Complete vehicle cavity noise prediction method

Abstract

The invention belongs to the technical field of whole vehicle noise prediction, and particularly relates to a whole vehicle cavity noise prediction method. Establishing a cavity three-dimensional geometry, and measuring cavity characteristic parameters; obtaining a simplified model of the closed cavity with the error meeting the requirement; establishing a cavity noise database, predicting the cavity noise through the cavity characteristic parameters, adding the predicted cavity noise into the whole vehicle wind noise calculation model, and predicting the influence of the cavity noise on the whole vehicle wind noise; obtaining a cavity wind noise contribution amount, calculating a correction coefficient, and multiplying the cavity noise prediction data of any to-be-predicted vehicle model by the correction coefficient to obtain predicted whole vehicle cavity noise of the to-be-predicted vehicle model; according to the method, the cavity wind noise level is predicted in the design stage, whole vehicle wind noise control is facilitated, the wind noise problem caused by the cavity noise is avoided, and meanwhile the influence of the cavity noise on the whole vehicle wind noise level can be predicted.

Description

Technical Field

[0001] The invention belongs to the technical field of whole vehicle noise prediction, and in particular relates to a whole vehicle cavity noise prediction method. Background Art

[0002] Wind noise is the main noise source when the vehicle is driving at high speed, which directly affects the quality of driving activities such as conversation and listening to music. There are three basic methods for wind noise research: theoretical analysis, experiment and numerical calculation. Among them, the research on wind noise of the whole vehicle mainly adopts the methods of numerical calculation and experimental analysis, and the research on the mechanism of wind noise is progressing slowly. The experimental analysis mainly tests the wind noise of the whole vehicle in a wind tunnel and evaluates the wind noise level through local contribution; the numerical calculation method mainly simulates the characteristics of the sound source, and cannot analyze the leakage noise caused by airtightness and the cavity noise caused by gaps and step differences.

[0003] In summary, cavity noise lacks prediction capabilities at the data stage, and currently there is no method in this field that can effectively predict the cavity noise of the entire vehicle. Summary of the invention

[0004] In order to overcome the above problems, the present invention provides a method for predicting the cavity noise of a whole vehicle; a cavity noise database is established, the cavity noise is predicted by cavity characteristic parameters, the predicted cavity noise is added to a whole vehicle wind noise calculation model, and the influence of cavity noise on the wind noise of the whole vehicle is predicted.

[0005] A method for predicting the cavity noise of a vehicle comprises the following steps:

[0006] Step 1 includes the following:

[0007] Step 1.1, import the vehicle door and back door models into the software through the Fileinput function in the Ansa software; then use the MESH-mesh generation-STL function to divide the mesh into triangular meshes; use the ShellMesh-Plane Cut function to perform equidistant cuts on the door and back door in the x, y, and z directions, turn on Releaseresult edges in the cutting settings, and obtain the cavity feature lines;

[0008] Step 1.2, convert the obtained cavity feature lines into solid curves through the TOPO-Curves-Cons2Curves function, and then connect the obtained solid curve boundaries to form a three-dimensional surface through the TOPO-Faces-New function to obtain the cavity three-dimensional geometry;

[0009] Step 1.3, measuring cavity characteristic parameters of the cavity three-dimensional geometry;

[0010] Step 2 includes the following:

[0011] Step 2.1, repair the cavity 3D geometry through the TOPO-Faces function to meet the fluid calculation requirements, then create a surface through the ANSA-surface function, and connect the surface with the repaired cavity geometry to form a closed cavity simplified model that meets the fluid calculation requirements;

[0012] Step 2.2, import the closed cavity simplified model into FIPS software, set the size of the encrypted area of ​​the cavity simplified model, the encrypted size body grid, select the turbulence model and boundary conditions, form the cavity noise calculation model, and submit the calculation to obtain the cavity noise;

[0013] Step 2.3, selecting vehicle door and back door prototypes that are the same as the vehicle door and back door models in step 1.1, and conducting a cavity noise test to obtain a truly measured cavity noise test value;

[0014] Step 2.4, adjusting the size of the encrypted area of ​​the simplified cavity model, the encrypted size body grid, selecting the turbulence model, and the boundary conditions to form a new closed cavity simplified model, submitting the adjusted closed cavity simplified model to calculate the cavity noise, until the error between the simulated cavity noise and the cavity noise test value obtained in step 2.3 meets the requirements; obtaining the closed cavity simplified model whose error meets the requirements;

[0015] Step three includes the following:

[0016] Step 3.1, construct the control body of the closed cavity simplified model obtained in step 2 through the ANSA-morph function, complete the geometric deformation control of the cavity feature parameters of the closed cavity simplified model through the Box-Morphing function, establish the parametric control of the cavity geometric features through the ANSA-morph-Parameter function, and set the upper and lower limits of each cavity feature parameter;

[0017] Step 3.2, using the ANSA-morph-Parameter function, evenly set multiple incremental steps for each cavity feature parameter according to its upper and lower limits, adjust different cavity feature parameters according to the incremental steps, execute the deformation function, and generate a closed cavity calculation simplified model of the new cavity feature parameters;

[0018] Step 3.3, import the closed cavity calculation simplified model with new parameters after deformation into FIPS software, establish noise monitoring points, and extract the noise data at the monitoring points under different vehicle speed conditions as cavity noise data;

[0019] Step 3.4, select different models and perform the operations of steps 1.1 to 3.3 respectively to obtain the cavity noise data of different models under different vehicle speed conditions, and classify and count all the cavity noise data and the corresponding models and cavity characteristic parameters into an excel table to form a data set of cavity calculation parameters and cavity noise;

[0020] Step 3.5, construct a cavity prediction proxy model through the optimization software Isight, and complete the training of the proxy model through the data set obtained in step 3.4;

[0021] Step 4 includes the following:

[0022] Step 4.1, establish a vehicle CFD model of a certain vehicle type, use FIPS software to perform vehicle CFD simulation, and extract the fluid velocity in the cavity of the vehicle door gap through the data processing function in the FIPS software;

[0023] Step 4.2, using CATIA software to measure the characteristic parameters of cavity step difference, cavity depth, cavity width, aspect ratio, cavity volume, and cavity shape for the vehicle data of the vehicle model;

[0024] Step 4.3, importing the fluid velocity of step 4.1 and the cavity characteristic parameters of step 4.2 into the proxy model trained in step 3.5 to obtain the cavity noise prediction data under the whole vehicle state;

[0025] Step 5 includes the following:

[0026] Step 5.1, conduct a whole vehicle acoustic wind tunnel test: first, seal all cavities of the same vehicle model as in step 4, measure the wind noise of the whole vehicle, then tear off the tape at the position of the cavity to be verified, and measure the wind noise of the whole vehicle in the unsealed cavity state;

[0027] Step 5.2, subtract the vehicle wind noise data of the sealed cavity and the unsealed cavity in step 5.1 to obtain the cavity wind noise contribution of the cavity position to be verified, and use the cavity wind noise contribution as the cavity noise measurement data;

[0028] Step 6, calculate the correction coefficient: the correction coefficient is the cavity noise measurement data obtained in step 5.2 divided by the cavity noise prediction data obtained in step 4.3;

[0029] Step 7: Obtain cavity noise prediction data for any vehicle model to be predicted through step 4, and multiply the cavity noise prediction data by the correction coefficient calculated in step 6 to obtain the predicted vehicle cavity noise of the vehicle model to be predicted.

[0030] The step 1.3 uses the Measure function to measure the cavity characteristic parameters of the cavity three-dimensional geometry.

[0031] The cavity characteristic parameters in step 1.3 include front and rear step difference, cavity depth, cavity width, aspect ratio, cavity volume, and cavity shape.

[0032] The step 2.4 is to adjust and form a new closed cavity simplified model by changing the encrypted area size and encrypted size of the closed cavity simplified model in step 2.1 and selecting different turbulence models.

[0033] In the step, 3.2, the ANSA-morph-Parameter function is used to evenly set each cavity feature parameter by 3 to 5 incremental steps according to its upper and lower limits.

[0034] The operating conditions of different vehicle speeds in step 3.3 are 80 kph, 100 kph, 120 kph, 140 kph, and 160 kph respectively.

[0035] In step 3.3, a noise monitoring point is established 1 m away from the cavity.

[0036] In the step 3.5, first add the proxy model Activities-Approximation in the software to the workflow of the software; select RBF as the approximation method of the proxy model, select Data File in the Sampling Option, select the data set obtained in step 3.4, and the error analysis option samples 10 data points by default. Click Continue to complete the proxy model training. The training result R-squared is higher than 0.95, and the model accuracy meets the requirements, and the training of the proxy model is completed.

[0037] When R-squared is lower than 0.95 in step 3.5, the amount of data of the cavity calculation parameters and cavity noise data set used in step 3.4 during training is increased.

[0038] Beneficial effects of the present invention:

[0039] The present invention realizes cavity noise prediction in the data stage and predicts the cavity wind noise level in the design stage, which is beneficial to the wind noise control of the whole vehicle, thereby avoiding the wind noise problem caused by cavity noise, and at the same time can predict the influence of cavity noise on the wind noise level of the whole vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings used in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the contents of the embodiments of the present invention and these drawings without creative work.

[0041] Figure 1 This is a simplified schematic diagram of the cavity model of the present invention.

[0042] Figure 2 It is a schematic diagram of the simulation benchmarking results of the present invention.

[0043] Figure 3 It is a schematic diagram of characteristic parameters of the present invention.

[0044] Figure 4 It is a schematic diagram of the characteristic parameter results of the present invention.

[0045] Figure 5 It is a schematic diagram of the velocity slice of the flow field of the whole vehicle of the present invention.

[0046] Figure 6 It is a schematic diagram of wind noise simulation of the present invention.

[0047] Figure 7 It is a schematic diagram of wind noise simulation results of the present invention. DETAILED DESCRIPTION

[0048] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0049] In the description of the present invention, unless otherwise clearly specified and limited, the terms "connected", "connected", and "fixed" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0050] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0051] In the description of this embodiment, the terms "upper", "lower", "left", "right" and other directions or positional relationships are based on the directions or positional relationships shown in the drawings, and are only for the convenience of description and simplification of operation, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operate in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used to distinguish in the description and have no special meaning.

[0052] Example 1

[0053] A method for predicting vehicle cavity noise, such as Figure 1-7 As shown, the following steps are included:

[0054] Step 1 includes the following:

[0055] Step 1.1, import the vehicle door and back door models into the software through the Fileinput function in the Ansa software; then use the MESH-mesh generation-STL function to divide the mesh into triangular meshes; use the ShellMesh-Plane Cut function to perform equidistant cuts on the door and back door in the x, y, and z directions, turn on Releaseresult edges in the cutting settings, and obtain the cavity feature lines;

[0056] Step 1.2, convert the obtained cavity feature lines into solid curves through the TOPO-Curves-Cons2Curves function, and then connect the obtained solid curve boundaries to form a three-dimensional surface through the TOPO-Faces-New function to obtain the cavity three-dimensional geometry;

[0057] Step 1.3, measuring cavity characteristic parameters of the cavity three-dimensional geometry;

[0058] Step 2 includes the following:

[0059] Step 2.1, continue to use the TOPO-Faces function of the ANSA software to repair the three-dimensional cavity geometry obtained in step 1 to meet the fluid calculation requirements, then use the ANSA-surface function to create a surface, and connect the surface with the repaired cavity geometry to form a closed cavity simplified model that meets the fluid calculation requirements;

[0060] Step 2.2, import the closed cavity simplified model into the FIPS software of the CFD software, set the size of the encrypted area of ​​the cavity simplified model, the encrypted size body grid, select the turbulence model, boundary conditions, form the cavity noise calculation model, and submit the calculation to obtain the cavity noise;

[0061] Step 2.3, carry out cavity noise test calibration, select the vehicle door and back door prototypes with the same vehicle door and back door models as those in step 1.1, carry out cavity noise tests under the same working conditions, obtain the actual measured cavity noise test value, compare it with the cavity noise obtained by simulation calculation, and carry out cavity noise simulation calibration;

[0062] Step 2.4, adjust the size of the encrypted area of ​​the simplified cavity model, encrypt the size body grid, select the turbulence model, and the boundary conditions to form a new closed cavity simplified model, and submit the adjusted closed cavity simplified model to calculate the cavity noise until the error between the simulated cavity noise and the cavity noise test value obtained in step 2.3 meets the requirements; the general error requirement is less than 5%; obtain the closed cavity simplified model with the error meeting the requirements;

[0063] Step three includes the following:

[0064] Step 3.1, construct the control body of the closed cavity simplified model obtained in step 2 through the ANSA-morph function of the ANSA software, complete the geometric deformation control of the cavity feature parameters of the closed cavity simplified model through the Box-Morphing function, establish the parametric control of the cavity geometric features through the ANSA-morph-Parameter function, and set the upper and lower limits of each cavity feature parameter according to the cavity features on the whole vehicle;

[0065] Step 3.2, using the ANSA-morph-Parameter function, evenly set multiple incremental steps for each cavity feature parameter according to its upper and lower limits, adjust different cavity feature parameters according to the incremental steps, execute the deformation function, and generate a closed cavity calculation simplified model of the new cavity feature parameters;

[0066] Step 3.3, import the closed cavity calculation simplified model with new parameters after deformation into FIPS software, set the volume grid, boundary conditions, and establish noise monitoring points, and extract the noise data at the monitoring points under different vehicle speed conditions as cavity noise data;

[0067] Step 3.4, select different models and perform the operations of steps 1.1 to 3.3 respectively to obtain the cavity noise data of different models under different vehicle speed conditions, and classify and count all the cavity noise data and the corresponding models and cavity characteristic parameters into an excel table to form a data set of cavity calculation parameters and cavity noise;

[0068] Step 3.5, construct a cavity prediction proxy model through the optimization software Isight, and complete the training of the proxy model through the data set obtained in step 3.4;

[0069] Step 4 includes the following:

[0070] Step 4.1, establish a vehicle CFD model of a certain vehicle type through ANSA software and FIPS software, use FIPS software to perform vehicle CFD simulation, and extract the fluid velocity in the cavity of the door gap of the vehicle through the data processing function in FIPS software;

[0071] Step 4.2, using CATIA software to measure the characteristic parameters of cavity step difference, cavity depth, cavity width, aspect ratio, cavity volume, and cavity shape for the vehicle data of the vehicle model;

[0072] Step 4.3, importing the fluid velocity of step 4.1 and the cavity characteristic parameters of step 4.2 into the proxy model trained in step 3.5 to obtain the cavity noise prediction data under the whole vehicle state;

[0073] Step 5 includes the following:

[0074] Step 5.1, conduct a whole vehicle acoustic wind tunnel test: verify the influence of cavity noise on the whole vehicle state, first seal all cavities of the whole vehicle of the same model as step 4, measure the wind noise of the whole vehicle, then tear off the tape at the position of the cavity to be verified, and measure the wind noise of the whole vehicle in the non-sealed cavity state;

[0075] Step 5.2, subtract the vehicle wind noise data of the sealed cavity and the unsealed cavity in step 5.1 to obtain the cavity wind noise contribution of the cavity position to be verified, and use the cavity wind noise contribution as the cavity noise measurement data;

[0076] Step 6, calculate the correction coefficient: the correction coefficient is the cavity noise measurement data obtained in step 5.2 divided by the cavity noise prediction data obtained in step 4.3;

[0077] Step 7: Obtain cavity noise prediction data for any vehicle model to be predicted through step 4, and multiply the cavity noise prediction data by the correction coefficient calculated in step 6 to obtain the predicted vehicle cavity noise of the vehicle model to be predicted.

[0078] The cavity noise database is updated through simulation and test data, and the cavity noise data of the whole vehicle is updated and iterated according to the Deming cycle principle, so as to improve the prediction accuracy.

[0079] The step 1.3 uses the Measure function to measure the cavity characteristic parameters of the cavity three-dimensional geometry.

[0080] The cavity characteristic parameters in step 1.3 include front and rear step difference, cavity depth, cavity width, aspect ratio, cavity volume, and cavity shape.

[0081] The step 2.4 is to adjust and form a new closed cavity simplified model by changing the encrypted area size and encrypted size of the closed cavity simplified model in step 2.1 and selecting different turbulence models.

[0082] In the step, 3.2, the ANSA-morph-Parameter function is used to evenly set each cavity feature parameter by 3 to 5 incremental steps according to its upper and lower limits.

[0083] The operating conditions of different vehicle speeds in step 3.3 are 80 kph, 100 kph, 120 kph, 140 kph, and 160 kph respectively.

[0084] In step 3.3, a noise monitoring point is established 1 m vertically away from the cavity.

[0085] In the step 3.5, first add the proxy model Activities-Approximation in the software to the workflow of the software; select RBF as the approximation method of the proxy model, select Data File in the Sampling Option, select the data set obtained in step 3.4, and the error analysis option samples 10 data points by default. Click Continue to complete the proxy model training. The training result R-squared is higher than 0.95, and the model accuracy meets the requirements, and the training of the proxy model is completed.

[0086] When R-squared is lower than 0.95 in step 3.5, the amount of data of the cavity calculation parameters and cavity noise data set used in step 3.4 during training is increased.

[0087] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the protection scope of the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention. These simple variations all belong to the protection scope of the present invention.

[0088] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0089] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A method for predicting vehicle cavity noise, characterized in that: The following steps are involved: Step 1 includes the following: Step 1.1, import the vehicle door and back door models into the software through the Fileinput function in Ansa software; then use the MESH-mesh generation-STL function to divide the mesh into triangular meshes; use the Shell Mesh-Plane Cut function to perform equidistant cuts on the door and back door in the x, y, and z directions, open Release resultedges in the cutting settings, and obtain the cavity feature lines; Step 1.2, convert the obtained cavity feature lines into solid curves through the TOPO-Curves-Cons2Curves function, and then connect the obtained solid curve boundaries to form a three-dimensional surface through the TOPO-Faces-New function to obtain the cavity three-dimensional geometry; Step 1.3, measuring cavity characteristic parameters of the cavity three-dimensional geometry; Step 2 includes the following: Step 2.1, repair the cavity 3D geometry through the TOPO-Faces function to meet the fluid calculation requirements, then create a surface through the ANSA-surface function, and connect the surface with the repaired cavity geometry to form a closed cavity simplified model that meets the fluid calculation requirements; Step 2.2, import the closed cavity simplified model into FIPS software, set the size of the encrypted area of ​​the cavity simplified model, the encrypted size body grid, select the turbulence model and boundary conditions, form the cavity noise calculation model, and submit the calculation to obtain the cavity noise; Step 2.3, selecting vehicle door and back door prototypes that are the same as the vehicle door and back door models in step 1.1, and conducting a cavity noise test to obtain a truly measured cavity noise test value; Step 2.4, adjust the size of the encrypted area of ​​the cavity simplified model, encrypt the size of the body mesh, and select turbulence The model and boundary conditions form a new closed cavity simplified model, and the closed cavity simplified model formed after adjustment is submitted to the calculation to obtain the cavity noise, until the error between the cavity noise obtained by simulation and the cavity noise test value obtained in step 2.3 meets the requirements; and the closed cavity simplified model with the error meeting the requirements is obtained; Step three includes the following: Step 3.1, construct the control body of the closed cavity simplified model obtained in step 2 through the ANSA-morph function, complete the geometric deformation control of the cavity feature parameters of the closed cavity simplified model through the Box-Morphing function, establish the parametric control of the cavity geometric features through the ANSA-morph-Parameter function, and set the upper and lower limits of each cavity feature parameter; Step 3.2, using the ANSA-morph-Parameter function, evenly set multiple incremental steps for each cavity feature parameter according to its upper and lower limits, adjust different cavity feature parameters according to the incremental steps, execute the deformation function, and generate a closed cavity calculation simplified model of the new cavity feature parameters; Step 3.3, import the closed cavity calculation simplified model with new parameters after deformation into FIPS software, establish noise monitoring points, and extract the noise data at the monitoring points under different vehicle speed conditions as cavity noise data; Step 3.4, select different models and perform the operations of steps 1.1 to 3.3 respectively to obtain the cavity noise data of different models under different vehicle speed conditions, and classify and count all the cavity noise data and the corresponding models and cavity characteristic parameters into an excel table to form a data set of cavity calculation parameters and cavity noise; Step 3.5, construct a cavity prediction proxy model through the optimization software Isight, and complete the training of the proxy model through the data set obtained in step 3.4; Step 4 includes the following: Step 4.1, establish a vehicle CFD model of a certain vehicle type, use FIPS software to perform vehicle CFD simulation, and extract the fluid velocity in the cavity of the vehicle door gap through the data processing function in the FIPS software; Step 4.2, using CATIA software to measure the characteristic parameters of cavity step difference, cavity depth, cavity width, aspect ratio, cavity volume, and cavity shape for the vehicle data of the vehicle model; Step 4.3, importing the fluid velocity of step 4.1 and the cavity characteristic parameters of step 4.2 into the proxy model trained in step 3.5 to obtain the cavity noise prediction data under the whole vehicle state; Step 5 includes the following: Step 5.1, conduct a whole vehicle acoustic wind tunnel test: first, seal all cavities of the same vehicle model as in step 4, measure the wind noise of the whole vehicle, then tear off the tape at the position of the cavity to be verified, and measure the wind noise of the whole vehicle in the unsealed cavity state; Step 5.2, subtract the vehicle wind noise data of the sealed cavity and the unsealed cavity in step 5.1 to obtain the cavity wind noise contribution of the cavity position to be verified, and use the cavity wind noise contribution as the cavity noise measurement data; Step 6, calculate the correction coefficient: the correction coefficient is the cavity noise measurement data obtained in step 5.2 divided by the cavity noise prediction data obtained in step 4.3; Step 7: Obtain cavity noise prediction data for any vehicle model to be predicted through step 4, and multiply the cavity noise prediction data by the correction coefficient calculated in step 6 to obtain the predicted vehicle cavity noise of the vehicle model to be predicted.

2. The method for predicting vehicle cavity noise according to claim 1, characterized in that: The step 1.3 uses the Measure function to measure the cavity characteristic parameters of the cavity three-dimensional geometry.

3. The method for predicting vehicle cavity noise according to claim 1, characterized in that: The cavity characteristic parameters in step 1.3 include front and rear step difference, cavity depth, cavity width, aspect ratio, cavity volume, and cavity shape.

4. The method for predicting vehicle cavity noise according to claim 1, characterized in that: The step 2.4 is to adjust and form a new closed cavity simplified model by changing the encrypted area size and encrypted size of the closed cavity simplified model in step 2.1 and selecting different turbulence models.

5. The method for predicting vehicle cavity noise according to claim 1, characterized in that: The step 3.2 uses the ANSA-morph-Parameter function to evenly set each cavity feature parameter by 3 to 5 incremental steps according to its upper and lower limits.

6. The method for predicting vehicle cavity noise according to claim 1, characterized in that: The operating conditions of different vehicle speeds in step 3.3 are 80 kph, 100 kph, 120 kph, 140 kph, and 160 kph respectively.

7. The method for predicting vehicle cavity noise according to claim 1, characterized in that: In step 3.3, a noise monitoring point is established 1 m away from the cavity.

8. The method for predicting vehicle cavity noise according to claim 1, characterized in that: In step 3.5, firstly, the proxy model in the software Activities-Approximation is added to the workflow of the software; select RBF as the approximation method of the proxy model, select Data File in Sampling Option, select the data set obtained in step 3.4, and the error analysis option samples 10 data points by default. Click Continue to complete the proxy model training. The training result R-squared is higher than 0.95, and the model accuracy meets the requirements, and the training of the proxy model is completed.

9. The method for predicting the cavity noise of a vehicle according to claim 1, characterized in that: When R-squared is lower than 0.95 in step 3.5, the amount of data of the cavity calculation parameters and cavity noise data set used in step 3.4 during training is increased.

Images