantitatively characterize the acoustic radiation or frequency response characteristics of the system on multiple orthogonal and specific spatial directions (+z, +x, +y axes and the 45° direction of x=z). The horizontal axis of the graph is defined as frequency (unit: Hz), using a logarithmic scale covering 10 Hz to 10 Hz. 4 The broadband acoustic analysis range is defined by the Hz band; the vertical axis represents the interpolated absolute sound pressure level (dB, reference standard air sound pressure 20 µPa), with a linear span set from -50 dB to 40 dB. In-depth observation of the morphological characteristics of the response spectrum in each direction reveals that in the low-frequency band (approximately below 200 Hz), the system exhibits significant discrete modes and low-frequency resonance characteristics, with several highly concentrated resonance peaks. Particularly near 100 Hz, the sound pressure level response along the +y axis and in the 45° direction reaches a global maximum, with peak values ​​approaching 30 dB; in contrast, the radiated energy in the +z direction in this low-frequency range is relatively weak and in the negative range. As the analysis frequency moves into the mid-to-high frequency region (above 200 Hz), the modal density of the system increases sharply, resulting in a spectral curve exhibiting severe sawtooth high-frequency oscillations and complex spatial interference patterns. Furthermore, the overall acoustic radiation energy of the system shows a macroscopic attenuation trend with increasing frequency, decreasing towards 10 dB. 4 At the high-frequency end of Hz, the sound pressure level in each direction has generally attenuated to the low-energy band of 0 dB to -50 dB, and the response curves of different spatial dimensions show extremely significant sound field anisotropy and spatial distribution differences throughout the entire wideband.

[0035] In step S2 above, based on the sound pressure level spectrum data, an equivalent point source model containing multipole characteristics is reconstructed, and the sound power spectrum of the equivalent point source is determined. Since macroscopic urban acoustic calculations cannot handle meshes with detailed blade geometry, the fluid results obtained above need to be equivalently reconstructed into an ideal multipole point sound source.

[0036] Assuming the aerodynamic noise source has axisymmetric characteristics and its sound field is mainly composed of monopoles, ), Dipole ) and Quadrupole ) Composition.

[0037] Define the angular directionality function of each order of multipole: for the observed polar angle (from (axis measurement), monopole directionality function Dipole Directionality Function Quadrupole Directionality Function .

[0038] For each frequency point in the sound pressure level spectrum data, the equivalent point source at that frequency is set at an arbitrary observation polar angle. The square of the total sound pressure level in the direction and the sound field at that frequency satisfy the following preset conditions:

[0039] in, To observe the polar angle The squared value of the total sound pressure level measured in the direction; Monopole strength; The strength of the dipole; The quadrupole strength.

[0040] Due to the directional function of the monopole Therefore, this preset condition can also be simplified to:

[0041] For each frequency point in the sound pressure level spectrum data, substitute the total squared sound pressure value at that frequency point under at least four observation polar angles into the formula of the preset conditions to establish a system of linear equations, and analytically solve for the monopole intensity, dipole intensity, and quadrupole intensity at that frequency point; for example, substitute into... (+) Direction, measured ), (+ Radial, measured )as well as (measured) The sound pressure level data at () is obtained. The full-rank linear equations are solved analytically to obtain the intensity of each multipole. .

[0042] Based on the source intensity results obtained at each frequency point, and combined with the conversion relationship between sound intensity and sound power, the sound power spectrum of the equivalent point sound source as a function of frequency is calculated; specifically, based on the spherical average direction factor, using the formula... Convert the source strength into the acoustic power spectrum of that frequency band.

[0043] The comparison and verification diagram between the original aerodynamic spectrum and the equivalent acoustic power spectrum reconstructed using multipole is shown in the figure below. Figure 4As shown, this figure is a spectrum of sound pressure level (SPL) response in the logarithmic frequency domain, aiming to qualitatively and quantitatively compare the actual measured values ​​(solid lines) and the algorithmically reconstructed values ​​(dashed lines) of SPL in multiple spatial directions (x, y, z axes and 45° direction). The horizontal axis of the graph is defined as frequency (unit: Hz), using a logarithmic scale covering 10 Hz to 10 Hz. 4 The analysis range is a wideband of Hz; the vertical axis represents the sound pressure level (in dB, with a reference standard air sound pressure of 20 µPa), with a linear span from -50 dB to 40 dB. Observation of the spectral distribution characteristics reveals that in the low-frequency band (below approximately 200 Hz), the system response exhibits obvious discrete modal characteristics, with several significant low-frequency resonance peaks. Particularly near 100 Hz, the sound pressure level reaches global maxima in the y-axis and 45° direction, approaching 30 dB. Notably, within this low-frequency range, the reconstruction curves in all directions highly coincide with the actual measured curves, strongly demonstrating the excellent prediction and reconstruction accuracy of the reconstruction model for low-frequency sound field characteristics. However, as the analysis frequency moves into the mid-to-high frequency region (above 200 Hz), the modal density of the system increases sharply, resulting in a spectral curve exhibiting severe sawtooth oscillations and complex spatial interference patterns. Furthermore, the overall acoustic radiation energy of the system shows a macroscopic attenuation trend with increasing frequency, decreasing towards 10 dB. 4 At the high-frequency end of Hz, the sound pressure levels in each direction are generally distributed in the low-energy range of 0 dB to -50 dB, and the response curves of different spatial dimensions show significant anisotropy and spatial distribution differences in a wide frequency band.

[0044] When the sound pressure level spectrum data in step S1 comes from a scaled-down rotor model, sound source amplification estimation is required to predict the noise of large-scale aircraft in cities. That is, before establishing a macroscopic acoustic propagation model that includes the geometry of urban buildings, the amplification ratio is determined based on the characteristic lengths of the rotorcraft and the scaled-down model (e.g., amplification ratio scale = 50 times, down to the 10m level). Under the same rotational speed, the blade tip velocity is proportional to the characteristic length. A scaling correction coefficient is determined based on the rotor aeroacoustic similarity criterion, such as the sixth power law of blade tip velocity. The sound power spectrum of the equivalent point source is amplified by the sixth power of the amplification ratio (scale^6) to serve as the final equivalent sound source input to the urban acoustic model.

[0045] In step S3 above, a macroscopic acoustic propagation model incorporating the geometry of urban buildings is established, and the equivalent point source model is positioned at a preset target location within the macroscopic acoustic propagation model; specifically: A macroscopic acoustic propagation model is established using a combined acoustic finite element / boundary element solver (such as COMSOL), incorporating the geometry of urban buildings, equivalent sound sources, and the FEM / BEM coupling surface. A schematic diagram of this macroscopic acoustic propagation model is shown below. Figure 5As shown in the figure, this image presents a visualization of a three-dimensional numerical simulation of the physical field radiation characteristics (presumably aerodynamic noise or sound pressure level distribution) of a quadcopter UAV under complex geometric boundary conditions. The spatial computational domain is established in a three-dimensional Cartesian coordinate system (x, y, z) with meters (m) as the unit of measurement, and the spatial scale of the horizontal plane ranges approximately from -200 meters to 200 meters. In the simulation environment, multiple rectangular cylinders and a centrally located large cylindrical structure are regularly distributed on the ground to represent typical urban building complexes or industrial facility terrain. The image uses multi-planar slicing (including bottom projection and multiple orthogonal vertical sections) combined with pseudo-color mapping technology to visualize the three-dimensional spatial distribution of the scalar field. The figure clearly and discretely labels the decibel (dB) values ​​(e.g., 40.6 dB, 37.1 dB, and 16.9 dB) at several spatial probe points, confirming that the evaluated physical quantities have logarithmic energy decay characteristics. A closer look at the topological distribution characteristics of the field variables reveals a distinct outward-expanding arc-shaped wavefront beneath the UAV radiation source. High-intensity energy concentration areas (exhibiting warm red and yellow hues) are primarily located in channels between structures and in open areas near the source. With increasing spatial propagation distance and the intervention of geometric obstacles, the amplitude of the field variables attenuates significantly, resulting in a low-energy distribution (exhibiting cool blue and green hues) in the "sound shadow zone" far from the sound source and on the leeward side of buildings. This spatial energy distribution pattern intuitively and profoundly reveals the complex acoustic propagation dynamics mechanisms arising from the interaction of radiated waves with complex three-dimensional topological obstacles, including spatial divergence, multipath reflection, edge diffraction, and geometric attenuation.

[0046] Next, the equivalent point source model is positioned at the preset target location (e.g., coordinates) in the macroscopic acoustic propagation model. ).

[0047] In step S4 above, a finite element calculation region is constructed around the equivalent point sound source at the target location, and the outer surface of the finite element calculation region is used as the coupling surface with the boundary element calculation region. like Figure 5 As shown, a small-sized spherical medium (e.g., with a radius of...) is wrapped around the equivalent point sound source. The interior of the sphere is used as the computational region for the pressure acoustic finite element method (FEM). The surface of the sphere is defined as the coupling surface between the finite element method and the boundary element method (BEM).

[0048] In the finite element method (FEM) computational domain, the amplified sound power spectrum is used as the input excitation of the equivalent point sound source, and the near-field acoustic response is calculated using the FEM method. The COMSOL-based pressure acoustic frequency method discussed here refers to a numerical analysis method for solving the scalar sound pressure field in the frequency domain for sound wave propagation problems in a stationary background fluid or where the macroscopic flow velocity of the fluid is assumed to be much less than the speed of sound. This method differs from the transient wave equation in the time domain; it assumes that the excitation experienced by the sound field system and the resulting sound pressure field are both steady-state responses that vary harmonicly (sinusoidally) with time. By solving for the complex form of the sound pressure value at the nodes of the finite element or boundary element mesh, this method can efficiently evaluate the radiation, scattering, diffraction, and transmission characteristics of mechanical vibration sources, fluid sound sources, etc., under complex propagation paths. The core dynamic model of the pressure acoustic frequency method lies in solving the non-homogeneous Helmholtz equation, which presupposes that the functional form of the sound pressure changing with time is defined as follows: ,in, Indicates spatial location At the moment The instantaneous sound pressure, measured in Pascals (Pa), reflects the pressure fluctuation at a point in the sound field over time. In the formula... The sound pressure complex amplitude (or complex sound pressure) describes the spatial distribution characteristics of sound pressure, containing both amplitude and phase information. Therefore, it can completely characterize the spatial variation of sound waves during propagation. (Exponential term) This is the time harmonic factor, used to describe the characteristic of sound pressure oscillating periodically over time. Among them, The imaginary unit satisfies Representing the vibration process in complex exponential form simplifies the acoustic wave equation and related calculations; actual physical sound pressure levels are typically represented by the real part of its complex expression. Parameters It represents time, with the unit being seconds (s), and is used to characterize the time process of sound wave propagation and vibration. (Parameters) Angular frequency is the frequency of a sound wave, measured in radians per second (rad / s). It represents the speed at which the sound wave vibrates and is related to its frequency. The relationship is A higher angular frequency indicates that the sound wave completes more vibration cycles per unit time, and thus a higher frequency. The unit is Hertz (Hz), which represents the number of vibrations per second of a sound wave. It is an important parameter that determines the pitch of a sound and is governed by the non-homogeneous Helmholtz equation. The specific derivation and definition of the three-dimensional spatial control equation are as follows:

[0049] in, Represents the complex sound pressure level in the acoustic propagation medium (unit: Pa); The complex equivalent density of a fluid is used to account for energy loss and damping caused by viscosity or thermal conduction within a system (unit: kg / m³). The complex equivalent sound velocity of the fluid (in m / s); The angular frequency of the external excitation; This is the equivalent volume sound source term in the frequency domain (such as a monopole sound source).

[0050] In step S5 above, the near-field acoustic response within the finite element calculation region is calculated based on the sound power spectrum using the finite element method, and the far-field sound pressure distribution and spectrum including building reflection conditions are solved by the coupled surface using the boundary element method.

[0051] The surfaces of urban buildings are set as rigid or symmetrical boundary conditions to calculate the reflection and diffraction effects of sound waves between buildings. Specifically, in the BEM region, building surfaces and the ground are equivalent to rigid reflective surfaces or corresponding acoustic impedance boundaries are set. Symmetrical surfaces are used to simplify the model (the governing equations of symmetrical surfaces are the same as those of rigid boundaries). The acoustic response of near-field excitation is calculated by FEM, and BEM is responsible for solving the far-field grid nodes that include building reflection and diffraction effects. Finally, the sound pressure distribution cloud map in the urban space and the total sound pressure level spectrum at a specific observation point (e.g., (20, 0, 1.5)) are output.

[0052] The final output includes a spatial distribution cloud map of urban sound pressure and a total sound pressure level spectrum at a specific observation point, as shown below. Figure 6 As shown, this figure illustrates the results of a numerical simulation (such as finite element analysis or computational fluid dynamics) of the spatial distribution of a three-dimensional sound pressure field at a specific frequency of 143.39 Hz. Specifically, the model is established in a three-dimensional Cartesian coordinate system with a spatial scale in meters. The computational domain ranges approximately from -50 meters to 50 meters horizontally and from 0 meters to 50 meters vertically. At the bottom center of the computational domain, there exists a geometric structure consisting of a central cylinder and several rectangular appendages, which acts as a sound source radiator or acoustic scatterer in this simulation. Visually, the image extracts and displays the sound wave propagation path and energy distribution within the three-dimensional volume through mutually orthogonal two-dimensional cross-sections. The color scale on the right quantifies the amplitude of the sound pressure, reaching an order of magnitude of 10. 5 Pa (range from -4 × 10) 5 Pa to 4×10 5The waveform (Pa) shows that the deep red area represents the positive sound pressure peak (i.e., the medium compression region), and the deep blue area represents the negative sound pressure trough (i.e., the medium rarefaction region). The alternating red and blue ripples on the slice visually reflect the periodic phase change and spatial interference pattern of the sound wave. Further observation of the wavefront morphology reveals that the sound wave energy is mainly concentrated above the geometric structure and exhibits a significant directional radiation beam along a specific direction (mainly along the positive z-axis to the right). This fully reveals the highly directional acoustic radiation or scattering characteristics of the central structure at the current frequency.

[0053] In summary, to address the extremely high computational cost of directly solving the Compressible Fluid Equation (CAA) in existing technologies at the urban scale, and the difficulty of rigorously handling multipath reflection and diffraction issues due to the lack of an exact Green's function in building complexes using traditional acoustic analogy (FW-H) methods, this invention provides a method for predicting urban noise propagation in rotorcraft. This method organically separates the extraction of microscopic high-fidelity aerodynamic noise from macroscopic large-scale urban building sound propagation. It overcomes the problem of Green's function failure in complex buildings using the FW-H method and avoids directly generating a global high-density fluid mesh in large-scale urban environments through FEM / BEM surface coupling. This achieves both rigorous consideration of building reflection and diffraction while significantly reducing computational costs, and is easily scalable to noise assessment and multi-condition parameter scanning in large-scale urban environments.

[0054] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0055] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting urban noise propagation from rotary-wing aircraft, characterized in that, include: Acquire the sound pressure level spectrum data at preset observation points in the near-field region of the rotorcraft; Based on the sound pressure level spectrum data, an equivalent point source model containing multipole characteristics is reconstructed, and the sound power spectrum of the equivalent point source is determined. A macroscopic acoustic propagation model incorporating the geometry of urban buildings is established, and the equivalent point source model is positioned at a preset target location within the macroscopic acoustic propagation model. A finite element calculation region is constructed around the equivalent point sound source at the target location, and the outer surface of the finite element calculation region is used as the coupling surface with the boundary element calculation region. Using the finite element method, the near-field acoustic response within the finite element calculation region is calculated based on the sound power spectrum. Combined with the boundary element method, the far-field sound pressure distribution and spectrum, including building reflection conditions, are solved through the coupling surface. The acquisition of sound pressure level spectrum data at preset observation points in the near-field region of the rotorcraft specifically includes: A local fluid computational domain for a rotorcraft is established, and the transient flow field of the local fluid computational domain is solved using the large eddy simulation method. During the solution process, multiple observation points at different angles are set in the three-dimensional space of the rotorcraft within the local fluid computational domain. Based on the FW-H acoustic analogy equation, the sound pressure level spectrum at each observation point is extracted; the number of observation points is no less than four. The frequencies corresponding to the sound pressure level spectrum at each observation point are subjected to cubic spline interpolation based on the logarithmic frequency axis to unify the frequency axis of different observation points.

2. The method for predicting urban noise propagation from rotary-wing aircraft as described in claim 1, characterized in that, The reconstructing of an equivalent point source model incorporating multipole characteristics based on the sound pressure level spectrum data specifically includes: The equivalent point sound source is set as an axisymmetric structure, and its sound field is composed of monopoles, dipoles and quadrupoles. For each frequency point in the sound pressure level spectrum data, the equivalent point source at that frequency is set at any observation polar angle. The square of the total sound pressure level in the direction and the sound field at that frequency satisfy the following preset conditions: in, To observe the polar angle The squared value of the total sound pressure level measured in the direction; It is a monopole directionality function, and ; Monopole strength; It is the dipole directionality function, and ; The strength of the dipole; It is the quadrupole directionality function, and ; The quadrupole strength.

3. The method for predicting urban noise propagation from rotary-wing aircraft as described in claim 2, characterized in that, The determination of the sound power spectrum of the equivalent point sound source specifically includes: For each frequency point in the sound pressure level spectrum data, the total squared sound pressure value at that frequency point under at least four observation polar angles is substituted into the formula of the preset condition to establish a linear equation system, and the monopole intensity, dipole intensity and quadrupole intensity at that frequency point are analytically solved. Based on the source intensity results at each frequency point obtained by the solution, and combined with the conversion relationship between sound intensity and sound power, the sound power spectrum of the equivalent point sound source as a function of frequency is calculated.

4. The method for predicting urban noise propagation from rotorcraft as described in claim 1, characterized in that, The sound pressure level spectrum data is derived from a scaled-down rotor model. Before establishing a macroscopic acoustic propagation model that includes the geometry of urban buildings, the process also includes: determining the amplification ratio based on the characteristic lengths of the rotorcraft and the scaled-down model; determining the scaling correction coefficient based on the rotor aeroacoustic similarity criterion of the blade tip velocity under the same rotational speed conditions; and amplifying the sound power spectrum of the equivalent point sound source to the sixth power of the amplification ratio.

5. The method for predicting urban noise propagation from rotorcraft as described in claim 1, characterized in that, The finite element calculation region is a spherical space region that encloses the equivalent point sound source.

6. The method for predicting urban noise propagation from rotorcraft as described in claim 1, characterized in that, In the finite element calculation region, the acoustic power spectrum is used as the input excitation of the equivalent point sound source to calculate the near-field acoustic response.

7. The method for predicting urban noise propagation from rotorcraft as described in claim 1, characterized in that, In the boundary element calculation region, the surfaces of urban buildings are set as rigid reflective boundary conditions or symmetric boundary conditions to calculate the reflection and diffraction effects of sound waves between buildings.

Images