A vehicle road transient fatigue key modal identification method and system

Abstract

The application relates to a vehicle road transient fatigue key modal identification method and system, S1: obtaining full-order frequency domain responses of a concerned position; S2: determining a power spectrum density cutoff value, determining a concerned frequency interval based on a curve segment corresponding to power spectrum density greater than or equal to the power spectrum density cutoff value in the full-order frequency domain responses; S3: selecting a certain concerned frequency interval, deleting modal stress results and modal coordinate result data corresponding to the concerned frequency interval, then obtaining each-order time domain responses and superimposing the same to obtain non-full-order time domain responses corresponding to the concerned frequency interval; S4: repeating S3 until non-full-order time domain responses corresponding to all concerned frequency intervals are obtained; S5: obtaining fatigue damage values of full-order time domain responses and fatigue damage values of each non-full-order time domain response, and the deleted modal characteristics corresponding to the maximum non-full-order time domain response fatigue damage contribution degree are key modes. The application accurately and efficiently calculates and identifies key modes of vehicle risk hotspots.

Description

Technical Field

[0001] This invention relates to the field of strength and durability technology, specifically to computer-aided design for automobiles. Background Technology

[0002] With the development of computer technology and the demand for lightweight automotive design, modern automotive product development is gradually shifting from experience-based design to precision design, avoiding redundant designs to achieve low energy consumption and high economic efficiency. Directly verifying designs on user roads or optimizing designs through test track road surface simulation is the most user-friendly approach. It can examine the impact of different road surfaces and frequency domain characteristics on the durability performance of vehicle structures and is currently the mainstream solution for domestic and international automakers to develop overall vehicle durability performance. Used in the early stages of product development, it allows for the conduct of transient fatigue simulation analysis and optimization of vehicles on roads, achieving durability performance targets for the entire vehicle system and components, reducing the risk of durability problems in prototypes and the market, and shortening the product development cycle.

[0003] While vehicle-to-road transient fatigue analysis offers high accuracy, its complex analysis process and low computational efficiency limit its widespread application in engineering. In practical engineering, modal superposition methods are typically used to improve analysis efficiency. However, processing the results and identifying problems using this modal superposition method often requires engineers' experience and repeated calculations.

[0004] Existing methods employ a fixed frequency bandwidth and an iterative approach involving global step-by-step deletion to calculate the contribution of each frequency to actual fatigue damage, thereby statistically determining the critical modes. Each iteration of this method requires a complete fatigue calculation process, resulting in low efficiency and high resource consumption in both computer software and hardware. Therefore, identifying the critical modes that contribute the most to fatigue damage from hundreds or thousands of modes, and obtaining their frequency, mode shape, and other characteristics as the basis for vehicle structural risk identification and optimization, is a dual requirement for both quality and efficiency in road transient fatigue analysis. Summary of the Invention

[0005] One objective of this invention is to provide a method for identifying key modes of transient fatigue in vehicles on the road, in order to solve the problems of low efficiency and high computer software and hardware occupancy in the prior art; another objective is to provide a system for identifying key modes of transient fatigue in vehicles on the road.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for identifying key modes of transient fatigue in vehicles on roads, wherein the identification method specifically comprises:

[0008] S1: Obtain the full-order frequency domain response of the location of interest;

[0009] S2: Determine the power spectral density cutoff value. Based on the curve segment corresponding to the power spectral density whose ordinate is greater than or equal to the power spectral density cutoff value in the full-order frequency domain response, determine the frequency range of interest.

[0010] S3: Select a frequency range of interest, delete the modal stress results and modal coordinate results corresponding to that frequency range, then obtain the time domain responses of each order and superimpose them to obtain the non-full-order time domain response corresponding to that frequency range of interest;

[0011] S4: Repeat S3 until the non-full-order time domain response corresponding to all frequency ranges of interest is obtained;

[0012] S5: Obtain the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response. The deleted modal feature corresponding to the fatigue damage contribution of the maximum non-full-order time domain response is the key mode.

[0013] Based on the above technical means, this method determines multiple frequency ranges of interest by determining the cutoff value. Then, for a single frequency range of interest, the modal stress results and modal coordinate results of that frequency range are deleted. Then, the non-full-order time domain response corresponding to that frequency range of interest is obtained, thereby forming the non-full-order time domain response corresponding to all frequency ranges of interest. Finally, the key mode is determined by obtaining the fatigue damage contribution of the maximum non-full-order time domain response.

[0014] This method selects the frequency range of interest and deletes the modes corresponding to that range, eliminating the need to delete each mode individually. This improves computational efficiency and reduces the utilization of computer software and hardware resources. Furthermore, by using actual response characteristics as a reference, it ensures the accuracy of key mode identification, making it more scientific and accurate than iterative methods that involve gradual deletion.

[0015] Furthermore, the method for obtaining the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response in S5 is as follows: based on the full-order frequency domain response and each non-full-order time domain response, the original SN curve of the position of interest is reconstructed; based on the reconstructed SN curve, the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response are obtained through a pseudo-damage fatigue calculation method.

[0016] Based on the above technical means, compared with the existing technology for obtaining the true fatigue damage value, this method uses pseudo-damage fatigue calculation to calculate the fatigue damage contribution corresponding to the focus frequency. Based on the actual response characteristics, it improves the efficiency of identifying the key modes of the vehicle's focus position without reducing the accuracy.

[0017] Furthermore, the method for reconstructing the original SN curve of the location of interest is as follows: based on the full-order frequency domain response and each non-full-order time domain response, obtain the full-order stress amplitude-cycle number matrix and each non-full-order stress amplitude-cycle number matrix, statistically determine the maximum stress amplitude in the full-order stress amplitude-cycle number matrix and the non-full-order stress amplitude-cycle number matrix, correct the stress amplitude, and combine the corrected stress amplitude to translate and reconstruct the original SN curve to obtain the reconstructed SN curve of the location of interest.

[0018] Furthermore, the method for determining the power spectral density cutoff value in S2 specifically involves: obtaining the maximum value of the power spectral density in the full-order frequency domain response, and multiplying the maximum value of the power spectral density by the cutoff factor to obtain the cutoff value.

[0019] Furthermore, the frequency value corresponding to the peak of the curve segment is recorded, and the frequency value is corrected in combination with the given frequency bandwidth to obtain the frequency range of interest.

[0020] Furthermore, S1 specifically involves: performing stress calculations for transient fatigue of the vehicle on the road using finite element software; obtaining the time-domain responses of each frequency at the location of interest based on the modal stress and modal coordinate results; finally obtaining the full-order time-domain response at the location of interest through linear superposition; and converting the full-order time-domain response at the location of interest into the full-order frequency-domain response using the fast Fourier transform method.

[0021] A vehicle road transient fatigue key mode recognition system based on the above recognition method includes a full-order frequency domain response acquisition module configured to acquire the full-order frequency domain response of the location of interest.

[0022] The frequency range acquisition module is configured to determine the power spectral density cutoff value. Based on the curve segment corresponding to the power spectral density that is greater than or equal to the power spectral density cutoff value in the full-order frequency domain response, the frequency range of interest is determined.

[0023] The non-full-order time domain response acquisition module is configured to acquire the non-full-order time domain response corresponding to all frequency intervals of interest. The method for acquiring the non-full-order time domain response corresponding to a single frequency interval of interest is as follows: select a frequency interval of interest, delete the modal stress result and modal coordinate result data corresponding to the frequency interval of interest, and then acquire and superimpose the time domain responses of each order to obtain the non-full-order time domain response corresponding to the frequency interval of interest.

[0024] The key mode acquisition module is configured to acquire the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response. The deleted modal feature corresponding to the fatigue damage contribution of the maximum non-full-order time domain response is the key mode.

[0025] Furthermore, the method by which the key mode acquisition module obtains the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response is as follows: based on the full-order frequency domain response and each non-full-order time domain response, the original SN curve of the position of interest is reconstructed; based on the reconstructed SN curve, the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response are obtained through a pseudo-damage fatigue calculation method.

[0026] Furthermore, the frequency range acquisition module obtains the maximum value of the power spectral density in the full-order frequency domain response, multiplies the maximum value of the power spectral density by a cutoff factor to obtain the cutoff value, records the frequency value corresponding to the peak of the curve segment, and corrects the frequency value in combination with the given frequency bandwidth to obtain the frequency range of interest.

[0027] Furthermore, the method by which the full-order frequency domain response acquisition module obtains the full-order frequency domain response of the location of interest is as follows: the stress calculation of the transient fatigue of the vehicle road is completed by finite element software, and the time domain response of each order of the location of interest is obtained according to the modal stress result and modal coordinate result. Finally, the full-order time domain response of the location of interest is obtained by linear superposition, and the full-order time domain response of the location of interest is converted into the full-order frequency domain response by the fast Fourier transform method.

[0028] The beneficial effects of this invention are:

[0029] This invention is based on the structural response location of the frequency of concern and uses a pseudo-damage calculation method to calculate the fatigue damage contribution corresponding to the frequency of concern. This allows for accurate and efficient calculation and identification of the key modes of vehicle risk hotspots, providing theoretical support for vehicle structural risk analysis and optimization.

[0030] Compared to existing technologies, this invention reduces the entire fatigue iteration calculation process. Based on structural response and pseudo-damage calculation methods, it achieves rapid and accurate identification of key transient fatigue modes of vehicles in road conditions, thus providing theoretical support for structural risk analysis and optimization. Compared to conventional iterative methods, the method of this invention improves the efficiency of identifying individual risk hotspots. Furthermore, the fatigue calculation in this invention does not rely on commercial fatigue calculation software, thereby significantly reducing the cost of computer software and hardware, achieving the effects of improving work efficiency and reducing usage costs. Attached Figure Description

[0031] Figure 1 This is a flowchart of Example 1;

[0032] Figure 2 This is a schematic diagram of the full-order time-domain response synthesis described in Example 1;

[0033] Figure 3 This is the flowchart for the frequency of interest identification described in Example 1;

[0034] Figure 4 This is a schematic diagram of the attention frequency identification described in Example 1;

[0035] Figure 5 This is the flowchart of non-full-order time-domain response synthesis described in Example 1;

[0036] Figure 6 This is a schematic diagram of non-full-order time-domain response synthesis as described in Example 1;

[0037] Figure 7 This is a case study of the reconstructed SN curve described in Example 1;

[0038] Figure 8 This is a structural diagram of the identification system described in Example 2.

[0039] Among them, 1-full-order frequency domain response acquisition module; 2-interested frequency range acquisition module; 3-non-full-order time domain response acquisition module; 4-key mode acquisition module. Detailed Implementation

[0040] The following description, with reference to the accompanying drawings and preferred embodiments, illustrates the implementation of the technical solution of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0041] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0042] Example 1

[0043] This embodiment proposes a method for identifying key modes of transient fatigue in vehicles on roads, such as... Figure 1 As shown, the specific identification method is as follows:

[0044] S1: Obtain the full-order frequency domain response of the location of interest. The location of interest is the single hot spot that requires modal identification, such as a sheet metal location on a vehicle body.

[0045] In this step, specifically, the stress calculation for transient fatigue of the vehicle under road conditions is first performed using finite element software, including but not limited to Nastran software, with solution type SOL112. The stress calculation output consists of two parts: modal stress results and modal coordinate results, which are output in f06 and pch file formats, respectively. Then, stress recovery is performed based on the modal stress and modal coordinate results to obtain the time-domain response of each frequency at the location of interest. Finally, the full-order time-domain response at the location of interest is obtained through linear superposition. Figure 2 As shown, A1 to An represent the time-domain responses of each order at the location of interest. The full-order time-domain response at the location of interest is obtained by linear superposition. Then, the full-order time-domain response at the location of interest is transformed into the full-order frequency-domain response using the Fast Fourier Transform method.

[0046] S2: Determine the power spectral density cutoff value. Based on the curve segment corresponding to the power spectral density that is greater than or equal to the power spectral density cutoff value in the full-order frequency domain response, determine the frequency range of interest.

[0047] The method for determining the power spectral density cutoff value is as follows: Obtain the maximum value of the power spectral density in the full-order frequency domain response, and multiply this maximum value by a cutoff factor to obtain the cutoff value. Then, record the frequency value corresponding to the peak of the curve segment, and correct the frequency value based on the given frequency bandwidth to obtain the frequency range of interest. The correction method is as follows: truncate the curve segment with the cutoff value as the ordinate, and simultaneously take half of the given frequency bandwidth on both sides of the frequency value as the reference to obtain the frequency range of interest.

[0048] This article uses a specific model's popular sheet metal sections as an example to illustrate how to calculate the frequency of attention for these sections. Figure 3 As shown, step S1 uses the Fast Fourier Transform method to transform the full-order time-domain response of the location of interest into the full-order frequency-domain response, and the result is as follows. Figure 4 The curve in the figure is shown.

[0049] Find the maximum power spectral density and its corresponding point from the full-order frequency domain response, such as Figure 4 P1, as shown, uses the maximum power spectral density multiplied by a cutoff factor to truncate the full-order frequency response. In this embodiment, the cutoff factor is generally within the range (0.5, 1.0). Figure 4 The maximum value of the medium power spectral density is 5493, and its cutoff line is the straight line corresponding to the vertical coordinate value of 3600.5 in the figure.

[0050] Record the curve segments where the power spectral density value exceeds the cutoff value, and find the frequency value corresponding to the peak of the peak, i.e., ... Figure 4The frequency value corresponding to point P1 is 10.5Hz. The frequency value obtained in step S23 is corrected based on the given frequency bandwidth to obtain the frequency range of interest. The frequency bandwidth is generally 1Hz to 5Hz. Figure 4 The frequency range △1 is shown in the figure. Repeat the above steps to obtain other frequency ranges of interest, as shown in the figure. Figure 4 The frequency ranges △2, △3, and △4 are shown in the figure.

[0051] S3: Select a frequency range of interest, delete the modal stress results and modal coordinate results corresponding to that frequency range, then obtain the time domain responses of each order and superimpose them to obtain the non-full-order time domain response corresponding to that frequency range of interest.

[0052] Specifically, such as Figure 5 and Figure 6 As shown, for each frequency range of interest obtained in S2, the data corresponding to the frequency of interest is searched in the modal stress and modal coordinate results in S1. The corresponding modal stress and modal coordinate results within the frequency range of interest are deleted, and the truncated time-domain responses of each order are obtained through stress recovery. The remaining time-domain responses after truncation and deletion are linearly superimposed to obtain the non-full-order time-domain response corresponding to the frequency range of interest.

[0053] For example, Figure 6 As shown, A1 to An represent the various modes of the full-time response. Figure 4 The modal stress results and modal coordinates corresponding to the frequency range △1 of interest are A2 to A3. Then, the time domain responses corresponding to the frequencies of interest, A2 to A3, are deleted from the time domain responses from A1 to An. The remaining time domain responses are linearly superimposed to obtain the non-full-order time domain response corresponding to the frequencies of interest.

[0054] Repeat the above steps until the non-full-order time-domain responses corresponding to all frequency intervals of interest, △1, △2, △3, and △4, are obtained.

[0055] S4: Repeat S3 until the non-full-order time domain response corresponding to all frequency ranges of interest is obtained.

[0056] S5: Obtain the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response. The deleted modal feature corresponding to the fatigue damage contribution of the maximum non-full-order time domain response is the key mode.

[0057] In this step, the original SN curve is first shifted based on the rainflow statistics of the full-order time-domain response and the non-full-order time-domain response to obtain the reconstructed SN curve.

[0058] Specifically, the full-order time-domain response obtained in step 1 and the various non-full-order time-domain responses obtained in step 3 are respectively subjected to rainflow calculations to obtain the full-order stress amplitude-cycle number matrix and the non-full-order stress amplitude-cycle number matrix. The maximum stress amplitude in the full-order stress amplitude-cycle number matrix and the non-full-order stress amplitude-cycle number matrix are statistically analyzed, and multiplied by an amplitude amplification factor to obtain the corrected stress amplitude. In this embodiment, the amplitude amplification factor is generally within the range (1.0, 2.0). Based on the material or analysis category of the location of interest, a corresponding original SN curve is preset, such as sheet metal fatigue, weld fatigue, and weld seam fatigue of steel or aluminum alloy. The original SN curve does not include the infinite life stage, and its curve form includes, but is not limited to, a single logarithmic form. The original SN curve is translated and reconstructed by combining the corrected stress amplitude to obtain the reconstructed SN curve for subsequent fatigue damage calculation.

[0059] An example is given using the reconstructed logarithmic SN curve of sheet metal fatigue data for a certain vehicle model. Figure 7 As shown. The original SN curve is as follows. Figure 7 As shown by the dashed line, the original SN curve is shifted by correcting the stress amplitude c to obtain the reconstructed SN curve, as shown in the figure. Figure 7 As shown by the solid line in the image.

[0060] The expressions for the original SN curve (Formula 1) and the reconstructed SN curve (Formula 2) are as follows:

[0061] S = alg(N) + b (Formula 1)

[0062] S′=alg(N)+c (Formula 2)

[0063] in:

[0064] S—Original stress amplitude;

[0065] a—Slope of the original SN curve;

[0066] N—Number of iterations;

[0067] b—the constant term of the original SN curve;

[0068] S′——Reconstructed stress amplitude;

[0069] c — the constant term of the reconstructed SN curve, i.e., the corrected stress amplitude.

[0070] Specifically, based on the reconstructed SN curve in S3 and the rainflow statistics results from the above steps (i.e., the full-order stress amplitude-cycle number matrix and the non-full-order stress amplitude-cycle number matrix), pseudo-damage fatigue calculations are performed to obtain the full-order fatigue damage value and the fatigue damage values ​​for each non-full-order stage. The ratio of the non-full-order fatigue damage value to the full-order fatigue damage value is calculated to obtain the non-full-order fatigue damage contribution. All non-full-order fatigue damage contributions are sorted in descending order, and the frequency range of interest corresponding to the largest non-full-order fatigue damage contribution is the key frequency, and the corresponding modal feature is the key mode.

[0071] The formula for calculating the contribution of non-full-order damage is as follows:

[0072]

[0073] in:

[0074] C i —Contribution of non-full-order damage, in percent.

[0075] D all — Total damage value.

[0076] D i —Incomplete damage.

[0077] In this embodiment, the efficiency of obtaining key modalities at the location of interest using this method is higher than that of existing methods, as shown in Table 1.

[0078] Table 1 compares the efficiency of acquiring key modalities for the same location of interest in different vehicle models.

[0079]

[0080]

[0081] Example 2

[0082] This embodiment proposes a key mode identification system for vehicle road transient fatigue, based on the identification method described in Embodiment 1, such as... Figure 8 As shown, the identification system includes:

[0083] Full-order frequency domain response acquisition module 1 is configured to acquire the full-order frequency domain response of the location of interest.

[0084] The method for the full-order frequency domain response acquisition module 1 to acquire the full-order frequency domain response of the location of interest is as follows: the stress calculation of the transient fatigue of the vehicle road is completed by finite element software, and the time domain response of each order of the location of interest is obtained according to the modal stress result and modal coordinate result. Finally, the full-order time domain response of the location of interest is obtained by linear superposition. The full-order time domain response of the location of interest is converted into the full-order frequency domain response by the fast Fourier transform method.

[0085] The frequency range acquisition module 2 is configured to determine the power spectral density cutoff value. Based on the curve segment corresponding to the power spectral density that is greater than or equal to the power spectral density cutoff value in the full-order frequency domain response, the frequency range of interest is determined.

[0086] The frequency range acquisition module 2 acquires the maximum value of the power spectral density in the full-order frequency domain response, multiplies the maximum value of the power spectral density by a cutoff factor to obtain the cutoff value, records the frequency value corresponding to the peak of the curve segment, and corrects the frequency value in combination with the given frequency bandwidth to obtain the frequency range of interest.

[0087] The non-full-order time domain response acquisition module 3 is configured to acquire the non-full-order time domain response corresponding to all frequency intervals of interest. The method for acquiring the non-full-order time domain response corresponding to a single frequency interval of interest is as follows: select a frequency interval of interest, delete the modal stress result and modal coordinate result data corresponding to the frequency interval of interest, and then acquire and superimpose the time domain responses of each order to obtain the non-full-order time domain response corresponding to the frequency interval of interest.

[0088] The key mode acquisition module 4 is configured to acquire the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response. The deleted modal feature corresponding to the fatigue damage contribution of the maximum non-full-order time domain response is the key mode.

[0089] The key mode acquisition module obtains the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response by: reconstructing the original SN curve of the position of interest based on the full-order frequency domain response and each non-full-order time domain response; and obtaining the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response by using the pseudo-damage fatigue calculation method based on the reconstructed SN curve.

[0090] The above embodiments are merely preferred embodiments provided to fully illustrate the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the present invention are all within the scope of protection of the present invention.

Claims

1. A method for identifying key modes of transient fatigue in vehicles on roads, characterized in that: The specific identification method is as follows: S1: Obtain the full-order frequency domain response of the location of interest; S2: Determine the power spectral density cutoff value, and take the curve segment corresponding to the power spectral density whose ordinate is greater than or equal to the power spectral density cutoff value in the full-order frequency domain response as the frequency range of interest. S3: Select a frequency range of interest, delete the modal stress results and modal coordinate results corresponding to that frequency range, then obtain the time domain responses of each order and superimpose them to obtain the non-full-order time domain response corresponding to that frequency range of interest; S4: Repeat S3 until the non-full-order time domain response corresponding to all frequency ranges of interest is obtained; S5: Obtain the fatigue damage value of the full-order time-domain response and the fatigue damage values ​​of each non-full-order time-domain response. The deleted modal feature corresponding to the largest fatigue damage contribution of the non-full-order time-domain response is the key mode, specifically: Rainflow calculations were performed on the full-order time-domain response and each non-full-order time-domain response to obtain the full-order stress amplitude-cycle number matrix and the non-full-order stress amplitude-cycle number matrix. The maximum stress amplitude in the full-order stress amplitude-cycle number matrix and the non-full-order stress amplitude-cycle number matrix is ​​statistically determined, and then multiplied by the amplitude amplification factor to obtain the corrected stress amplitude; Based on the material or analysis category of the location of interest, a corresponding original SN curve is preset, and the original SN curve is translated and reconstructed by combining the modified stress amplitude to obtain the reconstructed SN curve. Based on the reconstructed SN curve, the full-order stress amplitude-cycle number matrix and the non-full-order stress amplitude-cycle number matrix, pseudo-damage fatigue calculation is performed to obtain the full-order fatigue damage value and the non-full-order fatigue damage values. The contribution of non-full-order fatigue damage is obtained by calculating the ratio of non-full-order fatigue damage value to full-order fatigue damage value. All non-full-order fatigue damage contributions are sorted in descending order. The frequency range corresponding to the largest non-full-order fatigue damage contribution is the key frequency, and the corresponding modal feature is the key mode.

2. The identification method according to claim 1, characterized in that: The method for determining the power spectral density cutoff value in S2 is as follows: obtain the maximum value of the power spectral density in the full-order frequency domain response, and multiply the maximum value of the power spectral density by the cutoff factor to obtain the cutoff value.

3. The identification method according to claim 1, characterized in that: Record the frequency value corresponding to the peak of the curve segment, and correct the frequency value in combination with the given frequency bandwidth to obtain the frequency range of interest.

4. The identification method according to claim 1, characterized in that: Specifically, S1 involves: performing stress calculations for transient fatigue of the vehicle on the road using finite element software; obtaining the time-domain responses of each frequency at the location of interest based on the modal stress and modal coordinate results; finally obtaining the full-order time-domain response at the location of interest through linear superposition; and converting the full-order time-domain response at the location of interest into the full-order frequency-domain response using the fast Fourier transform method.

5. A vehicle road transient fatigue key mode recognition system based on the recognition method of any one of claims 1-4, characterized in that: The full-order frequency domain response acquisition module (1) is configured to acquire the full-order frequency domain response of the location of interest; The frequency range acquisition module (2) is configured to determine the power spectral density cutoff value and determine the frequency range of interest based on the curve segment corresponding to the power spectral density that is greater than or equal to the power spectral density cutoff value in the full-order frequency domain response. The non-full-order time domain response acquisition module (3) is configured to acquire the non-full-order time domain response corresponding to all frequency intervals of interest. The method for acquiring the non-full-order time domain response corresponding to a single frequency interval of interest is as follows: select a frequency interval of interest, delete the modal stress result and modal coordinate result data corresponding to the frequency interval of interest, and then acquire and superimpose the time domain responses of each order to obtain the non-full-order time domain response corresponding to the frequency interval of interest. The key mode acquisition module (4) is configured to acquire the fatigue damage value of the full-order time domain response and the fatigue damage value of each non-full-order time domain response. The deleted modal feature corresponding to the fatigue damage contribution of the maximum non-full-order time domain response is the key mode.

6. The identification system according to claim 5, characterized in that: The method for the full-order frequency domain response acquisition module (1) to acquire the full-order frequency domain response of the location of interest is as follows: the stress calculation of the transient fatigue of the vehicle road is completed by finite element software, and the time domain response of each order of the location of interest is obtained according to the modal stress result and modal coordinate result. Finally, the full-order time domain response of the location of interest is obtained by linear superposition, and the full-order time domain response of the location of interest is converted into the full-order frequency domain response by fast Fourier transform method.

Images