Micro-vibration evaluation method, system and equipment considering time weighting and storage medium

Abstract

The invention discloses a micro-vibration evaluation method, system and device considering time weighting, and a storage medium. The method comprises the following steps: preprocessing original micro-vibration data; setting the length of the analysis blocks and the overlapping ratio between the adjacent analysis blocks; carrying out block processing on the whole data, and if data less than one analysis block exists at last, abandoning the processing; calculating the duration of each analysis block, and taking the central moment of the analysis block as the representative moment of the analysis block; performing fast Fourier transform on all the analysis blocks, adding window function operation and multiplying by an energy recovery coefficient; effective values of all the analysis blocks at all the center frequencies are calculated; and drawing a time-center frequency-micro-vibration grade color cloud picture by taking time as an X coordinate, the center frequency as a Y coordinate and the velocity spectrum effective value as a Z coordinate. According to the method, the evaluation dimensions are more, the expression information is more complete, and the change rule of the speed effective value of each center frequency at different center frequencies and along with the time change of the micro-vibration can be displayed at the same time.

Description

Technical Field

[0001] The present invention belongs to the field of micro-vibration control and relates to a micro-vibration evaluation method, system, device and storage medium considering time weighting. Background Art

[0002] Precision process equipment has very high requirements for micro-vibration. If the environmental vibration exceeds the micro-vibration limit of the precision equipment, it will affect the normal operation of the precision equipment. Currently, 1 / 3 octave velocity spectrum is used for evaluation, such as Figure 1 However, the velocity spectrum only provides information on the change of the effective value at each center frequency and cannot express the change pattern of the micro-vibration level over time. For environmental micro-vibrations caused by vibration sources such as random vibration sources and mobile vibration sources, which vary greatly over time, in addition to clarifying the micro-vibration levels in different frequency bands, it is often necessary to clarify the change pattern over time when conducting micro-vibration evaluation. Summary of the Invention

[0003] Purpose of the invention: The purpose of the present invention is to provide a micro-vibration evaluation method, system, device and storage medium that takes time weighting into account, which has more evaluation dimensions and more complete information expression, and can simultaneously display the change law of the effective value of the micro-vibration velocity at different center frequencies and at each center frequency as time changes.

[0004] Technical solution: The micro-vibration evaluation method of the present invention comprises the following steps:

[0005] Preprocess the original micro-vibration data to eliminate the abnormal drift of the signal and obtain the overall data set after signal preprocessing;

[0006] Set the length of the analysis block and the overlap ratio between analysis blocks. Based on the length of the analysis block and the overlap ratio between analysis blocks, the entire data is divided into blocks. If there is less than the length of an analysis block, it is discarded, and finally the conversion from the entire data set to analysis blocks is realized.

[0007] The duration of each analysis block is calculated based on the analysis block length and data sampling rate, and the center moment of the analysis block is taken as the representative moment of the analysis block to form a time series of the analysis block;

[0008] Applying window functions to all analysis blocks in sequence and performing energy recovery on all analysis blocks; then performing fast Fourier transform on all analysis blocks to obtain the complex spectra of all analysis blocks;

[0009] Determine whether the original signal is a vibration acceleration signal. If not, directly calculate the effective value of each center frequency of all analysis blocks based on the complex spectrum. If so, first integrate the complex spectrum of the vibration acceleration signal, and then calculate the effective value of each center frequency of all analysis blocks.

[0010] The time series of the analysis block is the X coordinate, the center frequency is the Y coordinate, and the effective value of the velocity spectrum is the Z coordinate. The color time-center frequency-micro-vibration level cloud map is visualized and output.

[0011] Furthermore, data preprocessing is performed based on the least squares method.

[0012] Furthermore, the entire data is processed in blocks, including:

[0013] Let the one-dimensional array B of the entire data set be {y1, y2, ...y N0}; y1,y2,…y N0 is the vibration response value at a discrete time point. Let the acquisition time of the entire data set array be T and the sampling rate be fs0. Then the length of the entire data set N0 = fs0*T;

[0014] Let the analysis block array B k ={z1,z2,…z N1}, k is the subscript of the analysis block, representing the kth analysis block, k = 1, 2, ..., k max , where k max is the number of analysis blocks; let the length of the analysis block be N1, then the duration of the analysis block t1 = N1 / fs0, let the data overlap rate of the analysis block be R ol , where 0<R ol <1, solve the following equation to get the number of analysis blocks k max :

[0015] N1*(1-R ol )(k max -1)+N1=fs0*T

[0016] Considering the overlap rate of analysis blocks, the overall data set is divided into k max analysis blocks, the kth analysis block B k Expressed as:

[0017]

[0018] Among them, N1*(1-R ol )(k-1)+1 is the subscript corresponding to the element in the analysis block in the overall data set array, For the kth analysis block B k The first element in For the kth analysis block B k The second element in For the kth analysis block B k The N1th element in

[0019] Furthermore, the time series of the analysis block is formed by taking the center time of the analysis block as the representative time of the analysis block, including:

[0020] Time series array of analysis blocks Among them, t k For the kth analysis block B k represents the time, k=1,2,…,k max , k max is the number of analysis blocks;

[0021] kth analysis block B k Represents time t k =N1 / fs0+(N1*(1-R ol )*(k-1)) / fs0; where N1 is the analysis block length, fs0 is the sampling rate, and R ol is the data overlap ratio of the analysis block.

[0022] Furthermore, the complex spectrum expressions of all analysis blocks are:

[0023] P Bk (f1, f2,..., fn)=FFT(fac E *(B k *win()))

[0024] Among them, P Bk (f1, f2, ..., fn) is the kth analysis block B k The complex spectrum of the nth spectrum, fn is the frequency of the corresponding nth spectrum line, FFT is the fast Fourier transform function, fac E is the energy recovery coefficient, B k is the kth analysis block, and win is the window function.

[0025] Furthermore, if the original signal is vibration acceleration, the signal is first integrated, including:

[0026] Using the integral function to integrate acceleration to velocity, the complex spectrum after integration of each analysis block is:

[0027] kth analysis block B k The complex spectrum after integration is: P Bk_v =integral(P Bk ), where P Bk For the kth analysis block B k The complex spectrum of P Bk_v For the kth analysis block B k The integrated complex spectrum.

[0028] Furthermore, the effective value at each center frequency of all analysis blocks is calculated as follows:

[0029] Let the kth analysis block B k The nth center frequency is fc _k_n , 1≤k≤k max , 1≤n≤n max , k max is the number of analysis blocks, n max is the number of center frequencies, and the lower limit frequency is fl _k_n , complex spectrum P Bk The amplitude corresponding to the lower limit frequency spectrum line is mag(fl _k_n ), where mag is the amplitude function, and the complex spectrum P Bk The amplitude of the upper frequency spectrum line is mag(fu _k_n ), frequency resolution △f=1 / t1, t1 is the duration of the analysis block, let the effective value at the center frequency of the kth analysis block be Vrms _k = {Vrms _k_1 , Vrms _k_2 ,……,Vrms _k_n , ... Vrms _k_20}, the effective value calculation results are as follows:

[0030] The effective value of the velocity at the first center frequency of the first analysis block B1 is:

[0031]

[0032] The effective value of the velocity at the nth center frequency of the first analysis block B1 is:

[0033]

[0034] kth analysis block B k The effective value of the velocity at the nth center frequency is:

[0035]

[0036] Let the result matrix Vrms be composed of the effective velocity values ​​of each center frequency of all analysis blocks:

[0037]

[0038] The micro-vibration evaluation system of the present invention comprises:

[0039] A data preprocessing unit is used to preprocess the original micro-vibration data to eliminate the abnormal drift of the signal and obtain the overall data set after signal preprocessing;

[0040] The block processing unit is used to set the length of the analysis block and the overlap ratio between the analysis blocks. Based on the length of the analysis block and the overlap ratio between the analysis blocks, the entire data is divided into blocks for processing. If there is less than the length of an analysis block at the end, it is discarded, and finally the conversion from the entire data set to the analysis blocks is realized;

[0041] A time series establishment unit is used to calculate the duration of each analysis block according to the analysis block length and the data sampling rate, and to form a time series of the analysis block by taking the center moment of the analysis block as the representative moment of the analysis block;

[0042] The analysis block data processing unit is used to sequentially perform windowing functions on all analysis blocks and perform energy recovery on all analysis blocks; then perform fast Fourier transform on all analysis blocks to obtain complex spectra of all analysis blocks;

[0043] The effective value calculation unit is used to determine whether the original signal is a vibration acceleration signal. If not, the effective value at each center frequency of all analysis blocks is directly calculated based on the complex spectrum. If so, the complex spectrum of the vibration acceleration signal is first integrated, and then the effective value at each center frequency of all analysis blocks is calculated.

[0044] The evaluation chart generation unit is used to analyze the time series of the block as the X coordinate, the center frequency as the Y coordinate, and the effective value of the velocity spectrum as the Z coordinate, and to visualize the output of the color time-center frequency-micro-vibration level cloud chart.

[0045] The electronic device described in the present invention includes a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor. When the computer program / instruction is executed by the processor, the steps of the micro-vibration evaluation method considering time weighting are implemented.

[0046] The computer-readable storage medium of the present invention stores computer instructions, which, when called, are used to execute the steps of the micro-vibration evaluation method considering time weighting.

[0047] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are: (1) considering the signal overlap rate, the vibration signal data is divided into blocks, and the 1 / 3 octave velocity effective values ​​of the data blocks are batch calculated through the cyclic block program, and the change law of the vibration velocity effective values ​​on the time axis and frequency axis are simultaneously displayed in the form of a VC level time-frequency cloud diagram; (2) there are more evaluation dimensions and the expression of information is more comprehensive; the change law of the micro-vibration level in the time domain and frequency domain can be displayed at the same time, that is, the change law of the micro-vibration at different center frequencies and the change law of the velocity effective values ​​of each center frequency over time can be displayed at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1This is a micro-vibration assessment method that does not consider time weighting at this stage;

[0049] Figure 2 The calculation process of the micro-vibration evaluation method considering time weighting is presented;

[0050] Figure 3 This is a result diagram of the micro-vibration evaluation method case considering time weighting. DETAILED DESCRIPTION

[0051] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] like Figure 2 As shown, the micro-vibration evaluation method considering time weighting of the present invention includes the following steps:

[0053] S1. Preprocessing the original micro-vibration data obtained by testing or calculation to eliminate abnormal phenomena such as signal drift and obtain the overall data set after signal preprocessing;

[0054] In this embodiment, the original data obtained by testing or calculation is a vibration acceleration signal data. Assuming that the sampling rate of a vibration acceleration signal is 512 Hz and the sampling time is 120 s, the discrete sequence of the vibration acceleration signal is {x i}, i=1,2,3,…,61440,x i is the i-th discrete vibration acceleration signal.

[0055] Data preprocessing is performed based on the least squares method to eliminate trend terms.

[0056] Let the pre-processed vibration acceleration signal be {y i}. Among them, y i is x i After preprocessing, the signal a j is the coefficient, j=1,2,3,…,m, m is the order of the set polynomial; the coefficients a0~a can be solved by the equation group j .

[0057] S2. Setting the length of the analysis block and the overlap ratio between analysis blocks;

[0058] In this embodiment, the analysis block length is set to 1024, and the overlap rate of the data blocks is set to 75%.

[0059] S3. Based on the length of the analysis block and the overlap rate between analysis blocks, the entire data set is divided into blocks. If there is any data that is less than the length of an analysis block, it is discarded, and the conversion from the entire data set to analysis blocks is finally achieved;

[0060] Let the one-dimensional array B of the entire data set be {y1, y2, ...y N0}; y1,y2,…y N0 is the vibration response value at a discrete time point. Let the acquisition time of the entire data set array be T and the sampling rate be fs0. Then the length of the entire data set N0 = fs0*T;

[0061] Let the analysis block array B k ={z1,z2,…z N1}, k is the subscript of the analysis block, representing the kth analysis block, k = 1, 2, ..., k max , where k max is the number of analysis blocks; let the length of the analysis block be N1, then the duration of the analysis block t1 = N1 / fs0, let the data overlap rate of the analysis block be R ol , where 0<R ol <1, solve the following equation to get the number of analysis blocks k max :

[0062] N1*(1-R ol )(k max -1)+N1=fs0*T

[0063] Considering the overlap rate of analysis blocks, the overall data set can be divided into k max Analysis blocks, each analysis block results are:

[0064] The first analysis block

[0065] The second analysis block

[0066] …

[0067] kth analysis block

[0068] …

[0069] kth max Analysis Block

[0070]

[0071] Among them, N1*(1-R ol )(k-1)+1 is the subscript of the first element (subscript 1) in the k-th analysis block in the overall data set, then and are the first, second and N1th elements in the first analysis block B1, respectively. and are the first, second and N1th elements in the second analysis block B2, and The kth analysis block B k The first, second and N1th elements in and kth max Analysis Block The first, second and N1th elements in .

[0072] In this embodiment, according to the above-set analysis block parameters, let the one-dimensional array composed of analysis blocks be {B k}(k=1,2,3,…,k max );

[0073] The first analysis block B1={y1,y2,…y 1024};

[0074] The second analysis block B2 = {y 257 ,y 258 ,…y 1280};

[0075] The third analysis block B3={y 513 ,y 514 ,…y 1536};

[0076] Until the kth max Analysis Block

[0077]

[0078] Solve a linear equation, 1024*(1-75%)(k max -1)+1024=512*120, we can get k max =237.

[0079] From this we can see that there are 237 data blocks in total.

[0080] The duration of each analysis block is calculated based on the analysis block length and data sampling rate, and the center moment of the analysis block is taken as the representative moment of the analysis block to form a time series array of the analysis block. The method is:

[0081] The first analysis block B1 represents the time t1 = N1 / fs0 + (N1*(1-R ol )*(1-1)) / fs0

[0082] The second analysis block B2 represents the time t2 = N1 / fs0 + (N1*(1-R ol )*(2-1)) / fs0

[0083] …

[0084] kth analysis block B kRepresents time t k =N1 / fs0+(N1*(1-R ol )*(k-1)) / fs0

[0085] …

[0086] kth max Analysis Block Representative Moment

[0087] In this embodiment, the center time t1 of the B1 analysis block is taken as the representative time of the analysis block. k (k=1,2,3,…,237) Analysis block center time t k is the representative time of the analysis block, then t1=1024 / 512=2, t2=t1+(1024*(1-75%)) / 512=2.5, t3=t1+(1024*(1-75%)*2) / 512=3, t 237 = t1 + (1024*(1-75%)*236) / 512 = 120.0. Let the one-dimensional array consisting of all analysis blocks representing the time be {t k}(k=1,2,3,…,237).

[0088] S4. To reduce spectrum leakage, improve the accuracy of spectrum analysis, and better handle signal boundary effects, a window function is applied to all analysis blocks in sequence, and energy recovery is performed on all analysis blocks. Fast Fourier transform is performed on all analysis blocks to obtain the complex spectrum of all analysis blocks. Specifically:

[0089] Add window function win operation to all analysis blocks and multiply them by energy recovery coefficient fac E Perform energy recovery; use the fast Fourier transform function FFT to obtain the complex spectrum P of all analysis blocks Bk (f1, f2, ..., fn), fn is the frequency of the corresponding nth spectral line, and the complex spectrum calculation results of each analysis block are:

[0090] The complex spectrum of the first analysis block B1 is P B1 (f1, f2,..., fn)=FFT(fac E *(B1*win())) The complex spectrum of the second analysis block B2 is P B2 (f1, f2,..., fn)=FFT(fac E *(B2*win()))

[0091] …

[0092] kth analysis block B k The complex spectrum of PBk (f1, f2,..., fn)=FFT(fac E *(B k *win()))

[0093] …

[0094] kth max Analysis block B kmax The complex spectrum of

[0095]

[0096] For steady-state and random vibration data, the Hanning window is recommended.

[0097] In this embodiment, a Hanning window function win (hanning) operation is added to each analysis block and multiplied by the energy recovery coefficient fac E Perform energy recovery followed by fast Fourier transform.

[0098] S5. When the original data is vibration acceleration, the signal needs to be integrated in the frequency domain;

[0099] Using the integral function to integrate acceleration to velocity, the complex spectrum after integration of each analysis block is:

[0100] The complex spectrum after integration of the first analysis block B1 is: P B1_v =integral(P B1 )

[0101] The complex spectrum after integration of the second analysis block B2 is: P B2_v =integral(P B2 )

[0102] …

[0103] kth analysis block B k The complex spectrum after integration is: P Bk_v =integral(P Bk )

[0104] …

[0105] kth max Analysis Block The complex spectrum after integration is:

[0106] The expression of the integral function integral is: integral() = A / ω, where A is the complex spectrum before integration and ω is the frequency corresponding to the position of the integral spectrum line.

[0107] S6. Calculate the effective value at each center frequency of all analysis blocks;

[0108] According to the micro-vibration evaluation criteria, the frequency band of micro-vibration evaluation is 1~80Hz, so only 20 center frequencies of 1 / 3 octave 1~80Hz are taken for analysis. Let the sequence of center frequencies be fc, fc={fc1, fc2, ..., fc 20}={1.00, 1.25, 1.60, 2.00, 2.50, 3.15, 4.00, 5.00, 6.30, 8.00, 10.0, 12.5, 16.0, 25.0, 30.0, 31.5, 40.0, 50.0, 63.0, 80.0}.

[0109] Let the kth (1≤k≤k max )Analysis block B k The center frequency of the nth (1≤n≤20) is fc _k_n , the lower limit frequency is fl _k_n , complex spectrum P Bk The amplitude corresponding to the lower limit frequency spectrum line is mag(fl _k_n ), where mag is the amplitude function, and the complex spectrum P Bk The amplitude of the upper frequency spectrum line is mag(fu _k_n ), frequency resolution △f (△f=1 / t1), let the effective value at the center frequency of the kth block be Vrms _k = {Vrms _k_1 , Vrms _k_2 ,……,Vrms _k_n , ... Vrms _k_20}, the effective value calculation results are as follows:

[0110] The effective value of velocity at the first center frequency of the first analysis block B1 is

[0111]

[0112] The effective value of the velocity at the nth center frequency of the first analysis block B1 is

[0113]

[0114] The effective value of the velocity at the nth center frequency of the kth analysis block Bk is

[0115]

[0116] Let the result matrix Vrms be composed of the effective velocity values ​​of each center frequency of all analysis blocks:

[0117]

[0118] In this embodiment, the matrix composed of the effective velocity values ​​of each center frequency (20 in total) of all analysis blocks (237 in total) is:

[0119] S7. With the generated analysis block time series as the X coordinate, the center frequency as the Y coordinate, and the effective value of the velocity spectrum as the Z coordinate, a color time-center frequency-micro-vibration level cloud diagram is visualized and output.

[0120] In this embodiment, the measured data of a certain project are used to obtain the graphical results as follows: Figure 3 shown. Figure 3 It has the characteristics of more evaluation dimensions and more complete information expression. It can simultaneously display the changing rules of the effective value of the micro-vibration velocity at different center frequencies and at each center frequency over time.

[0121] The method described in the present invention can simultaneously display the changing patterns of micro-vibration levels in the time domain and frequency domain, solving the problem of incomplete information in traditional micro-vibration assessment methods. It can be applied to industries such as semiconductors and flat panel displays that have micro-vibration requirements.

[0122] The micro-vibration evaluation system considering time weighting of the present invention comprises:

[0123] A data preprocessing unit is used to preprocess the original micro-vibration data to eliminate the abnormal drift of the signal and obtain the overall data set after signal preprocessing;

[0124] The block processing unit is used to set the length of the analysis block and the overlap ratio between the analysis blocks. Based on the length of the analysis block and the overlap ratio between the analysis blocks, the entire data is divided into blocks for processing. If there is less than the length of an analysis block at the end, it is discarded, and finally the conversion from the entire data set to the analysis blocks is realized;

[0125] A time series establishment unit is used to calculate the duration of each analysis block according to the analysis block length and the data sampling rate, and to form a time series of the analysis block by taking the center moment of the analysis block as the representative moment of the analysis block;

[0126] The analysis block data processing unit is used to sequentially perform windowing functions on all analysis blocks and perform energy recovery on all analysis blocks; then perform fast Fourier transform on all analysis blocks to obtain complex spectra of all analysis blocks;

[0127] The effective value calculation unit is used to determine whether the original signal is a vibration acceleration signal. If not, the effective value at each center frequency of all analysis blocks is directly calculated based on the complex spectrum. If so, the complex spectrum of the vibration acceleration signal is first integrated, and then the effective value at each center frequency of all analysis blocks is calculated.

[0128] The evaluation chart generation unit is used to analyze the time series of the block as the X coordinate, the center frequency as the Y coordinate, and the effective value of the velocity spectrum as the Z coordinate, and to visualize the output of the color time-center frequency-micro-vibration level cloud chart.

[0129] The electronic device of the present invention includes a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor. When the computer program / instruction is executed by the processor, the steps of the micro-vibration evaluation method considering time weighting are implemented.

[0130] The computer-readable storage medium of the present invention stores computer instructions, which, when called, are used to execute the steps of the micro-vibration evaluation method considering time weighting.

Claims

1. A micro-vibration evaluation method considering time weighting, characterized in that: The following steps are involved: Preprocess the original micro-vibration data to eliminate the abnormal drift of the signal and obtain the overall data set after signal preprocessing; Set the length of the analysis block and the overlap ratio between analysis blocks. Based on the length of the analysis block and the overlap ratio between analysis blocks, the entire data is divided into blocks. If there is less than the length of an analysis block, it is discarded, and finally the conversion from the entire data set to analysis blocks is realized. The duration of each analysis block is calculated based on the analysis block length and data sampling rate, and the center moment of the analysis block is taken as the representative moment of the analysis block to form a time series of the analysis block; Applying window functions to all analysis blocks in sequence and performing energy recovery on all analysis blocks; then performing fast Fourier transform on all analysis blocks to obtain the complex spectra of all analysis blocks; Determine whether the original signal is a vibration acceleration signal. If not, directly calculate the effective value of each center frequency of all analysis blocks based on the complex spectrum. If so, first integrate the complex spectrum of the vibration acceleration signal, and then calculate the effective value of each center frequency of all analysis blocks. The time series of the analysis block is the X coordinate, the center frequency is the Y coordinate, and the effective value of the velocity spectrum is the Z coordinate. The color time-center frequency-micro-vibration level cloud map is visualized and output.

2. The micro-vibration evaluation method considering time weighting according to claim 1, characterized in that: Data preprocessing was performed based on the least squares method.

3. The micro-vibration evaluation method considering time weighting according to claim 1, characterized in that: The entire data is processed in blocks, including: Let the one-dimensional array B of the entire data set be {y1, y2, ...y N0 };y1,y2,…y N0 is the vibration response value at a discrete time point. Let the acquisition time of the entire data set array be T and the sampling rate be fs0. Then the length of the entire data set N0 = fs0*T; Let the analysis block array B k ={z1,z2,…z N1 }, k is the subscript of the analysis block, representing the kth analysis block, k = 1, 2, ..., k max , where k max is the number of analysis blocks; let the length of the analysis block be N1, then the duration of the analysis block t1 = N1 / fs0, let the data overlap rate of the analysis block be R ol , where 0<R ol <1, solve the following equation to get the number of analysis blocks k max : N1*(1-R ol )(k max -1)+N1=fs0*T Considering the overlap rate of analysis blocks, the overall data set is divided into k max analysis blocks, the kth analysis block B k Expressed as: Among them, N1*(1-R ol )(k-1)+1 is the subscript corresponding to the element in the analysis block in the overall data set array, For the kth analysis block B k The first element in For the kth analysis block B k The second element in For the kth analysis block B k The N1th element in .

4. The micro-vibration evaluation method considering time weighting according to claim 1, characterized in that: Taking the central moment of the analysis block as the representative moment of the analysis block, the time series of the analysis block is formed, including: Time series array of analysis blocks Among them, t k For the kth analysis block B k represents the time, k=1,2,…,k max , k max is the number of analysis blocks; kth analysis block B k Represents time t k =N1 / fs0+(N1*(1-R ol )*(k-1)) / fs0; where N1 is the analysis block length, fs0 is the sampling rate, and R ol is the data overlap ratio of the analysis block.

5. The micro-vibration evaluation method considering time weighting according to claim 1, characterized in that: The complex spectrum expressions of all analysis blocks are: P ,k (f1,f2,……,fn)=FFT(fac7*(B k *win())) Among them, P ,k (f1, f2, ..., fn) is the kth analysis block B k The complex spectrum of the nth spectrum, fn is the frequency of the corresponding spectrum line, FFT is the fast Fourier transform function, fac7 is the energy recovery coefficient, B k is the kth analysis block, and win is the window function.

6. The micro-vibration evaluation method considering time weighting according to claim 1, characterized in that: If the original signal is vibration acceleration, first perform signal integration, including: Using the integral function to integrate acceleration to velocity, the complex spectrum after integration of each analysis block is: kth analysis block B k The complex spectrum after integration is: P ,k_＝ =integral(P ,k ), where P ,k For the kth analysis block B k The complex spectrum of P ,k_＝ For the kth analysis block B k The integrated complex spectrum.

7. The micro-vibration evaluation method considering time weighting according to claim 1, characterized in that: Calculate the effective value at each center frequency of all analysis blocks as follows: Let the kth analysis block B k The nth center frequency is fc _k_> , 1≤k≤k max , 1≤n≤n max , k max is the number of analysis blocks, n max is the number of center frequencies, and the lower limit frequency is Fl _k_> , complex spectrum P ,k The amplitude corresponding to the lower limit frequency spectrum line is mag(fl _k_> ), where mag is the amplitude function, and the complex spectrum P ,k The amplitude of the upper frequency spectrum line is mag(fu _k_> ), frequency resolution △f=1 / t1, t1 is the duration of the analysis block, let the effective value at the center frequency of the kth analysis block be Vrms _k = {Vrms _k_1 , Vrms _k_2 ,……,Vrms _k_> , ... Vrms _k_20 }, the effective value calculation results are as follows: The effective value of the velocity at the first center frequency of the first analysis block B1 is: The effective value of the velocity at the nth center frequency of the first analysis block B1 is: kth analysis block B k The effective value of the velocity at the nth center frequency is: Let the result matrix Vrms be composed of the effective velocity values ​​of each center frequency of all analysis blocks:

8. A micro-vibration evaluation system considering time weighting, characterized in that: include: A data preprocessing unit is used to preprocess the original micro-vibration data to eliminate the abnormal drift of the signal and obtain the overall data set after signal preprocessing; The block processing unit is used to set the length of the analysis block and the overlap ratio between the analysis blocks. Based on the length of the analysis block and the overlap ratio between the analysis blocks, the entire data is divided into blocks for processing. If there is less than the length of an analysis block at the end, it is discarded, and finally the conversion from the entire data set to the analysis blocks is realized; A time series establishment unit is used to calculate the duration of each analysis block according to the analysis block length and the data sampling rate, and to form a time series of the analysis block by taking the center moment of the analysis block as the representative moment of the analysis block; The analysis block data processing unit is used to sequentially perform windowing functions on all analysis blocks and perform energy recovery on all analysis blocks; then perform fast Fourier transform on all analysis blocks to obtain complex spectra of all analysis blocks; The effective value calculation unit is used to determine whether the original signal is a vibration acceleration signal. If not, the effective value at each center frequency of all analysis blocks is directly calculated based on the complex spectrum. If so, the complex spectrum of the vibration acceleration signal is first integrated, and then the effective value at each center frequency of all analysis blocks is calculated. The evaluation chart generation unit is used to analyze the time series of the block as the X coordinate, the center frequency as the Y coordinate, and the effective value of the velocity spectrum as the Z coordinate, and to visualize the output of the color time-center frequency-micro-vibration level cloud chart.

9. An electronic device, characterized in that: The invention comprises a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor, wherein when the computer program / instruction is executed by the processor, the steps of the micro-vibration evaluation method considering time weighting according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the steps of the micro-vibration evaluation method considering time weighting according to any one of claims 1 to 7.

Images