An optimization method of a body noise transfer function with a third variable

Abstract

The application discloses a kind of optimization methods of introducing third variable car body noise transfer function, based on finite element software completes the calculation of car body noise transfer function;Extract the sound pressure size curve at human ear in car body noise transfer function, into sound pressure level form;Define target line;Third variable β is defined, and constraint function F1 is established, and target function F2 of variable β is established;Establish the design space Xi of the thickness parameter of main component of car body, material parameter;Define optimization function, iteration number, realize the optimization of noise transfer function.The application solves the optimization problem of noise transfer function in the optimization of car body NVH, effectively improves work efficiency, can avoid the situation that noise sound pressure peak value in 20~200Hz interval generally exists "this disappears the other's growth";Under the premise that all indexes reach standard, through the method, the size thickness etc. of car body structure piece can be redistributed, under the premise that full-band satisfies car body noise transfer function target, reduce overall weight to reduce cost.

Description

Technical Field

[0001] This invention relates to the technical field of vehicle NVH, and more specifically, to an optimization method for the vehicle body noise transfer function by introducing a third variable. Background Technology

[0002] The Noise Transfer Function (NTF) primarily refers to the functional relationship between the input excitation load and the output noise. It is used to evaluate the sensitivity of a structure to vibration-induced noise and is a major factor affecting the overall vehicle noise, vibration, and harshness (NVH) performance. Whether the excitation comes from the road surface or the engine, it is ultimately transmitted to the occupants through the vehicle body. In automotive NVH performance development, the vehicle body's NVH performance is crucial. The NTF is an integrated indicator in vehicle body NVH performance development, directly related to vehicle body modes, rigidity at the origin, local modes of body panels, and internal acoustic cavity characteristics.

[0003] Generally speaking, in vehicle body structure optimization, the noise transfer function (NTF) is very difficult to optimize due to its high performance integration and the "fluctuating" frequency range during the optimization process. It is very challenging to solve and there is an urgent need for optimization methods or algorithms that are highly efficient, consider full-band optimization, and can achieve fine-tuning of parameters within a small range and global compliance.

[0004] Currently, the industry commonly uses methods such as PFpanel (board contribution) and PFGrid (node ​​contribution) to identify the main contributing boards, and ODS (Operating Mode Array) analysis to pinpoint weak points. Then, improvements are made based entirely on engineers' experience through local optimization or stiffness increases. However, while these methods are effective for single peak noise issues, they are largely ineffective for common "bulge" problems—where all excitation points exhibit noticeable bulges at a certain frequency (50Hz is common). Furthermore, given the increasing prevalence of in-vehicle noise issues, lower frequencies cause greater customer discomfort, a crucial area of ​​focus in automotive R&D. Therefore, a highly efficient, automated optimization method or algorithm specifically targeting low-frequency, common-bulge noise transfer function peak noise issues, capable of fine-tuning design parameters for global improvement and avoiding inconsistent optimization results, is of paramount importance in the current automotive R&D system. Summary of the Invention

[0005] To address the aforementioned technical problems in related technologies, this invention studies the characteristics of such problems and specifically provides an optimization method for the vehicle body noise transfer function by introducing a third variable, which can effectively solve the aforementioned intractable problems. It effectively improves the peak noise transfer function in the low-frequency band, increases optimization efficiency, and resolves the "constant fluctuation" problem.

[0006] To achieve the above-mentioned technical objectives, the technical solution of the present invention is implemented as follows:

[0007] An optimization method for the vehicle body noise transfer function by introducing a third variable includes the following steps: S1. Building a Trimmed Bode model of the whole vehicle and an in-vehicle acoustic cavity model based on the preprocessing of the finite element software Hypermesh and the post-processing of the software Nastran, and calculating the vehicle body noise transfer function using the Sol111 solver; S2. Automatically extracting the sound pressure level curve at the human ear from the vehicle body noise transfer function through a program, in Pa, with a frequency range of 20~200Hz, and converting it into sound pressure level form through logarithmic transformation, denoted as SPL, in dB, with a frequency range of 20~200Hz; S3. Defining a target line, where the target sound pressure value in the frequency range of 20~200Hz does not exceed the target line set decibel value Rmax; S4. Defining a third variable β and establishing a constraint function F1, F1 = (SPL / Rmax) - β + 1.0, where β ranges from 0.01 to 1.01, establish the objective function F2 with β as the variable, F2 = 1000β; S5, establish the design space Xi for the thickness parameters and material parameters of the main body components; S6, define the optimization function and iteration number based on the Nastran software, where the optimization function is F0 = α*F1+γ*F2, the iteration number is 10, and call the finite element software to program and implement the optimization of the noise transfer function; S7, view the optimization results through Hyperview software to confirm the optimization effect.

[0008] Furthermore, the sound pressure conversion formula in step S1 is SPL=20*log(P / P0), where P is in Pa and represents the pressure curve obtained by the CAE model calculating the noise transfer function; P0 represents the reference pressure.

[0009] Furthermore, the optimization function F0 in step S6 represents the integration of the constraint function F1 and the objective function F2, where α+γ=1, and α=0.5 by default during initialization.

[0010] The beneficial effects of this invention are as follows: This application effectively solves the problem of noise transfer function optimization in vehicle body NVH optimization, which can effectively improve work efficiency and avoid the situation where the noise sound pressure peak in the 20~200Hz range generally "ebbs and flows"; at the same time, under the premise that all indicators meet the standards, this method can realize the redistribution of parameters such as the size and thickness of vehicle body structural components, and reduce the overall weight and thus reduce costs while meeting the vehicle body noise transfer function target across the entire frequency band. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] The present invention will now be described in further detail with reference to the accompanying drawings.

[0013] Figure 1 This is a flowchart of an optimization method for the vehicle body noise transfer function that introduces a third variable, as described in an embodiment of the present invention.

[0014] Figure 2 It is an unoptimized sound pressure level curve;

[0015] Figure 3 This is a sound pressure level curve optimized using this invention;

[0016] Figure 4 Yes Figure 2 Code diagram for noise transfer function optimization;

[0017] Figure 5 This is a schematic diagram of the finite element model of the TB car body;

[0018] Figure 6 This is a schematic diagram of the TB acoustic cavity finite element model;

[0019] Figure 7 This is a structural diagram of the main structural components of the car body with thickness as a variable;

[0020] Figure 8 It is a diagram of the calculation file for the optimization iteration process. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0022] like Figure 1 As shown in the embodiment of the present invention, an optimization method for the vehicle body noise transfer function by introducing a third variable includes the following steps: S1, building a whole vehicle Trimmed Bode model and an in-vehicle acoustic cavity model based on the preprocessing of the finite element software Hypermesh and the post-processing of the software Nastran, and calculating the vehicle body noise transfer function using the Sol111 solver; S2, automatically extracting the sound pressure level curve at the human ear from the vehicle body noise transfer function through a program, in Pa, with a frequency range of 20~200Hz, and converting it into sound pressure level form through logarithmic transformation, the sound pressure level value is denoted as SPL, in dB, with a frequency range of 20~200Hz; S3, defining a target line, where the target sound pressure value in the frequency range of 20~200Hz is not higher than the target line set decibel value Rmax; S4, defining a third variable β, and establishing a constraint function F1, F1 = (SPL / Rmax) - β +1.0, where the value of β ranges from 0.01 to 1.01, establish the objective function F2 with β as the variable, F2 = 1000β; S5, establish the design space Xi for the thickness parameters and material parameters of the main body components; S6, define the optimization function and iteration number based on the software Nastran, where the optimization function is F0 = α*F1+γ*F2, the iteration number is 10, and call the finite element software to program and implement the optimization of the noise transfer function; S7, view the optimization results through the Hyperview software to confirm the optimization effect.

[0023] In one specific embodiment of the present invention, such as Figure 5-6 As shown, a Trimmed Bode model of the whole vehicle and an in-vehicle acoustic cavity model are built using the preprocessing of the finite element software Hypermesh and the postprocessing of the software Nastran. The vehicle body noise transfer function is then calculated using the Sol111 solver.

[0024] In one specific embodiment of the present invention, such as Figure 4 As shown, a program is developed to automatically extract the sound pressure level curve at the human ear (usually the right ear of the driver and the left ear of the passenger) from the vehicle noise transfer function. The unit is Pa, and the frequency range is 20~200Hz (this range is the main frequency range of vehicle structural noise and booming, and is the key frequency range for acoustic-structure interaction analysis). Figure 2The diagram shows the conversion to sound pressure level (SPL) through logarithmic transformation. The SPL value is denoted as SPL, with units of dB and a frequency range of 20~200Hz. The SPL conversion formula is SPL=20*log(P / P0), where P is in Pa and represents the pressure curve obtained by calculating the noise transfer function using the CAE model. P0 represents the reference pressure, typically the pressure value corresponding to 0 dB noise, usually P0=2e-10 Pa.

[0025] In one specific embodiment of the present invention, a target line is defined, such as... Figure 3 As shown, the target sound pressure level should not exceed the target line set decibel value Rmax in the frequency range of 20~200Hz (the target line set decibel value is RmaxdB, requiring that the actual noise level SPL dB of the driver's inner ear obtained from the vehicle noise transfer function calculated by the CAE model should not exceed RmaxdB). Rmax is generally taken as 64dB or 65dB. Figure 2 or Figure 3 As shown, the target line is set to a decibel value Rmax of 64 dB.

[0026] In a specific embodiment of the present invention, a third variable β is defined, and a constraint function F1 is established, F1 = (SPL / Rmax) - β + 1.0, where the value of β ranges from 0.01 to 1.01. An objective function F2 is established with β as the variable, F2 = 1000β (F2 is the positive correlation function of β. The reason for increasing it by 1000 times is to increase the weight ratio of F2 in the optimization function).

[0027] In one specific embodiment of the present invention, such as Figure 7 As shown, the design space Xi for the thickness parameters of the main body components is established.

[0028] In one specific embodiment of the present invention, an optimization function and the number of iterations are defined based on the software Nastran, wherein the optimization function is F0 = α*F1+γ*F2, the number of iterations is 10, and finite element software is called to programmatically implement the optimization of the noise transfer function (e.g. Figure 8 The file shown illustrates the calculation process, illustrating the optimization iteration process, the number of iterations, and the changes in thickness. F0 represents the integration of constraint function F1 and objective function F2. Since the program internally calculates gradient optimization by weighting the constraint and objective functions, considering factors such as function optimization convergence, optimization effect, and optimization efficiency, the final optimization function is defined as F0 = α*F1 + γ*F2, where α + γ = 1. The default initialization value is α = 0.5. Adjusting the value of α improves the optimization effect and efficiency. Figure 2 The target sound pressure level shown has an RMAX of 64 dB in the 20 Hz to 200 Hz range. Figure 2As can be seen from the data, the sound pressure level curves converted from the noise transfer function exceed the target values ​​in the 50Hz and 130-150Hz ranges. The noise transfer function is optimized using the method described in this application. Figure 4 As shown, the aforementioned logical framework is implemented through programming, and commercial software is called to achieve automatic optimization of the noise transfer function.

[0029] In one specific embodiment of the present invention, the optimization results are viewed and the optimization effect is confirmed using Hyperview software, such as... Figure 3 As shown, the noise transfer function was optimized, and the optimized sound pressure decibel values ​​were all below the target line. The "one-to-one" situation was effectively resolved during the iteration process.

[0030] In summary, this application has the following advantages: it transforms a multi-indicator problem (such as inner ear noise of front and back row members) into a single-indicator problem; it innovatively introduces a third variable β as the objective value, solving the problem of not being able to extract the optimization peak point, and using all peaks as optimization targets; it establishes an objective function about the introduced third variable β, and by increasing the weight of the β function in the "objective-constraint" function, it quickly completes the optimization, reduces the number of iterations, and makes all peaks on the curve satisfy the objective value.

[0031] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method of optimizing a body noise transfer function with introduction of a third variable, characterized in that, Comprising the following steps: S1, based on the finite element software Hypermesh software pretreatment and Nastran software post-processing to build Trimmed Bode model and the sound cavity model in the vehicle, and through the Sol111 solver to complete the calculation of the body noise transfer function; S2, through the program to automatically extract the sound pressure curve of the human ear in the body noise transfer function, unit Pa, frequency range of 20~200Hz, through the logarithmic transformation, into the sound pressure level form, sound pressure level value is recorded as SPL, unit dB, frequency range of 20~200Hz, SPL=20*log(P / P0), wherein P unit Pa, indicates the pressure curve obtained by the CAE model calculation noise transfer function, P0 represents the reference pressure; S3, define the target line, the sound pressure target value in the frequency range of 20~200Hz is not higher than the target line set decibel value Rmax; S4, define the third variable β, and establish the constraint function F1, F1=(SPL / Rmax) -β+ 1.0, wherein the value range of β is 0.01~1.01, the objective function F2 of the variable β is established, F2=1000β; S5, establish the design space Xi of the thickness parameters and material parameters of the main components of the vehicle body; S6, based on the software Nastran to define the optimization function and the number of iterations, wherein the optimization function is F0 =α*F1+γ*F2, the number of iterations is 10 times, and the finite element software is called to realize the optimization of the noise transfer function; The optimization function F0 in step S6 represents the integration of the constraint function F1 and the objective function F2, wherein α+γ=1, and the initialization calculation defaults α=0.5; S7, view the optimization results through the Hyperview software, and confirm the optimization effect.

Images