Far-field noise simulation prediction model for single-stage axial flow compressor component of turboshaft engine

Abstract

The invention discloses a far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine, relates to the technical field of engine noise prediction, and aims at solving the problems that a traditional noise prediction method is large in computing resource consumption and limited in simulation precision. The far-field noise simulation prediction model comprises a geometric modeling module used for constructing a geometric model; the flow field calculation module is used for constructing a flow field calculation model; the flow field solving module is used for executing steady-state flow field steady solution and transient flow field unsteady solution; the acoustic calculation module is used for constructing an acoustic calculation model, and the sound source mapping module is used for selecting a large turbulence energy area in a flow field as a Lightthill body sound source and mapping flow field data to the sound source; and the sound propagation calculation module is configured to set a gas compressor fluid domain as a finite element grid area, and an infinite element two-dimensional surface grid unit is arranged at the outlet boundary so as to calculate the sound pressure level frequency spectrum of the far-field monitoring point. According to the method, on the premise that the precision is guaranteed, efficient prediction of the far-field noise of the gas compressor is achieved with low cost.

Description

Technical Field

[0001] The present invention relates to engine noise prediction, and in particular to a single-stage axial flow compressor component of a turboshaft engine. The invention discloses a far-field noise simulation prediction model, and belongs to the technical field of finite element analysis. Background Art

[0002] With the rapid development of the aviation and automotive industries, airplanes and cars have become a part of most people's lives. Axial-flow compressors, as core components in aircraft and automotive engines, are a significant source of noise. Predicting compressor noise has become a crucial issue in compressor design.

[0003] Traditional noise prediction using direct numerical modeling is theoretically feasible, but it requires enormous computing resources, which are beyond the reach of current computing power and time. The only method for predicting compressor aerodynamic noise simulation is based on acoustic analogy, which is both theoretically and practically feasible. However, based on available public data, the overall simulation accuracy is low. Summary of the Invention

[0004] The present invention aims to solve the problem of large computational resource consumption and limited simulation accuracy of traditional noise prediction methods, and proposes A far-field noise simulation and prediction model for single-stage axial compressor components of a turboshaft engine.

[0005] The technical solution adopted by the present invention to solve the above problems is: The present invention provides a far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine, comprising: A geometric modeling module is used to construct a geometric model, which includes moving blades, stationary blades, a hub, and a shroud. The model ignores non-flow channel structures and maintains a smooth surface. The flow field calculation module is used to build the flow field calculation model. It uses Turbogrid to divide the high-fidelity flow field grid including the blade tip clearance, and then uses ICEM to perform periodic expansion to form a complete circumferential model. Flow field solution module, used to perform steady-state flow field steady solution and transient flow field unsteady solution; Acoustic calculation module, used to build acoustic calculation models, The sound source mapping module is used to select the area with large turbulent energy in the flow field as the Lighthill body sound source and map the flow field data to the sound source; The sound propagation calculation module is configured to set the compressor fluid domain as a finite element grid area, and the outlet boundary is provided with infinite element two-dimensional surface grid units to calculate the sound pressure level spectrum of the far-field monitoring point and obtain the noise situation at the far-field monitoring point.

[0006] Furthermore, the number of moving blades in the geometric modeling module is 25, the number of stationary blades is 38, and the blade tip clearance height is 0.8 mm.

[0007] Furthermore, the flow field calculation module uses Turbogrid to divide the flow channel grid and restores it through ICEM to form a complete circumferential model.

[0008] Furthermore, in the flow field solution module, the number of steps N and the time step of the transient solution satisfy the Nyquist sampling theorem, and the velocity field and density field data are exported in Ensight format.

[0009] Furthermore, the K-oumegaSST turbulence model is used in the flow field solution module, and the convergence residual is set to 1e-5.

[0010] Furthermore, the acoustic prediction model uses a tetrahedral grid, and the grid size in the near-field area is refined to 1 / 2-1 / 3 of the far-field area; the calculation frequency range of the acoustic prediction mode sound is 50-20000Hz, and the step size is set to 10-100Hz.

[0011] Furthermore, the body sound source region in the sound source mapping module includes the trailing edge of the rotor blade and the blade tip clearance region, and the sound source data includes the time domain file time.nff and the frequency domain file freq.nff.

[0012] Furthermore, in the sound propagation calculation module, the boundary of the infinite element surface grid unit is set at the outlet, and the monitoring points are arranged on a semicircle with the center of the outlet surface as the origin, with an axial distance of 2.7m and a circumferential interval of 1°.

[0013] The beneficial effects of the present invention are: 1. In the model of the present invention, the equivalent sound source adopts the Lighthill acoustic analogy method, and the sound propagation calculation is based on the simulation prediction of finite element unit grid cells and infinite element surface grid cells to achieve fast far-field noise prediction, significantly reduce computing resource consumption, avoid the huge computing cost of traditional direct numerical simulation, and achieve fast and efficient noise prediction; 2. This invention accurately captures flow field details through high-fidelity flow field meshing (such as blade tip clearance encryption) and the K-oumega SST turbulence model. The transient unsteady solution is combined with the Nyquist sampling theorem to ensure the sound pressure level spectrum resolution, making the noise prediction results closer to the actual operating conditions. 3. While ensuring accuracy, this invention can efficiently and cost-effectively predict the far-field noise at the outlet of a single-stage axial-flow compressor in a turboshaft engine. This method can be applied to noise prediction in the design of single-stage axial-flow compressors in turboshaft engines, thereby avoiding the problem of excessive noise in these compressors. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a solid model of a single-stage axial flow compressor test piece of a turboshaft engine according to the present invention.

[0015] Figure 2 This is a flow chart of the flow field and acoustic field simulation calculation model based on Fluent and Actran in the present invention.

[0016] Figure 3 The velocity distribution cloud map and pressure distribution cloud map of the moving blade flow field at span-0.99 predicted by Fluent of the present invention.

[0017] Figure 4 This is a time domain diagram of the pressure fluctuation at the leading edge of the moving blade of the present invention.

[0018] Figure 5 This is a frequency domain diagram of the pressure fluctuation at the leading edge of the moving blade of the present invention.

[0019] Figure 6 This is the Lighthill surface sound source cloud map extracted by the present invention.

[0020] Figure 7 This is the Lighthill volume sound source cloud map extracted by the present invention.

[0021] Figure 8 This is the sound field boundary condition set by the present invention.

[0022] Figure 9 This is the far-field sound pressure level spectrum of aerodynamic noise predicted by the flow field and sound field simulation calculation model based on Fluent and Actran in this invention. DETAILED DESCRIPTION

[0023] Specific embodiment 1: The far-field noise simulation prediction model of a single-stage axial flow compressor component of a turboshaft engine described in this embodiment includes: The geometric modeling module is used to construct a geometric model, including the rotor blades, stationary blades, hub, and shroud of a turboshaft engine's single-stage axial-flow compressor. During simulation, the shape and dimensions of the actual turboshaft engine's single-stage axial-flow compressor are preserved as faithfully as possible. Furthermore, the constructed model should be smooth, disregarding the gaps between components encountered during actual manufacturing and assembly, and capable of being recognized and processed by the simulation software.

[0024] Furthermore, SolidWorks 3D modeling software was used to construct a solid model based on the actual analysis model.

[0025] The flow field calculation module is used to build the flow field calculation model, specifically including: S201) The created geometric entity model is imported into the geometric structure operation interface within Workbench, and the DesginModeler interface is selected to read the model.

[0026] S202) After reading from the DesginModeler interface, sketch lines of the inlet, outlet, hub, and outer cover of the corresponding impeller stage of the geometric model are created.

[0027] S203) Associating the successfully generated lines to the Turbgrid grid division module.

[0028] S204) In the Turbgrid meshing module, use an appropriate mesh size, specify the impeller flow path mesh size, and specify a smaller mesh size for the impeller region. Generate a mesh that meets the mesh quality requirements and export the mesh in CGNS format.

[0029] S205) Import the exported CGNS mesh into ICEM meshing software to perform periodic rotation array.

[0030] S206) Export the correct and complete mesh into a .msh mesh file compatible with Fluent. Repeat the above steps for the other stages of the axial flow compressor to complete the meshing of the corresponding stages.

[0031] Furthermore, the step S204) specifically further includes: For the impeller part including the blade tip clearance, an additional step is required to fill in the corresponding blade tip clearance height in the column of Shorud Tip. The blade tip clearance height can be set to any number of millimeters; The flow field solution module is used to perform steady-state flow field steady-state solution and transient flow field unsteady solution, including: S301) Import all impeller mesh models included in a single-stage axial flow compressor of a turboshaft engine, and set an Interface mesh interface.

[0032] S302) Setting a suitable inlet pressure at the inlet interface and a suitable outlet pressure at the outlet interface.

[0033] S303) Select the fluid domain mesh of the impeller, rotate the MRF method, and set the correct selection direction, rotation speed, and rotation center.

[0034] S304) Select the K-oumegaSST model to perform steady-state flow field simulation calculations until the steady-state results converge.

[0035] S305) Based on the steady-state convergence result, the calculation type is switched from steady-state to transient, and a transient unsteady flow field simulation calculation is performed.

[0036] S306) Before the transient run, switch the MRF method in the fluid domain mesh to the sliding mesh method, leaving all other settings unchanged. Set the correct time step and number of steps for the transient run. During the run, export the Ensight file. The Ensight file should contain at least the velocity file and the density file.

[0037] Furthermore, the correct time step is determined based on the resolution required for the sound pressure level spectrum in the far-field noise calculation, the number of steps is determined by the maximum analysis frequency, and the time step satisfies the Nyquist sampling theorem.

[0038] The acoustic calculation module is used to build an acoustic calculation model, including: S401) Process the flow field model of the turboshaft engine single-stage axial compressor to be calculated, perform appropriate mesh division using Meshing, and then export the acoustic mesh file.

[0039] Furthermore, the acoustic grid size of the appropriate grid division should satisfy the following formula:

[0040] Where: h max is the maximum acoustic grid size, λ 0 is the wavelength of the sound wave at the analysis frequency.

[0041] The sound source mapping module is used to select areas with large turbulent energy in the flow field as Lighthill sound sources and map the flow field data to the sound sources. The specific modules include the following: S501) Import the acoustic mesh file into Actran and select the area with the largest turbulent kinetic energy in the flow field calculation as the Lighthill body sound source.

[0042] S502) Enter the Ensight file path into the CAA item of the ICFD column in Actran, calculate the sound source information, and obtain the time.nff sound source time domain file and the freq.nff sound source frequency domain file.

[0043] The sound propagation calculation module is configured to set the compressor fluid domain as a finite element mesh region, with infinite element surface mesh units set at the outlet boundary to calculate the sound pressure level spectrum of the far-field monitoring point and obtain the noise situation at the far-field monitoring point. Specifically, it includes: S601) Using Actran's Direct Frequency Response, enter the expected starting analysis frequency and maximum analysis frequency, and fill in the analysis frequency step size based on the expected resolution in the flow field calculation.

[0044] S602) In the Components option, select Acoustics as the entire acoustic component, each area of ​​the acoustic calculation grid as a finite element area, and the two-dimensional surface mesh at the outlet as an infinite element surface mesh. S603) In Direct Frequency Response, select the Lighthill volume source region calculated by ICFD as the sound source, and select the finite element region and infinite element surface.

[0045] S604) Set the far-field monitoring points, export the EDAT file, and calculate the EDAT file to obtain the corresponding spectrum file. Finally, use the PLTViewer in the Actran VI to view the obtained PLT result file.

[0046] Example: The technical solution of the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] (1) According to Figure 1 The simulation model construction process involved in the single-stage axial flow compressor test specimen of the turboshaft engine is as follows: Figure 2 As shown in the figure, a 3D geometric model of a single-machine axial compressor was established using Solidworks. The main focus was on the moving blades and stationary blades. Non-flow path structures such as bolt holes and assembly gaps were ignored. The blade surface was ensured to be smooth, and a 0.8mm tip clearance was set for the moving blades.

[0048] (2) Save the model in a format recognizable by ANSYS and import it into the DesignModeler module of ANSYS Workbench for subsequent processing.

[0049] (3) In DesignModeler, create sketch lines of the inlet, outlet, hub, and outer cover of the corresponding impeller stage of the geometric model, extract the flow path areas of the moving blades and stationary blades, and generate a flow path sketch.

[0050] (4) Select the flow path of the impeller stage, enter the correct number of blades (25 blades), and correctly select the blade surface of the impeller stage to scan out the blades in the flow path.

[0051] (5) Then use the Turbogrid module in Workbench to divide the flow channel grid and set the parameters that need to be set, such as the global grid size of the moving blade flow channel and the encrypted size of the blade tip clearance area. At the same time, when drawing the moving blade grid, the blade tip clearance should be set to 0.8mm and the first layer grid thickness should be 0.05mm.

[0052] (6) After exporting the single channel mesh, the ICEM module in Workbench is used to periodically expand the mesh to complete the model.

[0053] (7) Finally, the complete mesh is exported as a Fluent-compatible .msh file with a total mesh size of approximately 30 million. The mesh quality is then checked to ensure that it meets the CFD solution requirements. The above steps are repeated for the other stages and inlets and outlets of the axial flow compressor to complete the meshing of the corresponding parts.

[0054] (8) Import the mesh generated in the previous step into Fluent and set the boundary conditions as required. The fluid domain mesh of the impeller is meshed using the MRF method, with a rotation speed of 5400 RPM and a rotation direction around the origin in the positive Z-axis direction. The fluid domain and inlet and outlet channels of the stationary impeller are set to the stationary reference system.

[0055] (9) The K-oumegaSST turbulence model is used to solve the steady-state flow field. The convergence residual is set to 1e-5. After 3000 iterations, the residual is stable. The transient unsteady flow field simulation calculation is performed based on this result. (10) Transient unsteady flow field simulation calculation: Switch the physical model to a sliding grid model, set the appropriate time step, total number of steps, and number of iterations, set the export of velocity and density field data, and save the time series file in .cas format to obtain the corresponding flow field data file.

[0056] (11) By Figure 2 The velocity distribution cloud diagram and pressure distribution cloud diagram of the blade flow field at span-0.99 are shown. The flow field distribution of the single-stage axial flow compressor component of this turboshaft engine can be observed, and it can be seen that the calculation is convergent and reasonable. At the same time, the unsteady calculation is monitored by Figure 3 The time domain diagram of the pressure fluctuation at the leading edge of the moving blade is shown. By looking at the flow field, we can observe periodic pressure fluctuations, which shows that the flow field calculation has successfully captured the weak fluctuations of the air. Then, the Fourier transform of the diagram is performed to obtain Figure 4 ,according to Figure 4 The frequency domain diagram of the pressure fluctuation amplitude at the leading edge of the moving blade shows peaks at the blade passing frequencies of 2250 Hz, 4500 Hz, and 6750 Hz, indicating that the pressure fluctuation is consistent with the characteristics of rotating machinery.

[0057] (12) Next, the geometric model is divided into tetrahedral acoustic meshes using ANSYS Meshing to set the global size and near-field area encryption size. The total number of meshes is checked to ensure that the mesh meets the computer computing power requirements and that there are 6-8 mesh units in one sound wave length. Then, the acoustic mesh is exported.

[0058] (13) Import the acoustic mesh into Actran, select the trailing edge of the moving blade and the blade tip gap area as the sound source domain, associate the Ensight file path in the ICFD module, and calculate and generate the time.nff time domain sound source and freq.nff frequency domain sound source files. (14) In the Direct frequency response module, fill in the frequency range of 50-20000Hz, the step size is 50Hz, there are 400 analysis points, and select the Lighthill surface and volume sound source area calculated by ICFD as the sound source. Figure 6 Lighthill surface sound source cloud map, Figure 7 is the Lighthill surface sound source cloud map. At the same time, define the boundary conditions as follows Figure 8 shown.

[0059] (15) Next, monitoring points were set up at 180 locations with a distance of 2.7 m in the axial direction and an interval of 1° in the circumferential direction from 0° to 180°.

[0060] (16) After submitting the calculation, the spectrum of the sound pressure level at the monitoring point is extracted through Actran VI and the appropriate points are selected. The spectrum results are as follows Figure 9 As shown in the far-field sound pressure level spectrum of aerodynamic noise predicted by the flow field and acoustic field simulation calculation model based on Fluent and Actran, the noise peaks appear at the blade passing frequencies of 2250Hz, 4500Hz, and 6750Hz, and the sound pressure levels at the simulated passing frequencies are 78.5dB, 70.8dB, and 67.5dB.

[0061] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any technician familiar with the present profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical content disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement of the above embodiments made according to the technical essence of the present invention, within the spirit and principles of the present invention, without departing from the content of the technical solution of the present invention, shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine, characterized in that: include: A geometric modeling module is used to construct a geometric model, which includes moving blades, stationary blades, a hub, and a shroud. The model ignores non-flow channel structures and maintains a smooth surface. The flow field calculation module is used to build the flow field calculation model. It uses Turbogrid to divide the high-fidelity flow field grid including the blade tip clearance, and then uses ICEM to perform periodic expansion to form a complete circumferential model. Flow field solution module, used to perform steady-state flow field steady solution and transient flow field unsteady solution; Acoustic calculation module, used to build acoustic calculation models, The sound source mapping module is used to select the area with large turbulent energy in the flow field as the Lighthill body sound source and map the flow field data to the sound source; The sound propagation calculation module is configured to set the compressor fluid domain as a finite element grid area, and the outlet boundary is provided with infinite element two-dimensional surface grid units to calculate the sound pressure level spectrum of the far-field monitoring point and obtain the noise situation at the far-field monitoring point.

2. The far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine according to claim 1, characterized in that: The number of moving blades in the geometric modeling module is 25, the number of stationary blades is 38, and the blade tip clearance height is 0.8 mm.

3. The far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine according to claim 1, characterized in that: The flow field calculation module uses Turbogrid to divide the flow channel grid and restores it through ICEM to form a complete circumferential model.

4. The far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine according to claim 1, characterized in that: In the flow field solution module, the number of steps N and the time step of the transient solution satisfy the Nyquist sampling theorem, and the velocity field and density field data are exported in Ensight format.

5. The far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine according to claim 1, characterized in that: The K-oumegaSST turbulence model is used in the flow field solution module, and the convergence residual is set to 1e-5.

6. The far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine according to claim 1, characterized in that: The acoustic prediction model uses a tetrahedral grid, and the grid size in the near field area is refined to 1 / 2-1 / 3 of that in the far field area. The calculation frequency range of the acoustic prediction mode sound is 50-20000 Hz, and the step size is set to 10-100 Hz.

7. The far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine according to claim 1, characterized in that: The body sound source area in the sound source mapping module includes the trailing edge of the moving blade and the blade tip clearance area, and the sound source data includes the time domain file time.nff and the frequency domain file freq.nff.

8. The far-field noise simulation prediction model for a single-stage axial flow compressor component of a turboshaft engine according to claim 1, characterized in that: In the sound propagation calculation module, the boundary of the infinite element surface grid unit is set at the outlet, and the monitoring points are arranged on a semicircle with the center of the outlet surface as the origin, with an axial distance of 2.7m and a circumferential interval of 1°.

Images