CN108334682A - A kind of batch processing method that frequency spectrum refines automatically - Google Patents
A kind of batch processing method that frequency spectrum refines automatically Download PDFInfo
- Publication number
- CN108334682A CN108334682A CN201810067689.8A CN201810067689A CN108334682A CN 108334682 A CN108334682 A CN 108334682A CN 201810067689 A CN201810067689 A CN 201810067689A CN 108334682 A CN108334682 A CN 108334682A
- Authority
- CN
- China
- Prior art keywords
- frequency
- frequency spectrum
- data
- refinement
- refines
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
- G06F2218/14—Classification; Matching by matching peak patterns
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Spectrometry And Color Measurement (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention relates to technical field of data processing more particularly to a kind of batch processing methods that frequency spectrum refines automatically, include the following steps：The data of acquisition；To data regularization；Obtain effective peak frequency；Judge whether to need micronization processes by effective frequency difference value, to needing the linear transform of carry out of micronization processes；It repeats the above steps, completes the micronization processes of all operating modes.Using the above method, multi-state in spectrum analysis can be solved, the problems such as data volume is big, data dimension is irregular, characteristic frequency domain spectral line is intensive, realize the Regularization batch processing and the automatic micronization processes of frequency spectrum of data, effectively obtain the frequecy characteristic of data.
Description
Technical field
The present invention relates to technical field of data processing more particularly to a kind of batch processing methods that frequency spectrum refines automatically.
Background technology
Spectrum analysis is most common method in signal processing, and traditional frequency spectrum analysis method generally uses fast Fourier
It converts (FFT), obtains the panoramic spectrum of entire band limits, and in practical applications, it generally requires to interested narrowband frequency
It composes section and carries out subtle observation and analysis, this just needs to improve frequency resolution using certain Frequence zooming method.
However in large-scale experiment or numerical simulation analysis, data generally existing following characteristics：Operating mode is more, data volume is big,
Data dimension is irregular, characteristic frequency domain spectral line is intensive etc..Frequency spectrum calculating and analysis are carried out according to manual type, sea can be reduced
The processing mistake or spectrum peak measured the extraction efficiency of data spectrum feature, and brought there are personnel's fatigue effect are omitted etc.
Potential uncertain factor.Once above-mentioned phenomenon occurs, it may be difficult to obtain the accurate description of system performance, production can be seriously affected
The design optimization of product.
Invention content
In view of above-mentioned analysis, the present invention is intended to provide a kind of batch processing method that frequency spectrum refines automatically, existing to solve
There is the problem that operating mode is more, data volume is big, data dimension is irregular, characteristic frequency domain spectral line is intensive.
The purpose of the present invention is mainly achieved through the following technical solutions：
A kind of batch processing method that frequency spectrum refines automatically, which is characterized in that include the following steps：
According to different experiment boundary conditions, classify to the gathered data of different operating modes；
To the gathered data of each operating mode, data regularization is carried out, multiple Regularization data are constituted；
According to the row of each Regularization data to dimension serial number, the frequency spectrum per column data is calculated, obtain the column data has
Imitate crest frequency；
The difference on the frequency score value for calculating adjacent effective peak in the column data frequency spectrum judges whether to need to carry out at frequency spectrum refinement
Reason is that then extraction needs to carry out the initial frequency of micronization processes frequency range and terminates frequency, carry out chirp z transform and preserve
As a result, obtaining the zoom FFT of corresponding frequency band；It is no, directly preserve spectrum results.
The present invention has the beneficial effect that：This method classifies to data according to experiment boundary condition, and carries out batch
Data padding regularization solves multi-state in mass data processing, the problems such as data dimension is irregular, and significantly
Improve data-handling efficiency；
It introduces spectral magnitude maximum value and screens effective peak as criterion, can utmostly ensure effective frequency feature
Extraction；
For difference on the frequency score value using effective peak as whether the criterion of frequency spectrum refinement processing is carried out, it is unnecessary to avoid
Frequence zooming process, reduce data processing work amount.
On the basis of said program, the present invention has also done following improvement：
Further, the gathered data of each operating mode imported to batch, carrying out data regularization includes：
The gathered data quantity of the operating mode is M, that is, includes M length data vector different in size, find therein adopt
Number of samples maximum value, as length maximum value, are denoted as N；
Zero padding processing is carried out to remaining M-1 data vector back so that the length of each data vector is N；
By after zero padding vector according to first row, secondary series ... m column arrange, constitute Regularization data.
Advantageous effect using above-mentioned further scheme is：By data vector zero padding different in size, become Regularization
Data preferably can do subsequent processing to each of which data.Maximum value data vector is chosen as length standard, was both protected
The validity of all data, and wasted caused by preventing data length to be set as the definite value beyond each data length.
Further, the frequency spectrum calculated per column data, and obtain effective peak frequency and include：
Fast Fourier Transform (FFT) is carried out to the column data；
It takes absolute value to transformation results, obtains spectral magnitude；
Advantageous effect using above-mentioned further scheme is：Using Fast Fourier Transform (FFT), data effective spectrum is calculated
Out, and by effective peak frequency it preserves, is made reference for the follow-up difference on the frequency score value that calculates.
Further, the effective peak frequency that obtains includes：
Using the maximum value of the spectral magnitude as the criterion of the row effective peak, when amplitude is higher than a times of maximum value,
It is determined as that effective peak, the corresponding frequency values of extraction effective peak save as effective peak frequency sequence.
Advantageous effect using above-mentioned further scheme is：The condition of effective peak is set, can save effective frequency after
Continuous refinement.
Further, described to judge whether to need to carry out frequency spectrum refinement processing, including sequence compares the difference on the frequency of effective peak
The size of sub-sequence and primary frequency resolution ratio b multiple values, if difference on the frequency score value is all higher than primary frequency resolution ratio b multiple values,
Without carrying out frequency spectrum refinement, spectrum results are directly preserved；
If difference on the frequency score value is respectively less than primary frequency resolution ratio b multiple values, need to carry out frequency spectrum refinement.
Advantageous effect using above-mentioned further scheme is：Frequency differential sequence and the primary frequency of effective peak compare,
For judging there is whether frequency spectrum needs to refine, the process of the frequency spectrum refinement of unnecessary refinement is avoided.
Further, the b value ranges are：3~5.
Advantageous effect using above-mentioned further scheme is：The value range that b is arranged is 3-5 as refinement standard so that
Not only meet refinement to require, but also wasted caused by not will produce excessively refinement.
Further, determine that the initial frequency of frequency spectrum refinement frequency range includes：
It when difference on the frequency score value is less than primary frequency resolution ratio b multiple values for the first time, extracts in the difference value, minuend corresponds to
Initial frequency of the effective peak Frequency point as first segment frequency spectrum refinement frequency range.
Advantageous effect using above-mentioned further scheme is：It is done using difference on the frequency score value and primary frequency resolution ratio b multiple values
Comparison, judging frequency spectrum, whether there is still a need for refinements.To limit frequency spectrum refinement frequency initial range, reduce unnecessary refinement
The process of frequency spectrum refinement.
Further, determine that the termination frequency of frequency spectrum refinement frequency range includes：
If until the last one serial number of difference sequence, difference on the frequency score value are respectively less than primary frequency resolution ratio b multiple values, then carry
Take the last one effective peak Frequency point as the termination frequency of this section of frequency spectrum refinement frequency range；Illustrate that the column data only has one section
Frequency range needs to carry out frequency spectrum refinement；
If in comparison process, when frequency of occurrences difference value is more than primary frequency resolution ratio b multiple values, extracting the difference value
In, termination frequency of the corresponding effective peak Frequency point of minuend as first segment frequency spectrum refinement frequency range.
Advantageous effect using above-mentioned further scheme is：When difference on the frequency score value is all higher than b times of primary frequency, you can stop
It only refines, limits the range of refinement, to avoid unnecessary refinement.Using the method for automatic decision, artificial judgment is avoided
Easy omission the shortcomings that.
Further, chirp z transform (CZT) is carried out according to the following formula, and preserves result；
Wherein：
X (n), fs are the column data and sample frequency of the operating mode after Regularization respectively；
fmin1,fmax1, initial frequency, the termination frequency of respectively this section frequency spectrum refinement frequency range；
R=(fmax1-fmin1)/Δ f, the points of this section of frequency spectrum refinement frequency range CZT；
R=0,1 ... R-1, the serial number of this section of frequency spectrum refinement frequency range CZT；
Δ f is the frequency resolution after frequency spectrum refinement.
Advantageous effect using above-mentioned further scheme is：To needing the frequencies of micronization processes in the way of linear transform
Rate carries out micronization processes so that resolution ratio reduces, and achievees the purpose that refinement.
Further, each column data in each operating mode is traversed, is handled to obtain the CZT of frequency spectrum refinement frequency range, and protect
Deposit result；Until institute refinement in need frequency range complete frequency spectrum refinement handle.
Advantageous effect using above-mentioned further scheme is：Using the method for automatic refinement batch processing, automatically process every
Each section of frequency range that need to be refined of one operating mode is completed until all frequency spectrum operating modes refine, and the speed for eliminating desk checking is slow, wrong
The accidentally high defect of rate.
It in the present invention, can also be combined with each other between above-mentioned each technical solution, to realize more preferred assembled schemes.This
Other feature and advantage of invention will illustrate in the following description, also, certain advantages can become from specification it is aobvious and
It is clear to, or understand through the implementation of the invention.The purpose of the present invention and other advantages can by write specification, right
Specifically noted structure is realized and is obtained in claim and attached drawing.
Description of the drawings
Attached drawing is only used for showing the purpose of specific embodiment, and is not considered as limitation of the present invention, in entire attached drawing
In, identical reference mark indicates identical component.
Fig. 1 is a kind of system block diagram for the batch processing method that frequency spectrum refines automatically of the present invention；
Fig. 2 is that the effective peak of the first column data of the specific embodiment of the invention screens schematic diagram；
Fig. 3 is that the effective peak of the first column data of the specific embodiment of the invention screens close-up schematic view；
Fig. 4 is that the refinement frequency range of the first column data of the specific embodiment of the invention judges schematic diagram；
Fig. 5 is that the frequency spectrum refinement of the first column data of the specific embodiment of the invention handles schematic diagram；
Fig. 6 is that the effective peak of the second column data of the specific embodiment of the invention screens schematic diagram；
Fig. 7 is that the refinement frequency range of the second column data of the specific embodiment of the invention judges schematic diagram.
Specific implementation mode
Specifically describing the preferred embodiment of the present invention below in conjunction with the accompanying drawings, wherein attached drawing constitutes the application part, and
It is used to illustrate the principle of the present invention together with embodiments of the present invention, be not intended to limit the scope of the present invention.
Embodiment one
The present invention a specific embodiment, disclose a kind of batch processing method that frequency spectrum refines automatically, specifically include with
Lower step：
Step 1, according to different experiment boundary conditions, classify to the gathered data of different operating modes, and batch import
The workspaces MATLAB；
Specifically, experiment boundary condition can be test period, load-up condition etc.；
Data are pre-processed, to judge to prepare for data refinement, to improve the accuracy judged later.
The gathered data of step 2, each operating mode imported to batch carries out data regularization；
Preferably, the step 2 includes the following steps：
Step 201 assumes that the gathered data quantity of an operating mode is M, that is, include M length data different in size to
Amount, finds sampling number maximum value therein i.e. length maximum value, is denoted as N；
Step 202 carries out zero padding processing to remaining M-1 data vector back so that the length of each data vector
It is n；
Step 203, by after zero padding vector according to first row, secondary series ... m column arrange, constitute Regularization data.
By data vector zero padding different in size, become the data of Regularization, after can preferably being done to each of which data
Continuous processing.Maximum value data vector is chosen as preferred length standard, not only protects the validity of all data, but also prevent data
Length is set as wasting caused by definite value.
Step 3, according to the row of Regularization data to dimension serial number, calculate the frequency spectrum of the first column data first, obtain effectively
Crest frequency；
Preferably, the frequency spectrum for calculating each column data, obtaining effective peak frequency includes：
Step 301 carries out Fast Fourier Transform (FFT) (FFT) to each column data；
Step 302, the result of FFT transform take absolute value, and obtain spectral magnitude；
Step 303 extracts criterion of the maximum value of spectral magnitude as effective peak, when amplitude is higher than a times of maximum value
When, it is determined as that effective peak, the corresponding frequency values of extraction effective peak save as effective peak frequency sequence.
Preferably, a value ranges are 0.1~0.2.
By finding out effective peak-to-peak value frequency, as the reference value with the frequency spectrum for needing to refine.Subsequently to calculate frequency
Difference value makes reference.Using Fast Fourier Transform (FFT), each column data effective spectrum is calculated, and frequently by effective peak
Rate preserves, and can be wasted caused by avoid all refining, improve efficiency.
Step 4, the difference on the frequency score value according to adjacent effective peak, judge whether the column data needs to carry out at frequency spectrum refinement
Reason.
Specifically, if it is judged that need to carry out frequency spectrum refinement processing, then extraction needs to carry out micronization processes frequency range
Initial frequency and terminate frequency, carry out chirp z transform simultaneously preserve as a result, obtaining the zoom FFT of corresponding frequency band；If
Judging result is then directly to preserve spectrum results woth no need to carry out frequency spectrum refinement processing.
Specifically, the step 4 includes the following steps：
Step 401, the frequency differential sequence for calculating effective peak frequency sequence；
The size of step 402, the frequency differential sequence and primary frequency resolution ratio b multiple values of sequence comparison effective peak, such as
Fruit difference on the frequency score value is all higher than primary frequency resolution ratio b multiple values, then is not necessarily to carry out frequency spectrum refinement, directly preserves spectrum results；It is no
Then enter step 403
Preferably, the value range of b is 3-5.
Step 403, the initial frequency for determining frequency spectrum refinement frequency range；
It when difference on the frequency score value is less than primary frequency resolution ratio b multiple values for the first time, extracts in the difference value, minuend corresponds to
Initial frequency of the effective peak Frequency point as first segment frequency spectrum refinement frequency range；
Step 404, the termination frequency for determining frequency spectrum refinement frequency range；
If until the last one serial number of difference sequence, difference on the frequency score value are respectively less than primary frequency resolution ratio b multiple values, then carry
Take the last one effective peak Frequency point as the termination frequency of this section of frequency spectrum refinement frequency range；Illustrate that the column data only has one section
Frequency range needs to carry out frequency spectrum refinement；
If in comparison process, when frequency of occurrences difference value is more than primary frequency resolution ratio b multiple values, extracting the difference value
In, termination frequency of the corresponding effective peak Frequency point of minuend as first segment frequency spectrum refinement frequency range；
Step 405 repeats step 403,404, extracts the initial frequency of next section of frequency spectrum refinement frequency range and terminates frequency, directly
To the last one serial number of difference sequence, the judgement of frequency spectrum refinement frequency range is completed；
Step 406, for each section of frequency spectrum refinement frequency range of each data in each operating mode after Regularization, according to
Following formula calculates its chirp z transform (CZT), and preserves result；
Wherein：
X (n), fs are the column data and sample frequency of the operating mode after Regularization respectively；
fmin1,fmax1, initial frequency, the termination frequency of respectively this section frequency spectrum refinement frequency range；
R=(fmax1-fmin1)/Δ f, the points of this section of frequency spectrum refinement frequency range CZT；
R=0,1 ... R-1, the serial number of this section of frequency spectrum refinement frequency range CZT；
Δ f is the frequency resolution after frequency spectrum refinement, electedly, is traditionally arranged to be 10~20 times of former resolution ratio；
Step 407 continues step 406, calculates each section of frequency spectrum of each column data in each operating mode after Regularization
The CZT of frequency range is refined, and preserves result；Until each column data institute refinement in need frequency range complete frequency spectrum refinement handle.
In step 4, it is compared with primary frequency using the frequency differential sequence of effective peak, for judging whether there is frequency spectrum
It needs refinement to limit the initial range of the frequency of frequency spectrum refinement and terminates range, avoid unnecessary thinning process.By second
The frequency spectrum that section need to refine does refinement identical with first segment spectral method and judges, with the method for automatic decision, avoids and manually sentences
The shortcomings that disconnected easy omission.To needing the frequency of micronization processes to carry out micronization processes in the way of linear transform so that point
Resolution reduces, and achievees the effect that refinement.Using the method for automatic refinement batch processing, automatically processing the first operating mode second segment needs carefully
The frequency range of change.
Step 5 traverses each column data in each operating mode, is handled to obtain the CZT of frequency spectrum refinement frequency range, and preserve
As a result；Until institute refinement in need frequency range complete frequency spectrum refinement handle.
Using same method, each row and each operating mode are refined, are completed until all operating modes refine, to
Realize the batch processing of frequency spectrum automation.
Embodiment two
The monitoring method using system described in embodiment one is described in detail in the present embodiment.
Step 1, according to different experiment boundary conditions, according to boundary conditions such as test period, load-up conditions to different works
The gathered data of condition is classified, and batch imports the workspaces MATLAB；
The gathered data of step 2, the operating mode imported to batch carries out data regularization；
In embodiment, the frequency spectrum refinement processing of an operating mode is illustrated.It is imported with numerical signal simulation real data
MATLAB carries out regularization.Two signal forms are as follows：
s1=cos (2 π 100t1)+cos(2π101.5t1)+cos(2π102.8t1)
+cos(2π153.2t1)+cos(2π154.7t1)+cos(2π156.2t1)；
t1=0.001:0.001:1s；
s2=cos (2 π 70t2)+cos(2π85t2)+cos(2π112t2)；
t2=0.001:0.001:0.92s；
Sampling number maximum value in two signals is 1000, by s1Automatic zero padding processing is 1000 points；
Step 3, according to the row of Regularization data to dimension serial number, calculate the frequency spectrum of the column data first, and search effective
Crest frequency；
It extracts spectral magnitude maximum value and is used as effective peak criterion, in embodiment, when the FFT amplitudes of the column data are higher than frequently
At 0.1 times of spectral amplitude ratio maximum value, it is determined as effective peak；It is as shown in Figure 1 that effective peak screens schematic diagram.As it can be seen that 100Hz is attached
Effective peak near close and 155Hz is more intensive.Effective peak the selection result is：
[98,99,100,101,102,103,104,105,106,151,152,153,154,155,156,157,158]；
Step 4, the difference on the frequency score value for calculating adjacent effective peak judge whether to need to carry out frequency spectrum refinement processing, are, then
Extraction needs to carry out the initial frequency of micronization processes frequency range and terminates frequency, carry out chirp z transform and preserve as a result, obtaining
The zoom FFT of corresponding frequency band；It is no, directly preserve spectrum results；
In embodiment, when the difference on the frequency score value of effective peak is less than 5 times of primary frequency resolution ratio, it is judged to needing into line frequency
Spectrum refinement.It is as shown in Figure 2 to refine frequency range schematic diagram.As it can be seen that the initial frequency of first segment frequency spectrum refinement frequency range is the 1st effective peak
The frequency 99Hz of value terminates the frequency 106Hz that frequency is the 9th effective peak；The initial frequency of second segment frequency spectrum refinement frequency range
For the frequency 151Hz of the 10th effective peak, the frequency 158Hz that frequency is the 18th effective peak is terminated.
In embodiment, the frequency resolution of refinement is 20 times of primary frequency resolution ratio.Frequency spectrum refinement handles schematic diagram such as Fig. 7
It is shown.As it can be seen that after CZT is refined, the characteristic frequency of signal can be recognized intuitively and effectively, and first segment refinement frequency range is mainly wrapped
Containing 3 characteristic frequencies such as 100Hz, 101.5Hz, 102.8Hz, second segment refine frequency range mainly and include 153.2Hz, 154.7Hz,
3 characteristic frequencies such as 156.2Hz are coincide with emulation input condition.
Step 5 repeats step 3 to step 4, and automatic micronization processes are carried out to the second column data of Regularization data；
In embodiment, the effective peak screening schematic diagram of the second column data is as shown in Fig. 2, refinement frequency range judges schematic diagram such as
Shown in Fig. 3.As it can be seen that the difference on the frequency score value of effective peak is all higher than 5 times of primary frequency resolution ratio, so directly preserving spectrum results.
So far, the batch process that frequency spectrum refines automatically is completed.
In conclusion an embodiment of the present invention provides a kind of batch processing method that frequency spectrum refines automatically, frequency spectrum can be solved
The problems such as multi-state, data volume in analysis are big, data dimension is irregular, characteristic frequency domain spectral line is intensive, realizes the rule of data
Integralization batch processing and the automatic micronization processes of frequency spectrum effectively obtain the frequecy characteristic of data.
It will be understood by those skilled in the art that realizing all or part of flow of above-described embodiment method, meter can be passed through
Calculation machine program is completed to instruct relevant hardware, and the program can be stored in computer readable storage medium.Wherein, institute
It is disk, CD, read-only memory or random access memory etc. to state computer readable storage medium.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Any one skilled in the art in the technical scope disclosed by the present invention, the change or replacement that can be readily occurred in,
It should be covered by the protection scope of the present invention.
Claims (10)
1. a kind of batch processing method that frequency spectrum refines automatically, which is characterized in that include the following steps：
According to different experiment boundary conditions, classify to the gathered data of different operating modes；
To the gathered data of each operating mode, data regularization is carried out, multiple Regularization data are constituted；
According to the row of each Regularization data to dimension serial number, the frequency spectrum per column data is calculated, effective peak of the column data is obtained
It is worth frequency；
The difference on the frequency score value for calculating adjacent effective peak in the column data frequency spectrum judges whether to need to carry out frequency spectrum refinement processing,
It is that then extraction needs to carry out the initial frequency of micronization processes frequency range and terminates frequency, carry out chirp z transform and preserve knot
Fruit obtains the zoom FFT of corresponding frequency band；It is no, directly preserve spectrum results.
2. the batch processing method that frequency spectrum according to claim 1 refines automatically, which is characterized in that it is described to batch import
The gathered data of each operating mode, carrying out data regularization includes：
The gathered data quantity of the operating mode is M, that is, includes M length data vector different in size, find sampled point therein
Number maximum value, as length maximum value, are denoted as N；
Zero padding processing is carried out to remaining M-1 data vector back so that the length of each data vector is N；
By after zero padding vector according to first row, secondary series ... m column arrange, constitute Regularization data.
3. the batch processing method that frequency spectrum according to claim 1 refines automatically, which is characterized in that described to calculate per column data
Frequency spectrum include：
Fast Fourier Transform (FFT) is carried out to the column data；
It takes absolute value to transformation results, obtains spectral magnitude.
4. the batch processing method that frequency spectrum according to claim 3 refines automatically, which is characterized in that described to obtain effective peak
Frequency includes：
Using the maximum value of the spectral magnitude as the criterion of the row effective peak, when amplitude is higher than a times of maximum value, judgement
For effective peak, the corresponding frequency values of extraction effective peak save as effective peak frequency sequence.
5. the batch processing method that frequency spectrum according to claim 1 refines automatically, which is characterized in that described to judge whether to need
Carry out frequency spectrum refinement processing, including frequency differential sequence and the primary frequency resolution ratio b multiple values of sequence comparison effective peak is big
It is small, if difference on the frequency score value is all higher than primary frequency resolution ratio b multiple values, without carrying out frequency spectrum refinement, directly preserve frequency spectrum knot
Fruit；
If difference on the frequency score value is respectively less than primary frequency resolution ratio b multiple values, need to carry out frequency spectrum refinement.
6. the batch processing method that frequency spectrum according to claim 5 refines automatically, which is characterized in that the b value ranges are：
3~5.
7. the batch processing method that frequency spectrum according to claim 5 or 6 refines automatically, which is characterized in that determine frequency spectrum refinement
The initial frequency of frequency range includes：
When difference on the frequency score value is less than primary frequency resolution ratio b multiple values for the first time, extract in the difference value, minuend is corresponding to be had
Imitate initial frequency of the crest frequency point as first segment frequency spectrum refinement frequency range.
8. the batch processing method that the frequency spectrum described according to claim 6 or 7 refines automatically, which is characterized in that determine frequency spectrum refinement
The termination frequency of frequency range includes：
If until the last one serial number of difference sequence, difference on the frequency score value are respectively less than primary frequency resolution ratio b multiple values, then extract most
Termination frequency of the latter effective peak Frequency point as this section of frequency spectrum refinement frequency range；Illustrate that the column data only has one section of frequency range
It needs to carry out frequency spectrum refinement；
If in comparison process, when frequency of occurrences difference value is more than primary frequency resolution ratio b multiple values, extracted in the difference value, quilt
Termination frequency of the corresponding effective peak Frequency point of subtrahend as first segment frequency spectrum refinement frequency range.
9. the batch processing method that frequency spectrum according to claim 8 refines automatically, which is characterized in that
Chirp z transform (CZT) is carried out according to the following formula, and preserves result；
Wherein：
X (n), fs are the column data and sample frequency of the operating mode after Regularization respectively；
fmin1,fmax1, initial frequency, the termination frequency of respectively this section frequency spectrum refinement frequency range；
R=(fmax1-fmin1)/Δ f, the points of this section of frequency spectrum refinement frequency range CZT；
R=0,1 ... R-1, the serial number of this section of frequency spectrum refinement frequency range CZT；
Δ f is the frequency resolution after frequency spectrum refinement.
10. the batch processing method that frequency spectrum according to claim 9 refines automatically, which is characterized in that further include：It traverses each
Each column data in a operating mode is handled to obtain the CZT of frequency spectrum refinement frequency range, and preserves result；Until institute's refinement in need
Frequency range complete frequency spectrum refinement processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810067689.8A CN108334682B (en) | 2018-01-24 | 2018-01-24 | Batch processing method for automatically refining frequency spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810067689.8A CN108334682B (en) | 2018-01-24 | 2018-01-24 | Batch processing method for automatically refining frequency spectrum |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108334682A true CN108334682A (en) | 2018-07-27 |
| CN108334682B CN108334682B (en) | 2021-08-17 |
Family
ID=62925490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810067689.8A Active CN108334682B (en) | 2018-01-24 | 2018-01-24 | Batch processing method for automatically refining frequency spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108334682B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114252698A (en) * | 2020-09-24 | 2022-03-29 | 广州慧睿思通科技股份有限公司 | Frequency spectrum processing method, system, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002236135A (en) * | 2001-02-07 | 2002-08-23 | Furuno Electric Co Ltd | Method for detecting signal frequency |
| CN102707275A (en) * | 2012-05-25 | 2012-10-03 | 北京航空航天大学 | Digital processing method of altimeter of linear frequency modulation continuous wave radar |
| CN103454495A (en) * | 2013-09-13 | 2013-12-18 | 电子科技大学 | Self-adaptive high-precision fast spectral analysis method |
| CN103743470A (en) * | 2013-12-23 | 2014-04-23 | 广西科技大学 | Automobile noise spectrum analyzing method |
-
2018
- 2018-01-24 CN CN201810067689.8A patent/CN108334682B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002236135A (en) * | 2001-02-07 | 2002-08-23 | Furuno Electric Co Ltd | Method for detecting signal frequency |
| CN102707275A (en) * | 2012-05-25 | 2012-10-03 | 北京航空航天大学 | Digital processing method of altimeter of linear frequency modulation continuous wave radar |
| CN103454495A (en) * | 2013-09-13 | 2013-12-18 | 电子科技大学 | Self-adaptive high-precision fast spectral analysis method |
| CN103743470A (en) * | 2013-12-23 | 2014-04-23 | 广西科技大学 | Automobile noise spectrum analyzing method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114252698A (en) * | 2020-09-24 | 2022-03-29 | 广州慧睿思通科技股份有限公司 | Frequency spectrum processing method, system, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108334682B (en) | 2021-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106679741B (en) | Processing method and system based on vortex-shedding meter anti-jamming signal | |
| CN103471841B (en) | A kind of rotating machinery vibrating failure diagnosis method | |
| CN107013449A (en) | Voice signal based on deep learning recognizes the method and system of compressor fault | |
| CN107620868A (en) | Pipeline leakage detection method and device | |
| CN107202028B (en) | A kind of turbocharger centrifugal compressor surge recognition methods | |
| CN109870282B (en) | Blade vibration feature identification method and system based on compressed sensing | |
| CN106405230A (en) | Frequency measuring method and apparatus | |
| CN101726414A (en) | Method and system thereof for measuring trial run parameter of aeroengine | |
| US7945606B2 (en) | Method and apparatus for evaluating a time varying signal | |
| Zhao et al. | Variational time-domain decomposition of reciprocating machine multi-impact vibration signals | |
| Li et al. | A VME method based on the convergent tendency of VMD and its application in multi-fault diagnosis of rolling bearings | |
| CN103941722A (en) | Method monitoring and diagnosing equipment failure through component characteristic frequency multiplication amplitude trend | |
| CN108334682A (en) | A kind of batch processing method that frequency spectrum refines automatically | |
| You et al. | Fault diagnosis system of rotating machinery vibration signal | |
| CN113128047A (en) | Aero-engine thrust calculation method, storage medium and calculation device | |
| CN111948454B (en) | Synchronous phasor measurement method and system for resisting attenuated direct current component interference | |
| CN106597095B (en) | A kind of frequency-tracking implementation method of dual processors equal interval sampling | |
| CN108106827A (en) | A kind of rotor platform rises amplifier terminals under rotary regimes and loosens recognition methods | |
| CN105700012B (en) | A kind of marine non-repeatability time-lapse seismic frequency dividing cross-equalization processing method | |
| KR20180022330A (en) | Data measure system for performance verification test of marine large engine and method thereof | |
| CN204759416U (en) | Gas production curve is synthesized to gas well and analytic system thereof | |
| CN104068851B (en) | Eeg signal acquisition system and method based on FPGA | |
| CN202770852U (en) | Velocity measuring device used for servo motor in low-speed operation | |
| CN102571032B (en) | Method for post-processing measured data by using scrolling smoothing filtering | |
| CN112232414B (en) | Triple concurrency fault analysis method based on X and Y dual-measurement-point spectrum data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |