Method for determining multi-task multi-frequency component periodic vibration environmental test conditions

Abstract

This invention provides a method for determining environmental test conditions for multi-task, multi-frequency component periodic vibration, belonging to the field of equipment environmental engineering. It includes the following steps: S1, obtaining a normalized vibration data set X containing n task profiles; S2, obtaining the first and second type components of the measured vibration data; S3, using X as a reference, calculating the equivalent duration t of the spectral values ​​corresponding to the remaining L-1 group of periodic frequencies fa using the following formula. L S4. Obtain a set of parameters with spectral values ​​and duration T; S5. Calculate the duration T of the random spectrum. R Based on this, the duration of the periodic frequency fa is converted to T. R S6. Calculate the converted spectral values; S7. Superimpose the time-unified periodic spectrum or narrowband spectrum with the random spectrum to form the vibration test conditions for vibration testing. This invention can directly determine the frequency of the periodic / narrowband component with higher identification accuracy.

Description

Technical Field

[0001] This invention relates to the field of equipment environmental engineering, specifically to a method for determining the environmental test conditions for multi-task, multi-frequency components of periodic vibration. Background Technology

[0002] Environmental testing conditions are a core component of equipment environmental adaptability design and verification. GJB 150.1A-2009, "Laboratory Environmental Testing Methods for Military Equipment Part 1: General Requirements," specifies the requirements for determining equipment environmental testing conditions: Dynamic environmental parameters such as vibration, shock, noise, and artillery fire are platform-induced environments, and their values ​​depend on the platform type and operating state. When determining dynamic environmental parameters, they should be based as much as possible on measured data from the equipment or similar equipment. The basic principle is to obtain a standard spectrum by statistically summarizing the measured data, and then convert it into equipment environmental testing conditions through engineering processing.

[0003] The measured working conditions include a variety of mission profiles, and the frequency components of the measured data of the platform vibration environment are also quite complex. Taking the vibration measured data of aviation equipment as an example, its characteristics are that it is composed of broadband random components superimposed with broadband periodic components. The broadband random components are generated by the excitation of aerodynamic forces during flight, while the periodic components are generated by periodic vibration sources such as rotating parts (propellers, engines, gearboxes, etc.) on the platform. When the rotation speed of the rotating parts fluctuates, the periodic components exhibit narrow-band characteristics.

[0004] When statistically summarizing measured vibration data of broadband random + periodic components, since the data processing methods used for the two components are different, it is first necessary to identify the broadband random component and the periodic component, then process them separately, and finally synthesize them again to form the vibration environment test conditions. HB 20236-2014 "Methods for Environmental Data Analysis and Processing of Military Aircraft Platforms" specifies the identification method for the periodic component in section 6.4.1.2: "For random vibration signals containing periodic components, several peaks will appear on their power spectral density. If the analysis bandwidth narrows, there are peaks with continuously increasing peak values. Calculating the corresponding peak frequencies can estimate the frequency value of the periodic component." Using this identification method requires repeatedly changing the analysis bandwidth and then visually comparing the changing trends of the peaks, which is time-consuming and labor-intensive, and relies excessively on the experience of technical personnel, thus having certain shortcomings in practical engineering applications.

[0005] Furthermore, changes in the operating conditions of rotating components in equipment can cause corresponding changes in the amplitude of periodic components. When formulating vibration durability test conditions based on measured data, it is necessary to combine and superimpose the periodic components with different amplitudes contained in different mission profiles to obtain comprehensive test conditions. Currently, there are no engineering conversion methods for periodic components involving multiple mission profiles in existing publicly available technical documents, making it difficult to better determine the test conditions for periodic vibration environments with multiple missions and multiple frequency components. Further research on this issue is urgently needed. Summary of the Invention

[0006] To address the shortcomings of the prior art, the present invention aims to provide a method for determining periodic vibration environment test conditions for multiple tasks and multiple frequency components. This method employs time-frequency analysis to identify and separate periodic components in non-stationary vibration signals, solving the problem that traditional methods cannot fully consider the time-varying characteristics of non-stationary vibration signals. It can directly determine the frequency of periodic / narrowband components, achieving higher identification accuracy. Furthermore, it can directly calculate the periodic components of the multi-task profile to obtain accurate periodic vibration environment test conditions for multiple tasks and multiple frequency components.

[0007] Specifically, the present invention provides a method for determining test conditions for periodic vibration environments with multiple tasks and multiple frequency components, comprising the following steps: S1. Standardize and preprocess the measured vibration data corresponding to the n task profiles to obtain a set of standardized vibration data X containing the n task profiles, i.e., X = {x n (t)}, where t is the time domain length; S2, Based on the vibration normalization data set X={x n From the n sets of vibration measured data in (t)}, we obtain the first type component and the second type component of the vibration measured data, where the first type component is a periodic component or a narrowband component, and the second type component is a random component; and we perform spectral analysis on the first type component to obtain the periodic spectrum or narrowband spectrum, and perform spectral analysis on the second type component to obtain the random spectrum. S3. Calculate the same period frequency in all periodic components or narrowband components for a given task profile. The corresponding maximum spectral value Calculate the periodic frequencies of the remaining L-1 groups respectively. The equivalent duration t of the spectral value L The calculation formula is as follows: ; Where m is the vibration durability equivalence factor, calculated based on the SN curve of the structural material, and P LLet L be the spectral value of the Lth component. Maximum spectral value ; S4, Change the periodic frequency The L groups of time values ​​are summed to obtain the spectral value. Duration is A set of parameters: ; S5, with the duration T of the random spectrum R Based on the reference, the period frequency Duration Converted to T R Calculate the converted spectral values The calculation formula is as follows: ; in, The spectral value of the random spectrum; S6. Repeat steps S3 to S5, converting the periodic or narrowband spectra of the n task profiles to have the same duration as the random spectrum and obtaining the converted spectral values. Superimpose the time-unified periodic or narrowband spectrum with the random spectrum to form the vibration test conditions. It was later applied to vibration platforms for vibration testing.

[0008] Preferably, step S2 specifically includes the following sub-steps: S21. Perform time-frequency analysis on the vibration normalized data x(t1) of the first task profile to generate the time spectrum of x(t1). The analysis formula is as follows: ; In the formula, a is a non-zero scaling coefficient, a = 1 / f, b is the translation amount, and CWT is the spectral value of the time spectrum. Let f be a wavelet function, and f be the frequency. S22. Observe the spectrum at x(t1), and analyze the frequencies corresponding to the peaks of the frequency domain curve hour by hour. The analysis formula is as follows: ; In the formula, argmax is the value of the independent variable or parameter that makes a function reach its maximum value; S23, Connecting the peaks at all times. This forms the time-frequency ridge of the vibration signal; S24. Repeat steps S22 and S23 to obtain n time-frequency ridges of the spectrum at x(t1); S25. Analyze the frequency range corresponding to each time-frequency ridge and identify and separate the periodic component / narrowband component and random component. S26. For the vibration normalization data set X={x(t) n For the remaining n-1 sets of vibration measurement data in the task profile, repeat steps S21~S25 to separate the periodic / narrowband and random components of the vibration measurement data for the n task profiles. S27. Perform spectral analysis on the periodic / narrowband and random components of the n task profiles respectively to obtain the periodic spectrum / narrowband spectrum and random spectrum.

[0009] Preferably, in step S1, data standardization refers to unifying the name, unit, and storage format of the vibration measurement data.

[0010] Preferably, in step S1, data preprocessing refers to checking and correcting the normalized vibration measurement data to remove abnormal signals and spurious trend terms.

[0011] Preferably, the time-frequency analysis method in step S21 refers to simultaneously decomposing the vibration measurement data in the time domain and frequency domain for analysis, including Gabor expansion, Cohen-type time-frequency distribution, Radon-Wigner transform, wavelet analysis, or fractional Fourier transform.

[0012] Preferably, the spectral analysis in step S27 specifically involves: using line spectrum analysis for periodic signals; and using power spectrum analysis for narrowband or random signals.

[0013] Preferably, the vibration durability equivalence factor refers to the slope of the SN curve of the structural material in a double logarithmic coordinate system.

[0014] Preferably, the duration T of the random spectrum R This is the result of proportionally adjusting and summing the random vibration times of all task profiles.

[0015] Preferably, step S25 specifically includes the following steps: S251. If the frequency corresponding to the time-frequency ridge is constant, then this component is a periodic component with a frequency fs. ki It is the periodic frequency; S252. If the frequency corresponding to the time-frequency ridge fluctuates within a small range, then extract the upper and lower limits of the floating frequency [fs]. ki fs k(i+1) As a narrowband component; S253, the remaining part of the signal is used as the broadband random part.

[0016] Preferably, the small-range fluctuation in step S252 refers to the fluctuation range being within ±5% of the periodic frequency.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) Compared with the traditional "two-step" method of using variable resolution + FFT transformation to determine the periodic components, the method of determining the periodic vibration environment test conditions of the present invention can quickly and accurately identify and separate the periodic components in the non-stationary vibration signal by using time-frequency analysis. This solves the problem that the traditional method cannot fully consider the time-varying characteristics of the non-stationary vibration signal, and can directly determine the frequency of the periodic / narrowband components, with higher identification accuracy.

[0018] (2) Compared with the traditional "two-step" method of using variable resolution + FFT transformation to identify periodic components, the method for determining the test conditions of periodic vibration environment of the present invention uses time-frequency analysis to process the vibration signal. By analyzing the change state of the time-frequency ridge, the periodic / narrowband components can be directly determined. The method has sufficient theoretical basis, high identification accuracy, high practical operability, and is simple and fast, with very high engineering application value.

[0019] (3) The method for determining the periodic vibration environment test conditions for multiple tasks and multiple frequency components of the present invention first unifies the amplitude of the periodic / narrowband components of multiple tasks with the same frequency, then unifies the time of the periodic / narrowband components and random components of multiple frequencies, and finally superimposes all spectral values ​​to obtain the vibration environment test conditions. Based on the fatigue characteristic parameters of equipment materials, this method solves for the first time the problem of converting the periodic / narrowband components of multiple tasks into vibration environment test conditions. The calculation results are highly reliable, conform to engineering practice, and have strong innovation, effectiveness and practicality. Attached Figure Description

[0020] Figure 1 This is an overall flowchart of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3 This is a schematic diagram illustrating the validity check of data samples under various states in an embodiment of the present invention; Figure 4 This is a schematic diagram of periodic vibrations identified in an embodiment of the present invention; Figure 5 This is a schematic diagram of narrowband random vibrations identified in an embodiment of the present invention; Figure 6 This is a schematic diagram of broadband random vibrations identified in an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the calculation of the upper limit spectrum of set statistics in an embodiment of the present invention; Figure 8 This is a schematic diagram of randomized experimental values ​​in an embodiment of the present invention; Figure 9This is a schematic diagram of the experimental spectrum in an embodiment of the present invention. Detailed Implementation

[0021] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0022] Specifically, this invention provides a method for determining the test conditions of a multi-task, multi-frequency component periodic vibration environment, which can be used to simulate the vibration conditions of a transport platform, such as an aircraft or ship. The method of this invention uses time-frequency analysis to process the vibration signal. By analyzing the changing state of the time-frequency ridge, the periodic / narrowband component can be directly determined. This method has a sound theoretical basis, high identification accuracy, high practical operability, and is simple and quick, possessing very high engineering application value. Figure 1 and Figure 2 As shown, the method specifically includes the following sub-steps: S1. Standardize and preprocess the measured vibration data corresponding to n task profiles to obtain a set of standardized vibration data X containing n task profiles, i.e., X = {x(t n )}, where t is the time domain length.

[0023] In a specific embodiment, data normalization in step S1 refers to standardizing the name, unit, and storage format of the vibration measurement data. Data preprocessing in step S1 refers to checking and correcting the normalized vibration measurement data to remove abnormal signals and spurious trend terms.

[0024] S2, Based on the vibration normalization data set X={x(t n From the n sets of vibration measurement data in the data, we obtain the first type of component and the second type of component of the vibration measurement data. The first type of component is a periodic component or a narrowband component, and the second type of component is a random component. We then perform spectral analysis on the first type of component to obtain the periodic spectrum or the narrowband spectrum, and perform spectral analysis on the second type of component to obtain the random spectrum.

[0025] In a specific embodiment, step S2 specifically includes the following sub-steps: S21. Perform time-frequency analysis on the vibration normalized data x(t1) of the first task profile to generate the time spectrum of x(t1). The analysis formula is as follows: ; In the formula, a is a non-zero scaling coefficient, a = 1 / f, b is the translation amount, and CWT is the spectral value of the time spectrum. Let f be the wavelet function and f be the frequency. The time-frequency analysis method in step S21 refers to simultaneously decomposing the measured vibration data into the time and frequency domains for analysis. Specific analysis methods include Gabor expansion, Cohen-type time-frequency distribution, Radon-Wigner transform, wavelet analysis, or fractional Fourier transform.

[0026] S22. Observe the spectrum at x(t1), and analyze the frequencies corresponding to the peaks of the frequency domain curve hour by hour. The analysis formula is as follows: ; In the formula, argmax is the value of the independent variable or parameter that makes a function reach its maximum value.

[0027] S23. Connect the frequencies corresponding to the peaks at all times. This forms the time-frequency ridge of the vibration signal.

[0028] S24. Repeat steps S22 and S23 to obtain n time-frequency ridges of the spectrum at x(t1).

[0029] S25. Analyze the frequency range corresponding to each time-frequency ridge, and identify and separate the periodic component, narrowband component, and random component.

[0030] In a specific embodiment, step S25 is as follows: S251. If the frequency corresponding to the time-frequency ridge is constant, then this component is a periodic component with a frequency fs. ki It is the periodic frequency.

[0031] S252. If the frequency corresponding to the time-frequency ridge fluctuates within a small range, then extract the upper and lower limits of the floating frequency [fs]. ki fs k(i+1) As a narrowband component; small-range fluctuations refer to fluctuations within ±5% of the period frequency.

[0032] S253, the rest of the signal is treated as a broadband random component.

[0033] S26. For the vibration normalization data set X={x(t) n For the remaining n-1 sets of vibration measurement data in the task profile, repeat steps S21 to S25 to separate the periodic / narrowband and random components of the vibration measurement data for the n task profiles.

[0034] S27. Perform spectral analysis on the periodic / narrowband and random components of the n task profiles respectively to obtain the periodic spectrum / narrowband spectrum and random spectrum. Specifically, the spectral analysis in step S27 is as follows: line spectrum analysis is used for periodic signals; power spectrum analysis is used for narrowband or random signals.

[0035] S3. Calculate the same period frequency in all periodic components or narrowband components for a given task profile. The corresponding maximum spectral value ,by Based on this, the remaining L-1 group period frequencies are calculated using the following formulas. The equivalent duration t of the corresponding spectral value L The calculation formula is as follows: ; Where m is the vibration durability equivalence factor, calculated based on the SN curve of the structural material, P L Let L be the spectral value of the Lth component. Maximum spectral value The duration of vibration; the vibration durability equivalence factor m refers to the slope of the SN curve of the structural material in a double logarithmic coordinate system.

[0036] S4, Change the periodic frequency The spectral values ​​are obtained by summing the L groups of time. And the duration is A set of parameters: ; S5, with the duration T of the random spectrum R Based on the reference, the period frequency Duration Converted to T R Calculate the converted spectral values The calculation formula is as follows: ; in, is the spectral value of the random spectrum.

[0037] In a specific embodiment, the duration T of the random spectrum R This is the result of proportionally adjusting and summing the random vibration times of all task profiles.

[0038] S6. Repeat steps S3 to S5, converting the periodic or narrowband spectra of the n task profiles to have the same duration as the random spectrum and obtaining the converted spectral values. Superimpose the time-unified periodic or narrowband spectrum with the random spectrum to form the vibration test conditions. Conduct vibration tests. Specific implementation examples: This embodiment takes the measured vibration data from a ground engine test of a certain type of accessory as the research object, and provides a method for determining the test conditions of the periodic vibration environment of a ground engine with multiple missions and multiple frequency components. The corresponding operating conditions are shown in the table below.

[0040] The method for determining the test conditions for the cyclic vibration environment of a ground engine with multiple missions and multiple frequency components in this embodiment specifically includes the following steps: S1. Standardize and preprocess the vibration measurement data corresponding to n mission profiles of the ground engine. First, divide the measured vibration data into samples according to typical operating conditions. The division results are shown in the table below.

[0041] Secondly, the data samples under each state are subjected to validity checks. Visual inspection or data analysis is used to determine if there are any anomalies such as false trends or outliers. If so, the data is corrected to ensure its validity. The validity checks for data samples under each state are as follows: Figure 3 As shown.

[0042] Finally, wavelet analysis was performed on the corrected data from the ground engine to identify time-frequency ridges in the time-frequency plot. See [link to details on time-frequency ridges] for more information. Figure 4 , Figure 5 and Figure 6 As shown by the red curve, the frequency of the periodic component is determined to be 564Hz in this embodiment.

[0043] S2, Vibration normalization data set based on ground engine X={x(t n From the n sets of vibration measurement data in the data, we obtain the first type of component and the second type of component of the vibration measurement data. The first type of component is a periodic component or a narrowband component, and the second type of component is a random component. We then perform spectral analysis on the first type of component to obtain the periodic spectrum or the narrowband spectrum, and perform spectral analysis on the second type of component to obtain the random spectrum.

[0044] In this embodiment, all periodic components, narrowband components, and random components are identified. Based on the periodic component identification results in the table above, the periodic components are extracted from the PSD spectrum of the measured vibration signal to obtain the broadband random spectrum. Statistical summarization is performed on the broadband random spectrum sample set for all states, and the upper limit spectrum of the set is calculated. For example... Figure 7 As shown.

[0045] S3. Calculate the same period frequency in all periodic components or narrowband components for a mission profile of a ground engine. The corresponding maximum spectral value P max ( ), with P max ( Using as a benchmark, the remaining L-1 group period frequencies are calculated using the following formulas. The equivalent duration t of the corresponding spectral value LIn this embodiment, for the narrowband random vibration of a certain accessory under multiple speed states, the normalized root mean square value of the narrowband spectrum at each center frequency is calculated; according to the usage mission status of the engine platform where the accessory is located, usage mission vibration samples containing each order of narrowband are statistically analyzed, and the durations of vibration samples with the same center frequency and bandwidth are equivalently normalized and accumulated as the duration of the narrowband random vibration of that order; the total duration of each order of narrowband random vibration is calculated.

[0046] S4, Change the periodic frequency The L groups of time are summed to obtain the spectral value P. max ( And the duration is A set of parameters. In this embodiment, the squared root mean square value of the narrowband peak is divided by the bandwidth of each narrowband to obtain the PSD value corresponding to the total test time of random vibration of each narrowband.

[0047] S5, with the duration T of the random spectrum R Based on the reference, the period frequency Duration Converted to T R Calculate the converted spectral values In this embodiment, a narrowband random vibration durability test time is set, and the Miner cumulative damage equivalent formula is used to convert the total narrowband random vibration test time and the corresponding PSD value to the set vibration durability test time to obtain the narrowband random vibration durability test value.

[0048] Subsequently, based on the aforementioned spectral peak truncation, the PSD value adjusted to the test time was calculated according to the Miner cumulative damage equivalent formula (see column I in Table 1). This value was then multiplied by a safety factor of 1.4 to obtain the randomized test value (see column J in Table 1). Detailed results are shown in Table 1 below. Figure 8 .

[0049] Table 1 For cases where frequencies overlap but spectral values ​​differ in different typical task profiles, it is necessary to further equivalence the spectral values ​​of the overlapping parts. The results after equivalence are shown in Table 2 below.

[0050] Table 2 S6. Superimpose the time-unified periodic spectrum or narrowband spectrum with the random spectrum to form the vibration test conditions. Vibration tests were conducted. In this embodiment, the accelerated broadband spectrum and narrowband spectrum were combined. Based on the design of the broadband and narrowband random vibration spectra, the broadband and narrowband spectra were combined. For some narrowband spectra with smaller values, if the broadband spectrum value could cover the narrowband spectrum value, the broadband spectrum value was used. Secondly, considering that the center frequencies of multiple narrowband spectra were close together, in order to increase the operability of the laboratory vibration test, this embodiment was engineered to merge narrowband spectra with similar center frequencies, increase the value of some narrowband spectra, and combine multiple narrowbands into a narrowband with a larger bandwidth. The vibration spectrum frequency inflection points and values ​​are shown in Table 3 below, and the test spectrum type is as follows. Figure 9 As shown.

[0051] Table 3 As can be seen from the above embodiments, the method of the present invention first unifies the amplitude of the multi-task, multi-frequency periodic / narrowband components, then unifies the time of the multi-frequency periodic / narrowband components and random components, and finally superimposes them to obtain the vibration environment test conditions. This method, using the fatigue characteristic parameters of equipment materials as a starting point, solves for the first time the problem of calculating vibration environment test conditions for multi-task, multi-frequency periodic / narrowband components. The calculation results are highly reliable, consistent with engineering practice, and have significant application value.

[0052] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for determining environmental test conditions for multi-task, multi-frequency component periodic vibration, characterized in that: It includes the following steps: S1, normalize and pretreat the vibration measured data corresponding to n kinds of task profiles to obtain a vibration normalized data set X containing n kinds of task profiles, that is, X={x n (t)}, wherein t is the time domain length; S2, Based on the vibration normalization dataset X={x n From the n sets of measured vibration data in (t)}, we obtain the first and second types of vibration data, where the first type is a periodic or narrowband component, and the second type is a random component. We then perform spectral analysis on the first type to obtain the periodic or narrowband spectrum, and perform spectral analysis on the second type to obtain the random spectrum. Specifically, this includes: S21. Perform time-frequency analysis on the vibration normalized data x(t1) of the first task profile to generate the time spectrum of x(t1). The analysis formula is as follows: ； In the formula, a is a non-zero scaling coefficient, a = 1 / f, b is the translation amount, and CWT is the spectral value of the time spectrum. Let f be a wavelet function, f be the frequency, and d be the derivative; S22. Observe the spectrum at x(t1), and analyze the frequencies corresponding to the peaks of the frequency domain curve hour by hour. The analysis formula is as follows: ； In the formula, argmax is the value of the independent variable or parameter that makes a function reach its maximum value; S23, Connect the peaks at all times. This forms the time-frequency ridge of the vibration signal; S24. Repeat steps S22 and S23 to obtain n time-frequency ridges of the spectrum at x(t1); S25. Analyze the frequency range corresponding to each time-frequency ridge and identify and separate the periodic component / narrowband component and random component. S26. For the vibration normalization data set X={x(t) n For the remaining n-1 sets of vibration measurement data in the task profile, repeat steps S21~S25 to separate the periodic / narrowband and random components of the vibration measurement data for the n task profiles. S27. Perform spectral analysis on the periodic / narrowband and random components of the n task profiles respectively to obtain the periodic spectrum / narrowband spectrum and random spectrum. S3. Calculate the same period frequency in all periodic components or narrowband components for a given task profile. Corresponding maximum spectral value ,by Based on this, the remaining L-1 group period frequencies are calculated using the following formulas. The equivalent duration t of the corresponding spectral value L The calculation formula is as follows: ； Where m is the vibration durability equivalence factor, calculated based on the SN curve of the structural material, and P L Let L be the spectral value of the Lth component. Maximum spectral value The duration; S4, Change the periodic frequency The spectral values ​​are obtained by summing the L groups of time. And the duration is A set of parameters: ； S5, with the duration T of the random spectrum R Based on the reference, the period frequency Duration Converted to T R Calculate the converted spectral values The calculation formula is as follows: ； in, The spectral value of the random spectrum; S6. Repeat steps S3 to S5, converting the periodic or narrowband spectra of the n task profiles to have the same duration as the random spectrum and obtaining the converted spectral values. Superimpose the periodic or narrowband spectra with the random spectrum after unifying the duration to form the vibration test conditions. It was later applied to vibration platforms for vibration testing.

2. The method for determining test conditions for multi-task, multi-frequency component periodic vibration environments according to claim 1, characterized in that: In step S1, data standardization refers to unifying the name, unit, and storage format of the vibration measurement data.

3. The method for determining test conditions for multi-task, multi-frequency component periodic vibration environments according to claim 1, characterized in that: In step S1, data preprocessing refers to checking and correcting the normalized vibration measurement data to remove abnormal signals and spurious trend terms.

4. The method for determining test conditions for multi-task, multi-frequency component periodic vibration environments according to claim 2, characterized in that: The time-frequency analysis method in step S21 refers to simultaneously decomposing the measured vibration data in the time and frequency domains for analysis. Specifically, it includes Gabor expansion, Cohen-type time-frequency distribution, Radon-Wigner transform, wavelet analysis, or fractional Fourier transform.

5. The method for determining test conditions for multi-task, multi-frequency component periodic vibration environments according to claim 2, characterized in that: The spectral analysis in step S27 specifically involves: using line spectrum analysis for periodic signals; and using power spectrum analysis for narrowband or random signals.

6. The method for determining test conditions for multi-task, multi-frequency component periodic vibration environments according to claim 1, characterized in that: The vibration durability equivalence factor refers to the slope of the SN curve of the structural material in a double logarithmic coordinate system.

7. The method for determining test conditions for multi-task, multi-frequency component periodic vibration environments according to claim 1, characterized in that: The duration T of the random spectrum R This is the result of proportionally adjusting and summing the random vibration times of all task profiles.

8. The method for determining test conditions for multi-task, multi-frequency component periodic vibration environments according to claim 1, characterized in that: The specific steps of step S25 are as follows: S251. If the frequency corresponding to the time-frequency ridge is constant, then this component is a periodic component with a frequency fs. ki It is the periodic frequency; S252. If the frequency corresponding to the time-frequency ridge fluctuates within a small range, then extract the upper and lower limits of the floating frequency [fs]. ki fs k(i+1) As a narrowband component; S253, the rest of the signal is treated as a broadband random component.

9. The method for determining environmental test conditions for multi-task, multi-frequency component periodic vibration according to claim 8, characterized in that: The small-range fluctuation in step S252 refers to the fluctuation range being within ±5% of the period frequency.

Images