Aircraft fuselage surface acoustic load time domain prediction method and system

Abstract

This invention relates to a method and system for predicting the time-domain acoustic load on the surface of an aircraft, belonging to the field of aircraft detection technology. To address the problem of high-precision calculation of acoustic scattering in aircraft, this invention establishes a flow field calculation grid and an acoustic field grid; initializes the grid; calculates the Jacobian determinant and grid cell metric, and completes the grid coordinate system transformation; determines the convergence of the numerical solutions of the grid cells; iteratively solves the linear Euler equations at each time step using the Runge-Kutta method; obtains the flow field data and acoustic field data for each grid cell; performs filtering using a spatial filtering format to eliminate numerical artifacts, obtaining the filtered flow field data and acoustic field data for each grid cell; iteratively updates the flow field data and acoustic field data over time until the acquired acoustic load data stabilizes, outputting the time-domain data of the acoustic load on the aircraft surface, a flow field acoustic field cloud map, and a flow field cloud map. This invention balances computational efficiency and accuracy.

Description

Technical Field

[0001] This invention belongs to the field of aircraft testing technology, specifically relating to a method and system for predicting the time-domain acoustic load on the surface of an aircraft. Background Technology

[0002] Propeller noise is one of the main sound sources of turboprop aircraft. Due to the close proximity of the propeller to the fuselage, the fuselage strongly scatters the propeller noise, thus affecting the sound field distribution. Therefore, accurately predicting the scattering effect of the fuselage on propeller noise is crucial for studying the noise characteristics of turboprop aircraft. The scattering effect can be solved using traditional CFD methods, fast prediction, or a hybrid method (CFD+CAA). CFD methods require a sufficiently large computational domain and high mesh resolution to accurately capture the passing frequencies of blades with high sound intensity. Furthermore, the complex structure of rotating components and the fuselage significantly increases the computational load due to boundary treatment. Fast prediction methods provide rapid predictions for components with specific structures but cannot explain the mechanistic characteristics of noise generation and propagation, making them highly targeted. Hybrid methods overcome the geometric limitations of fast prediction while reducing computational load and improving efficiency, making them an effective prediction method that balances high accuracy and efficiency. However, acoustic pulsations differ significantly in scale from macroscopic changes in flow fields (such as vortex structures and boundary layer structures). The acoustic quantities involved in a sound field are typically 5 to 6 orders of magnitude smaller than those in fluid dynamics, and sound often travels long distances. If not handled properly, these quantities may be masked by numerical errors, dissipating the true acoustic signal. Furthermore, the magnitude of sound intensity varies by 10 from the hearing threshold to the pain threshold. 14 The frequency range is 20–20 kHz. Simulating motion at all these scales is extremely difficult. Therefore, there is an urgent need to develop a sound field solution method with high accuracy, low dispersion, and low dissipation, capable of accurately capturing sound waves to meet the practical engineering requirements for predicting the acoustic loads of turboprop aircraft. Summary of the Invention

[0003] The problem to be solved by this invention is to perform high-precision calculation of the acoustic scattering problem of aircraft, and proposes a time-domain prediction method and system for the acoustic load on the surface of an aircraft.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A method for time-domain prediction of acoustic load on the surface of an aircraft fuselage includes the following steps:

[0006] S1. Establish the aircraft flow field calculation grid and sound field grid, read in the aircraft parameter file, the aircraft flow field calculation grid and sound field grid file and the flow field data file, initialize the information transmission interface, and establish the connectivity between grid cells;

[0007] S2. Calculate the boundary parameters according to the boundary condition type in the aircraft parameter file obtained in step S1, and initialize the boundary conditions, including the use of unconsolidated wall boundary conditions on the aircraft wall and the use of a combination of mesh stretching and absorption layer conditions in the far field of the aircraft.

[0008] S3. Calculate the Jacobian determinant and grid cell metric, complete the grid coordinate system transformation, and transform the linear Euler equation into an equation in general curvilinear coordinates;

[0009] S4. For each time step, traverse all grid cells, calculate the maximum value of the convergence condition check number at each time step, and determine the convergence of the numerical solution of the grid cell;

[0010] S5. The Runge-Kutta method is used to iteratively solve the linear Euler equations at each time step. Specifically, this includes applying boundary conditions to each grid cell and performing MPI communication; using a high-order, low-dissipation, low-dispersion difference scheme to calculate spatial derivatives; combining the sound source parameters to solve the linear Euler equations; and achieving the solution for the entire grid cell region, thereby obtaining the flow field data and sound field data for each grid cell.

[0011] S6. For the flow field data and sound field data of each grid cell obtained in step S5, a spatial filtering format is used to filter the data to eliminate numerical pseudo-waves, and the filtered flow field data and sound field data of each grid cell are obtained.

[0012] S7. Repeat steps S4-S6 to update the flow field data and sound field data over time until the acquired acoustic load data stabilizes, and output the time-domain data of the acoustic load on the aircraft surface, the sound field cloud map of the watershed, and the flow field cloud map.

[0013] Furthermore, the specific implementation method of step S1 includes the following steps:

[0014] S1.1. The flow field and sound field calculation grids of the aircraft are established using a structured grid; the overall region is divided into multiple sub-regions using multi-block meshing technology, and a structured grid is established for each sub-region, with the grid distribution at the boundary of adjacent sub-regions remaining consistent;

[0015] S1.2. Read in the aircraft parameter file, mesh file, and flow field data file;

[0016] The aircraft parameter file includes Mach number, temperature, density, angle of attack, monitoring point coordinates, grid information, sound source information, and flow field information. Among them, the grid information is used to identify and process the grid file and establish the connectivity of grids in different regions; the sound source information includes sound source type, sound source amplitude, and sound source frequency; the flow field information is used to identify the flow field data file and allocate data space for each flow field data.

[0017] S1.3. Initialize the information transmission interface and establish connectivity between grid cells.

[0018] Furthermore, the specific implementation method of step S2 includes the following steps:

[0019] S2.1. The aircraft wall adopts a non-cohesive wall boundary condition, that is, the normal velocity and its partial derivative with respect to time are zero;

[0020] All physical points in the boundary region are divided into three categories: internal points, boundary points, and virtual points. Internal points are obtained by solving the governing equations; boundary points are specified by boundary conditions while simultaneously satisfying the governing equations; virtual points are obtained by specifying pressure values, such that… To simulate solid wall boundary conditions;

[0021] S2.2. The far-field conditions of the aircraft employ a combination of grid stretching and an absorbing layer;

[0022] The absorption layer adds three virtual mesh layers outside the computational domain mesh. In the absorption layer, the outward-propagating wave passes through an attenuation factor function. Decay to a specified value, that is, an explicitly decayed solution vector after each time step iteration:

[0023]

[0024] in, U is the initial solution vector after the (n+1)th iteration time step. targent Let U be the target decay solution vector. n+1 This is the explicitly decaying solution vector after the (n+1)th iteration time step. It is the attenuation factor;

[0025] The attenuation factor function transitions smoothly in the following manner, expressed as:

[0026]

[0027] Where L is the width of the absorption layer. It is the maximum value of the attenuation factor. It is a constant of the decay factor function;

[0028] The shape of the decay function is determined manually.

[0029] While using the absorption layer method, a mesh stretching method is simultaneously used to reduce the mesh density in the far field to attenuate the propagation of sound waves.

[0030] Furthermore, the formula for calculating the CFL number in step S4 is as follows:

[0031]

[0032] in, It is the flow field at each time step. It is the time step. It is the spatial step size;

[0033] Determine the convergence of the numerical solution of the mesh element. If the numerical solution of the mesh element converges, proceed to the next step. If the numerical solution of the mesh element does not converge, adjust the mesh read in step S1 until the convergence condition is met.

[0034] Furthermore, the specific implementation method of step S5 includes the following steps:

[0035] S5.1. Apply boundary conditions to each converged mesh cell obtained in step S5 and perform MPI communication;

[0036] S5.2. A high-order, low-dissipation, low-dispersion difference scheme is used to calculate the spatial derivative. Combined with the sound source parameters, the linear Euler equation is solved to solve the grid cells of the entire region, and the flow field data and sound field data of each grid cell are obtained.

[0037] S5.2.1. The linearized Euler equation is used to solve for the acoustic load on the body surface; all variables in the calculation are dimensionless according to the following reference values: the reference length is... The reference speed is The reference density is The reference pressure is Reference time is ;

[0038] S5.2.2. Spatial derivative calculation adopts the high-order, low-dissipation, low-dispersion OPCS scheme, decomposing the derivative into the sum of the forward and backward derivatives; the outer boundary of the grid block adopts the explicit template of the third-order four-point scheme, and the interface of the block adopts the explicit template of the fourth-order eleven-point scheme, forming a recursive system in the matrix form of the fourth-order OPCS. The OPCS matrix is ​​a double diagonal matrix. The lower diagonal matrix is ​​solved by forward sweep, and the upper diagonal matrix is ​​solved by backward sweep; in the three-dimensional flow domain of dimension (Nx, Ny, Nz), a total of Ny×Nz upper and lower diagonal matrices are solved;

[0039] S5.2.3. The time discretization adopts the explicit multi-intermediate-step Runge-Kutta method, which alternates between 4-step and 6-step LDDRK formats during time advancement, and optimizes the coefficients for the alternating format. This ensures that the advancement format maintains low dissipation and low dispersion characteristics while achieving 4th-order precision at each time step, and also realizes a low-memory application form.

[0040] S5.2.4. The sound source term is given by sound source information transformed from computational fluid dynamics solution or analytical sound source model. The sound source information transformed from computational fluid dynamics solution is obtained by CFD calculation to obtain unsteady turbulent velocity, and the sound source information is obtained by taking its derivative.

[0041] The sound source is simplified for the sound scattering problem and given a classical sound source model, which is composed of rotating monopoles or dipoles superimposed.

[0042] S5.2.5. Combining the sound source term from step S5.2.4, solve the energy equation, momentum equation, and continuity equation through spatial and temporal discretization to obtain sound field data.

[0043] Furthermore, the spatial filtering format in step S6 is an explicit spatial filter using the classic tenth-order format, expressed as:

[0044]

[0045] in, Let i be the i-th filtered sound field or flow field. Let be the i-th original and unfiltered flow or sound field.

[0046] A time-domain prediction system for the acoustic load on the surface of an aircraft is provided, based on the aforementioned time-domain prediction method for the acoustic load on the surface of an aircraft. The system includes a processor and a computer-readable storage medium communicatively connected to the processor. The computer-readable storage medium further includes a data list for each grid cell and stores computer instructions. When the computer instructions are executed by the processor, the aforementioned time-domain prediction method for the acoustic load on the surface of an aircraft is implemented.

[0047] The beneficial effects of this invention are:

[0048] This invention discloses a time-domain prediction method for the acoustic load on an aircraft surface. It employs a high-order scheme to achieve high-precision numerical simulation of noise propagation and combines spatial filtering to improve computational stability. For time discretization, an explicit low-dispersion, low-dissipation time scheme is used to increase the time step size and reduce computational load. Simultaneously, a non-reflective boundary condition is introduced in the far field to reduce sound wave reflection. Overall, this method improves sound wave resolution, accurately simulates the sound scattering problem of rotating sound sources while maintaining computational efficiency and accuracy, and enhances the accuracy of time-domain prediction of the acoustic load on the aircraft surface.

[0049] The present invention discloses a time-domain prediction method for the acoustic load on the surface of an aircraft, which can take into account the complex geometric details of the aircraft and achieve high-precision numerical simulation of the acoustic load on the surface of a turboprop aircraft through high-order numerical format and filtering. The present invention exhibits low dispersion and dissipation, achieving a balance between computational efficiency and accuracy. Attached Figure Description

[0050] Figure 1 This is a flowchart of a method for time-domain prediction of acoustic load on the surface of an aircraft body, as described in this invention.

[0051] Figure 2 A schematic diagram of a full-size wing-body assembly model and its solution domain;

[0052] Figure 3 For a full-size wing-body assembly grid;

[0053] Figure 4 Method for dividing the boundary points of the near-wall region;

[0054] Figure 5 A comparison chart of numerical analytical wavenumbers in the difference scheme;

[0055] Figure 6 The figures show a comparison of dispersion error and dissipation error for different time discretization schemes, where (a) is a comparison of dispersion error and (b) is a comparison of dissipation error.

[0056] Figure 7 The images show the sound field contours before and after filtering, where (a) is the sound field contour before filtering and (b) is the sound field contour after filtering.

[0057] Figure 8 The image shows the instantaneous acoustic pressure cloud of the turboprop plane.

[0058] Figure 9 This is a curve showing the sound pressure level distribution on the fuselage surface. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described specific embodiments are merely a part of the embodiments of the invention, and not all of them. The components of the specific embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations, and the invention may also have other embodiments.

[0060] Therefore, the following detailed description of specific embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected specific embodiments of the invention. All other specific embodiments obtained by those skilled in the art based on these specific embodiments without inventive effort are within the scope of protection of this invention.

[0061] To further understand the invention's content, features, and effects, the following specific embodiments are provided, along with accompanying drawings. Figure 1 - Appendix Figure 9 Detailed explanation is as follows:

[0062] Example 1:

[0063] A method for time-domain prediction of acoustic load on the surface of an aircraft fuselage includes the following steps:

[0064] S1. Establish the aircraft flow field calculation grid and sound field grid, read in the aircraft parameter file, the aircraft flow field calculation grid and sound field grid file and the flow field data file, initialize the information transmission interface, and establish the connectivity between grid cells;

[0065] Furthermore, the specific implementation method of step S1 includes the following steps:

[0066] S1.1. The flow field and sound field calculation grids of the aircraft are established using a structured grid; the overall region is divided into multiple sub-regions using multi-block meshing technology, and a structured grid is established for each sub-region, with the grid distribution at the boundary of adjacent sub-regions remaining consistent;

[0067] Furthermore, the flow field mesh considers the influence of the boundary layer, that is, the thickness of the first boundary layer is calculated based on the actual operating conditions of the aircraft:

[0068]

[0069] in, The coefficient of dynamic viscosity, The coefficient of friction, The value is 1;

[0070] S1.2. Read in the aircraft parameter file, mesh file, and flow field data file;

[0071] The aircraft parameter file includes Mach number, temperature, density, angle of attack, monitoring point coordinates, grid information, sound source information, and flow field information. Among them, the grid information is used to identify and process the grid file and establish the connectivity of grids in different regions; the sound source information includes sound source type, sound source amplitude, and sound source frequency; the flow field information is used to identify the flow field data file and allocate data space for each flow field data.

[0072] Furthermore, to accurately simulate the near field, it is necessary to consider phenomena such as scattering, refraction, and absorption of sound waves by turbulence and shear flow, i.e., solving the linearized Euler equations based on a non-uniform background flow field. Based on the flight parameter file, CFD calculations are used to obtain... (pressure), (density), , , (Three-directional velocity);

[0073] S1.3. Initialize the information transmission interface and establish connectivity between grid cells;

[0074] Furthermore, taking the calculation of propeller noise scattering on a full-size fuselage as an example, such as... Figure 2 A schematic diagram of the model and computational domain is provided. The model consists of two parts: the wing and the fuselage. The propeller plane is perpendicular to the fuselage, and both propellers rotate counterclockwise. The background flow Mach number is 0.2. This invention uses the finite difference method to solve the linearized Euler equations, and a structured mesh is used for acoustic calculations. For flow field calculations, to account for the influence of the boundary layer, a boundary layer mesh needs to be set near the aircraft. Both the acoustic and flow field calculation meshes use structured meshes to reduce interpolation errors. Simultaneously, a multi-block meshing technique is used to partition the geometry. The mesh size is adjusted at the interface of the partitioned meshes to ensure the continuity of mesh metrics. Figure 3 A full-size wing-body assembly mesh is presented, with an acoustic mesh density (PPW) of 7. Smooth transition meshes are used at the nose and tail, while O-shaped meshes are employed at the leading and trailing edges of the wing to improve mesh quality.

[0075] S2. Calculate the boundary parameters according to the boundary condition type in the aircraft parameter file obtained in step S1, and initialize the boundary conditions, including the use of unconsolidated wall boundary conditions on the aircraft wall and the use of a combination of mesh stretching and absorption layer conditions in the far field of the aircraft.

[0076] Furthermore, the specific implementation method of step S2 includes the following steps:

[0077] S2.1. The spacecraft wall adopts an unconsolidated wall boundary condition, that is, the normal velocity and its partial derivative with respect to time are zero, expressed as:

[0078] ;

[0079] in It is the magnitude of the normal vector on the solid wall. , , These are the horizontal, vertical, and longitudinal variables within the computational domain; , , , , , , , and These are the partial derivatives of the horizontal variables with respect to the x-axis, the partial derivatives of the horizontal variables with respect to the y-axis, the partial derivatives of the horizontal variables with respect to the z-axis, the partial derivatives of the vertical variables with respect to the x-axis, the partial derivatives of the vertical variables with respect to the y-axis, the partial derivatives of the vertical variables with respect to the z-axis, the partial derivatives of the vertical variables with respect to the x-axis, the partial derivatives of the vertical variables with respect to the y-axis, and the partial derivatives of the vertical variables with respect to the z-axis. , and These are the x-axis component, y-axis component, and z-axis component of the wall normal vector, respectively.

[0080] like Figure 4 All physical points in the boundary region are divided into three categories: internal points, boundary points, and virtual points. Internal points are obtained by solving the governing equations; boundary points are specified by boundary conditions while simultaneously satisfying the governing equations; and virtual points are determined by specifying pressure values ​​to ensure... To simulate solid wall boundary conditions;

[0081] S2.2. The far-field conditions of the aircraft employ a combination of grid stretching and an absorbing layer;

[0082] The absorption layer adds three virtual mesh layers outside the computational domain mesh. In the absorption layer, the outward-propagating wave passes through an attenuation factor function. Decay to a specified value, that is, an explicitly decayed solution vector after each time step iteration:

[0083]

[0084] in, U is the initial solution vector after the (n+1)th iteration time step. targent Let U be the target decay solution vector. n+1 This is the explicitly decaying solution vector after the (n+1)th iteration time step. It is the attenuation factor;

[0085] The attenuation factor function transitions smoothly in the following manner, expressed as:

[0086]

[0087] Where L is the width of the absorption layer. It is the maximum value of the attenuation factor. It is a constant of the decay factor function;

[0088] The shape of the decay function is determined manually; for example... Figure 2The computational domain is a cuboid, and an outflow buffer is set on each of the six faces of the cuboid.

[0089] While using the absorbing layer method, a mesh stretching method is simultaneously employed to reduce the mesh density in the far field, thereby attenuating sound wave propagation. The far field utilizes a combination of mesh stretching and the absorbing layer to reduce the size of the physical domain while minimizing reflections from artificial boundaries. When the average background flow field is not uniform or the sound disturbance does not propagate perpendicular to the wall, traditional non-reflective boundary conditions are ineffective in attenuating sound wave propagation. Therefore, the absorbing layer boundary condition, through an attenuation function... The outward-propagating wave is attenuated to the target value, thereby reducing false reflections at the boundary;

[0090] S3. Calculate the Jacobian determinant and grid cell metric, complete the grid coordinate system transformation, and transform the linear Euler equation into an equation in general curvilinear coordinates;

[0091] Furthermore, the specific implementation method of step S3 includes the following steps:

[0092] S3.1. The expression for the Jacobian determinant is:

[0093]

[0094] in, It is a Jacobian determinant;

[0095] S3.2. According to the standard rules of inverse matrices, the expression is:

[0096] ;

[0097] S3.3. Compare the corresponding elements of the two matrices in the equation of step S3.2 to obtain the following relational expression for the direct metric represented by the inverse metric:

[0098] .

[0099] S4. For each time step, traverse all grid cells, calculate the maximum value of the convergence condition check number at each time step, and determine the convergence of the numerical solution of the grid cell;

[0100] Furthermore, the formula for calculating the CFL number in step S4 is as follows:

[0101]

[0102] in, It is the flow field at each time step. It is the time step. It is the spatial step size;

[0103] Determine the convergence of the numerical solution of the mesh element. If the numerical solution of the mesh element converges, proceed to the next step. If the numerical solution of the mesh element does not converge, adjust the mesh read in step S1 until the convergence condition is met.

[0104] S5. The Runge-Kutta method is used to iteratively solve the linear Euler equations at each time step. Specifically, this includes applying boundary conditions to each grid cell and performing MPI communication; using a high-order, low-dissipation, low-dispersion difference scheme to calculate spatial derivatives; combining the sound source parameters to solve the linear Euler equations; and achieving the solution for the entire grid cell region, thereby obtaining the flow field data and sound field data for each grid cell.

[0105] Furthermore, the specific implementation method of step S5 includes the following steps:

[0106] S5.1. Apply boundary conditions to each converged mesh cell obtained in step S5 and perform MPI communication;

[0107] S5.2. A high-order, low-dissipation, low-dispersion difference scheme is used to calculate the spatial derivative. Combined with the sound source parameters, the linear Euler equation is solved to solve the grid cells of the entire region, and the flow field data and sound field data of each grid cell are obtained.

[0108] S5.2.1. The linearized Euler equation is used to solve for the acoustic load on the body surface. The calculation formula is as follows:

[0109]

[0110] in, , and They represent , , Component of velocity in direction, and These represent density and pressure, respectively. This represents specific heat, set to steady state. The specific heat of air is 1.4. The superscript ' indicates a disturbed field, and the subscript 0 represents the background flow field. arrive These are the first source term, the second source term, the third source term, the fourth source term, and the fifth source term, respectively.

[0111] All variables in the calculations are dimensionless according to the following reference values: Reference length is... The reference speed is The reference density is The reference pressure is Reference time is ;

[0112] S5.2.2. Spatial derivative calculation adopts the high-order, low-dissipation, low-dispersion OPCS scheme, decomposing the derivative into the sum of the forward and backward derivatives; the outer boundary of the grid block adopts the explicit template of the third-order four-point scheme, and the interface of the block adopts the explicit template of the fourth-order eleven-point scheme, forming a recursive system in the matrix form of the fourth-order OPCS. The OPCS matrix is ​​a double diagonal matrix. The lower diagonal matrix is ​​solved by forward sweep, and the upper diagonal matrix is ​​solved by backward sweep; in the three-dimensional flow domain of dimension (Nx, Ny, Nz), a total of Ny×Nz upper and lower diagonal matrices are solved;

[0113] Furthermore, the fourth-order OPCS along the x-direction can be given as a recursive system:

[0114]

[0115] in, , ,and Let these represent the original derivative, forward derivative, and backward derivative at the grid points, respectively. Represents the known original field. Coefficients , , , , , , and All are constants. For the outer boundary of the mesh blocks, a third-order four-point explicit template (combined forward and backward templates) is used:

[0116]

[0117] At the interface of the blocks, an explicit template of eleven points of order four (combining forward and backward) is used:

[0118]

[0119] in, , and All coefficients are definite, and the above formulas form a recursive system. This can be represented in matrix form as follows:

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127] Similarly, along and The fourth-order OPCS in the direction can be changed by... and Let me find a solution. Figure 5 By comparing the resolvable wavenumbers of different formats, it can be seen that the OPCS format applied in this invention can improve the maximum resolvable wavenumber of the format.

[0128] S5.2.3. The time discretization adopts the explicit multi-intermediate-step Runge-Kutta method, which alternates between 4-step and 6-step LDDRK formats during time advancement, and optimizes the coefficients for the alternating format. This ensures that the advancement format maintains low dissipation and low dispersion characteristics while achieving 4th-order precision at each time step, and also realizes a low-memory application form.

[0129] Furthermore, the low-memory P-step LDDRK format can be represented as:

[0130] cycle :

[0131] ;

[0132] Incremental stacking:

[0133] ;

[0134] in It is an intermediate step increment. It is the solution vector. It is a constant coefficient, calculated according to the following formula:

[0135]

[0136] Figure 6 By comparing the dispersion error and dissipation error of different time discretization schemes, it can be seen that the method used in this invention has higher accuracy and stability.

[0137] S5.2.4. The sound source term is given using sound source information derived from computational fluid dynamics solutions or analytical sound source models. The sound source information derived from computational fluid dynamics solutions is obtained by CFD calculation of unsteady turbulent velocities, and the sound source information is obtained by differentiating these velocities.

[0138]

[0139] in, It is the velocity of the i-th sound source at time t. It is the average velocity of the i-th sound source at time t. This represents the information of the i-th sound source, and j is the position of the grid point;

[0140] To simplify the sound source problem, a classical sound source model is adopted, which consists of a superposition of rotating monopoles or dipoles, with the center coordinates of the rotation axis being... If the position of the i-th rotating sound source is perpendicular to the yoz plane, then the position is as follows:

[0141]

[0142] in, Indicates the radius of the annulus. Indicates the relative position of the sound source. Indicates rotational angular velocity;

[0143] The velocity of the sound source is expressed as:

[0144]

[0145] Set the monopole strength to The strength of the dipole is The expression for the sound source term is rewritten as:

[0146]

[0147] S5.2.5. Combining the sound source term from step S5.2.4, solve the energy equation, momentum equation, and continuity equation through spatial and temporal discretization to obtain sound field data.

[0148] S6. For the flow field data and sound field data of each grid cell obtained in step S5, a spatial filtering format is used to filter the data to eliminate numerical pseudo-waves, and the filtered flow field data and sound field data of each grid cell are obtained.

[0149] Furthermore, the spatial filtering format in step S6 is an explicit spatial filter using the classic tenth-order format, expressed as:

[0150]

[0151] in, Let i be the i-th filtered sound field or flow field. Let i be the i-th original and unfiltered flow field or sound field;

[0152] Figure 7 The filtering effect was analyzed using a point source example. It can be seen that the filtered numerical calculation results suppress non-physical numerical oscillations caused by numerical calculation, and the contour plots of isobars are smoother and closer to the true values.

[0153] S7. Repeat steps S4-S6 to update the flow field data and sound field data over time until the acquired acoustic load data stabilizes, and output the time-domain data of the acoustic load on the aircraft surface, the sound field cloud map of the watershed, and the flow field cloud map.

[0154] Furthermore, Figure 8 Instantaneous sound pressure cloud diagrams of the turboprop plane are presented. The counterclockwise rotation of the sound pressure in the turboprop plane over time indicates the counterclockwise rotation of the sound source. The calculation results of this invention can clearly calculate the sound scattering effect of the fuselage on the propeller. From Figure 9 By comparing the sound pressure level distribution curves of uniform incoming flow and RANS, this invention can take into account the refraction effect of the incoming flow on the propeller sound waves and predict the time-domain results of the acoustic load on the aircraft surface.

[0155] Example 2:

[0156] A time-domain prediction system for the acoustic load on the surface of an aircraft is provided, based on the time-domain prediction method for the acoustic load on the surface of an aircraft described in Embodiment 1. The system includes a processor and a computer-readable storage medium communicatively connected to the processor. The computer-readable storage medium further includes a data list for each grid cell and stores computer instructions. When the computer instructions are executed by the processor, the aforementioned time-domain prediction method for the acoustic load on the surface of an aircraft is implemented.

[0157] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0158] Although this application has been described above with reference to specific embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of this application. In particular, as long as there is no structural conflict, the features in the specific embodiments disclosed in this application can be combined with each other in any way. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, this application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for time-domain prediction of acoustic loads on a surface of an aircraft body, characterized in that, Includes the following steps: S1. Establish the aircraft flow field calculation grid and sound field grid, read in the aircraft parameter file, the aircraft flow field calculation grid and sound field grid file and the flow field data file, initialize the information transmission interface, and establish the connectivity between grid cells; S2. Calculate the boundary parameters according to the boundary condition type in the aircraft parameter file obtained in step S1, and initialize the boundary conditions, including the use of unconsolidated wall boundary conditions on the aircraft wall and the use of a combination of mesh stretching and absorption layer conditions in the far field of the aircraft. S3. Calculate the Jacobian determinant and grid cell metric, complete the grid coordinate system transformation, and transform the linear Euler equation into an equation in general curvilinear coordinates; S4. For each time step, traverse all grid cells, calculate the maximum value of the convergence condition check number at each time step, and determine the convergence of the numerical solution of the grid cell; S5. The Runge-Kutta method is used to iteratively solve the linear Euler equations at each time step. Specifically, this includes applying boundary conditions to each grid cell and performing MPI communication; using a high-order, low-dissipation, low-dispersion difference scheme to calculate spatial derivatives; combining the sound source parameters to solve the linear Euler equations; and achieving the solution for the entire grid cell region, thereby obtaining the flow field data and sound field data for each grid cell. The specific implementation method of step S5 includes the following steps: S5.

1. Apply boundary conditions to each converged mesh cell obtained in step S5 and perform MPI communication; S5.

2. A high-order, low-dissipation, low-dispersion difference scheme is used to calculate the spatial derivative. Combined with the sound source parameters, the linear Euler equation is solved to solve the grid cells of the entire region, and the flow field data and sound field data of each grid cell are obtained. S5.2.

1. The linearized Euler equation is used to solve for the acoustic load on the body surface. The calculation formula is as follows: ； in, , and They represent , , Component of velocity in direction, and These represent density and pressure, respectively. This represents specific heat, set to steady state. The specific heat of air is 1.

4. The superscript ' indicates a disturbed field, and the subscript 0 represents the background flow field. arrive These are the first source term, the second source term, the third source term, the fourth source term, and the fifth source term, respectively. All variables in the calculations are dimensionless according to the following reference values: Reference length is... The reference speed is The reference density is The reference pressure is Reference time is ; S5.2.

2. Spatial derivative calculation adopts the high-order, low-dissipation, low-dispersion OPCS scheme, decomposing the derivative into the sum of the forward and backward derivatives; the outer boundary of the grid block adopts the explicit template of the third-order four-point scheme, and the interface of the block adopts the explicit template of the fourth-order eleven-point scheme, forming a recursive system in the matrix form of the fourth-order OPCS. The OPCS matrix is ​​a double diagonal matrix. The lower diagonal matrix is ​​solved by forward sweep, and the upper diagonal matrix is ​​solved by backward sweep; in the three-dimensional flow domain of dimension (Nx, Ny, Nz), a total of Ny×Nz upper and lower diagonal matrices are solved; S5.2.

3. The time discretization adopts the explicit multi-intermediate-step Runge-Kutta method, which alternates between 4-step and 6-step LDDRK formats during time advancement, and optimizes the coefficients for the alternating format. This ensures that the advancement format maintains low dissipation and low dispersion characteristics while achieving 4th-order precision at each time step, and also realizes a low-memory application form. S5.2.

4. The sound source term is given using sound source information derived from computational fluid dynamics solutions or analytical sound source models. The sound source information derived from computational fluid dynamics solutions is obtained by CFD calculation of unsteady turbulent velocities, and the sound source information is obtained by differentiating these velocities. ； in, It is the velocity of the i-th sound source at time t. It is the average velocity of the i-th sound source at time t. This represents the information of the i-th sound source, and j is the position of the grid point; To simplify the sound source problem, a classical sound source model is adopted, which consists of a superposition of rotating monopoles or dipoles, with the center coordinates of the rotation axis being... If the position of the i-th rotating sound source is perpendicular to the yoz plane, then the position is as follows: ； wherein, denotes the radius of the circular ring, denotes the relative position of the sound source, denotes the rotational angular velocity; The velocity of the sound source is expressed as: ； The monopole strength is set to and the dipole strength to The expression for the sound source term is rewritten as ； S5.2.

5. Combining the sound source term from step S5.2.4, solve the energy equation, momentum equation, and continuity equation through spatial and temporal discretization to obtain sound field data; S6. For the flow field data and sound field data of each grid cell obtained in step S5, a spatial filtering format is used to filter the data to eliminate numerical pseudo-waves, and the filtered flow field data and sound field data of each grid cell are obtained. S7. Repeat steps S4-S6 to update the flow field data and sound field data over time until the acquired acoustic load data stabilizes, and output the time-domain data of the acoustic load on the aircraft surface, the sound field cloud map of the watershed, and the flow field cloud map.

2. The method of claim 1, wherein, The specific implementation method of step S1 includes the following steps: S1.

1. The flow field and sound field calculation grids of the aircraft are established using a structured grid; the overall region is divided into multiple sub-regions using multi-block meshing technology, and a structured grid is established for each sub-region, with the grid distribution at the boundary of adjacent sub-regions remaining consistent; S1.

2. Read in the aircraft parameter file, mesh file, and flow field data file; The aircraft parameter file includes Mach number, temperature, density, angle of attack, monitoring point coordinates, grid information, sound source information, and flow field information. Among them, the grid information is used to identify and process the grid file and establish the connectivity of grids in different regions; the sound source information includes sound source type, sound source amplitude, and sound source frequency; the flow field information is used to identify the flow field data file and allocate data space for each flow field data. S1.

3. Initialize the information transmission interface and establish connectivity between grid cells.

3. A method according to claim 1 or 2, characterised in that, The specific implementation method of step S2 includes the following steps: S2.

1. The spacecraft wall adopts an unconsolidated wall boundary condition, that is, the normal velocity and its partial derivative with respect to time are zero, expressed as: ； in, It is the magnitude of the normal vector on the solid wall. , , These are the horizontal, vertical, and longitudinal variables within the computational domain; , , , , , , , and These are the partial derivatives of the horizontal variables with respect to the x-axis, the partial derivatives of the horizontal variables with respect to the y-axis, the partial derivatives of the horizontal variables with respect to the z-axis, the partial derivatives of the vertical variables with respect to the x-axis, the partial derivatives of the vertical variables with respect to the y-axis, the partial derivatives of the vertical variables with respect to the z-axis, the partial derivatives of the vertical variables with respect to the x-axis, the partial derivatives of the vertical variables with respect to the y-axis, and the partial derivatives of the vertical variables with respect to the z-axis. , and These are the x-axis component, y-axis component, and z-axis component of the wall normal vector, respectively. All the physical points in the boundary region are divided into three categories, including interior points, boundary points and virtual points, where the interior points are obtained by solving the governing equations; the boundary points are specified by the boundary conditions while satisfying the governing equations; the virtual points are specified by pressure values to simulate the solid wall boundary conditions. ​ S2.

2. The far-field conditions of the aircraft employ a combination of grid stretching and an absorbing layer; The absorbing layer is a three virtual grid layer added outside the computational domain grid, in which the outward propagating wave is attenuated by a decay factor function Decay to a specified value, i.e. explicit decay of the solution vector after each time step iteration: ； wherein, is the initial solution vector after the n+1 iteration time step, U target is the target attenuated solution vector, U n+1 is the explicit attenuated solution vector after the n+1 iteration time step, is the attenuation factor; The attenuation factor function transitions smoothly in the following manner, expressed as: ； where L is the width of the absorption layer, is the maximum value of the attenuation factor, is a constant of the attenuation factor function; The shape of the decay function is determined by manual setting; While using the absorption layer method, a mesh stretching method is simultaneously used to reduce the mesh density in the far field to attenuate the propagation of sound waves.

4. The method of claim 3, wherein, The specific implementation method of step S3 includes the following steps: S3.

1. The expression for the Jacobian determinant is: ； wherein J is the Jacobian determinant; S3.

2. According to the standard rules of inverse matrices, the expression is: ； S3.

3. Compare the corresponding elements of the two matrices in the equation of step S3.2 to obtain the following relational expression for the direct metric represented by the inverse metric: 。 5. The method of claim 4, wherein, Step S4: Solve for the formula for calculating the CFL number. ； wherein, is the flow field at each time step, is the time step, is the spatial step; Determine the convergence of the numerical solution of the mesh element. If the numerical solution of the mesh element converges, proceed to the next step. If the numerical solution of the mesh element does not converge, adjust the mesh read in step S1 until the convergence condition is met.

6. The method of claim 5, wherein, The spatial filtering format in step S6 is an explicit spatial filter using the classic tenth-order format, with the expression: ； wherein, is the i-th filtered acoustic or flow field, is the i-th original and unfiltered flow or acoustic field.

7. An aircraft surface acoustic load time domain prediction system, relying on an aircraft surface acoustic load time domain prediction method according to any one of claims 1-6, characterized in that, The system includes a processor and a computer-readable storage medium communicatively connected to the processor. The computer-readable storage medium further includes a data list for each grid cell and stores computer instructions that, when executed by the processor, implement the aforementioned method for time-domain prediction of acoustic loads on the surface of an aircraft.

Images