CN105652084B - Wind power plant 3p CFVFs detection method based on EIAMD - Google Patents
Wind power plant 3p CFVFs detection method based on EIAMD Download PDFInfo
- Publication number
- CN105652084B CN105652084B CN201511031105.4A CN201511031105A CN105652084B CN 105652084 B CN105652084 B CN 105652084B CN 201511031105 A CN201511031105 A CN 201511031105A CN 105652084 B CN105652084 B CN 105652084B
- Authority
- CN
- China
- Prior art keywords
- frequency
- eiamd
- amplitude
- follows
- row
- 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.)
- Expired - Fee Related
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 25
- 239000011159 matrix material Substances 0.000 claims abstract description 18
- 238000001228 spectrum Methods 0.000 claims abstract description 9
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 4
- 238000005070 sampling Methods 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 9
- 238000000926 separation method Methods 0.000 claims description 8
- 241001269238 Data Species 0.000 claims description 3
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 12
- 238000000354 decomposition reaction Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000001914 filtration Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R23/00—Arrangements for measuring frequencies; Arrangements for analysing frequency spectra
- G01R23/16—Spectrum analysis; Fourier analysis
Landscapes
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
- Measuring Frequencies, Analyzing Spectra (AREA)
Abstract
本发明涉及风电场电压波动检测方法，具体是一种基于EIAMD的风电场3p CFVFs检测方法。本发明解决了现有风电场电压波动检测方法无法有效检测由多个电压波动组成的3p CFVFs的问题。基于EIAMD的风电场3p CFVFs检测方法，该方法是采用如下步骤实现的：步骤S1：电压信号u(t)的采集；步骤S2：等间隔宽度Δfb的选择；步骤S3：检测频率区间[fmin,fmax]的选择；步骤S4：各区间截止频率fbi的选择；步骤S5：区间的选择；步骤S6：EIAMD分量矩阵的形成；步骤S7：EIAMD分量矩阵的列重构；步骤S8：EIAMD分量矩阵的行简化；步骤S9：瞬时幅值ai(t)和瞬时频率fi′(t)的计算；步骤S10：瞬时频率fi′(t)的修正；步骤S11：边际谱h(t)的计算；步骤S12：m个电压波动的频率和幅值的求取。本发明适用于风电场3p CFVFs检测。
The invention relates to a method for detecting voltage fluctuations in wind farms, in particular to a method for detecting 3p CFVFs in wind farms based on EIAMD. The invention solves the problem that the existing wind farm voltage fluctuation detection method cannot effectively detect 3p CFVFs composed of multiple voltage fluctuations. EIAMD-based wind farm 3p CFVFs detection method, the method is realized by the following steps: step S1: acquisition of voltage signal u(t); step S2: selection of equal interval width Δf b ; step S3: detection frequency interval [f min , f max ] selection; Step S4: selection of cut-off frequency f bi of each interval; Step S5: selection of interval; Step S6: formation of EIAMD component matrix; Step S7: column reconstruction of EIAMD component matrix; Step S8: Row simplification of EIAMD component matrix; step S9: calculation of instantaneous amplitude a i (t) and instantaneous frequency f i '(t); step S10: correction of instantaneous frequency f i '(t); step S11: marginal spectrum h Calculation of (t); Step S12: Obtaining the frequency and amplitude of m voltage fluctuations. The invention is suitable for the detection of 3p CFVFs in wind farms.
Description
Technical field
It is specifically a kind of based on EIAMD (resolution modalities point at equal intervals the present invention relates to wind power plant voltage fluctuation detection method
Solution) wind power plant 3p CFVFs detection method.
Background technique
Voltage fluctuation can generate light flash, and bring harm to sensibility load, cause equipment normal when serious
Work.Wind power plant can generate 3p tight frequency voltage fluctuation (Close in actual operation, at points of common connection
Frequency Voltage Fluctuations, CFVFs), concrete reason is as follows: voltage fluctuation is mainly drawn by power swing
It rises, for currently used 3 blade blower, wind shear, tower shadow effect and yaw system can often change the line of production in blower and give birth to 3 times of output works
The oscillation of rate, thus generate 3p voltage fluctuation (p is blower rotational frequency).Even if being become in the case that wind speed is certain by wind speed
Power swing caused by changing is zero, but still can generate 3p voltage fluctuation.The wind speed of more Fans is different in a wind power plant, because
And its revolving speed is also different, one group of 3p being made of multiple voltage fluctuations can be generated at points of common connection around center wind speed at this time
CFVFs.Studies have shown that the frequency range of this group of 3p CFVFs is mostly within 1Hz, the lowest difference of each frequency in 0.01-0.02Hz,
It is even more small.
Voltage fluctuation bring harm in order to prevent, at present both at home and abroad mainly using Fourier transformation, in short-term Fourier's change
It changes, wavelet transformation, Hilbert-Huang transform and parsing the methods of mode decomposition examine the frequency and amplitude of voltage fluctuation
It surveys.However practice have shown that, existing detection method cannot achieve effective inspection to this group of 3p CFVFs since itself principle is limited
It surveys, is specifically described as follows: first, Fourier transformation, Short Time Fourier Transform are due to by temporal resolution and frequency resolution
Influence and be not suitable for the detections of multiple signals；Second, wavelet transformation is suitble to the detection of multiple signals, but with Decomposition order
Increase, the signal length decomposited is reduced at 2 power side, leverages the accuracy of amplitude calculating；Third, Martin Hilb
Spy-Huang is also suitble to the detection of multiple signals, passes through empirical mode decomposition (Empirical Mode first
Decomposition, EMD) by original signal be decomposed into each intrinsic mode function (Intrinsic Mode Function,
IMF), but it can not effectively decompose frequency and pass through side due to the decomposition deviation of each IMF than the CFVFs (as shown in Figure 3) less than 1.5
The frequency and amplitude that border is composed can not effectively reflect the essential attribute (as shown in Figure 4) of original signal；Fourth, resolution modalities point
Solution (Analytic Mode Decomposition, AMD) can be equivalent to a digital filter, and filtering performance will receive
The influence of the end effect generated because the sampling time is shorter, although avoid EMD decompose when due to cubic spline interpolation and
The end effect of generation, but Hilbert transformation still can be in AMD (as shown in Figure 5, Figure 6), instantaneous amplitude (as shown in Figure 7) and wink
When frequency (as shown in Figure 8) calculating process in generate end effect and cause modal overlap, its filtering performance is influenced, thus nothing
Method realizes effective detection to CFVFs.Based on this, it is necessary to a kind of completely new wind power plant voltage fluctuation detection method is invented, with
Solve the problems, such as that existing wind power plant voltage fluctuation detection method can not effectively detect the 3p CFVFs being made of multiple voltage fluctuations.
Summary of the invention
The present invention can not be detected effectively to solve existing wind power plant voltage fluctuation detection method by multiple voltage fluctuation groups
At 3p CFVFs the problem of, provide a kind of wind power plant 3p CFVFs detection method based on EIAMD.
The present invention is achieved by the following technical scheme: the wind power plant 3p CFVFs detection method based on EIAMD, the party
Method is realized using following steps:
Step S1: voltage signal u (t) acquisition；Voltage signal can both be expressed as power-frequency voltage and m CFVFs modulation
It forms, 2m+1 tight frequency m-Acetyl chlorophosphonazo (Close Frequency Inter Harmonics, CFIHs) can also be expressed as
It is added；When acquisition, sample frequency fsMore than or equal to 2 times highest detection frequencies, frequency resolution are fallen equal to sampling time T's
Number, sample frequency fsHigher with frequency resolution, sampling number N is more, and calculation amount is bigger；
Step S2: width Delta f at equal intervalsbSelection；When selection, it then follows following principle:
1) it is equal to the integral multiple of frequency resolution；
2) it is less than or equal to the half of CFVFs minimum frequency difference；
3) take into account the error calculated of amplitude: interval selection is smaller, and the minimum frequency difference being capable of detecting when is smaller, but width
It is bigger to be worth error；Conversely, the minimum frequency difference being capable of detecting when is bigger, amplitude error is smaller；Generally, certain in order to guarantee
The minimum frequency being capable of detecting when is poor, at equal intervals width Delta fbIt can be suitably smaller；
Step S3: detection frequency separation [fmin,fmax] selection；fminFor the lower limit for detecting frequency, fmaxTo detect frequency
The upper limit, be typically chosen in the integral multiple of frequency resolution；
Step S4: each section cutoff frequency fbiSelection；Selection course is as follows:
According to width Delta f at equal intervalsbWith detection frequency separation [fmin,fmax], determine each section cutoff frequency fbi, determine public
Formula is as follows:
fbi=fmin+(i-1)·Δfb；
I=2,3 ... n；
fb1=fmin；
fb(n+1)=fmax；
Step S5: the selection in section；Selection course is as follows:
According to each section cutoff frequency fbi, determine that the n+2 section of EIAMD, n+2 section are expressed as: (0, fb1],
[fb1,fb2], [fb2,fb3] ..., [fb(n-1),fbn],[fbn,fb(n+1)], [fb(n+1),+∞]；
The formation of step S6:EIAMD Component Matrices；Forming process is as follows:
According to each section cutoff frequency fbi, calculate each component u in n+2 sectioni(t), calculation formula is as follows:
ui(t)=si(t)-si-1(t)；
si(t)=2 π f of sinbit·H[u(t)cos 2πfbit]-cos 2πfbit·H[u(t)sin 2πfbit]；
I=1,2 ... n+1；
s0(t)=0；
un+2(t)=u (t)-sn+1(t)；
In above formula: H [] is Hilbert transformation；
According to the component u in n+2 sectioni(t), the row of EIAMD Component Matrices is determined；According to sampling number N, determine
The column of EIAMD Component Matrices；EIAMD Component Matrices indicate are as follows: u(n+2)×N；
The column of step S7:EIAMD Component Matrices reconstruct；Column restructuring procedure is as follows:
Assuming that the intermediate data of each row of EIAMD Component MatricesAnd it adopts
Reconstitute EIAMD Component Matrices after replacing two side datas with intermediate data；After reconstituting the matrix byA N0Composition,
And indicate are as follows: N=[N0,N0,…,N0]；
The row of step S8:EIAMD Component Matrices simplifies；It is as follows that row simplifies process:
The 2i-1 row of EIAMD Component Matrices and 2i row are added, threshold value appropriate is selected, looks for
Each row maximum value is less than going and leaving out for threshold value in trip, finds out the row that each row maximum value in row is more than or equal to threshold value
And retain, the EIAMD decomposed component matrix after being thus simplified；Assuming that the row that maximum value is less than threshold value in row has l,
Then simplified EIAMD Component Matrices indicate are as follows:
Step S9: instantaneous amplitude ai(t) and instantaneous frequency fiThe calculating of ' (t)；Calculation formula is as follows:
Step S10: instantaneous frequency fiThe amendment of ' (t)；Makeover process is as follows:
Count each instantaneous frequency fiThe frequency that each frequency occurs in ' (t), and the highest frequency of frequency of occurrence is selected to replace
Thus all instantaneous frequencys obtain revised instantaneous frequency fi(t),
Step S11: marginal spectrum h (t) calculating；Calculation formula is as follows:
In marginal spectrum h (t), the frequency of amplitude non-zero is the m-Acetyl chlorophosphonazo frequency calculated, and amplitude is m-Acetyl chlorophosphonazo
Amplitude corresponding to frequency picks out the preceding 2m+1 amplitude and its corresponding frequency of amplitude maximum, the as amplitude of CHIHs
And frequency；
The frequency of S12:m voltage fluctuation of step and seeking for amplitude；Finding process is as follows:
The frequency of 2m+1 CFIHs obtained in step S11 is arranged according to ascending sequence, then between m+1 is a
The frequency f of harmonic wavem+1As power-frequency voltage frequency, corresponding amplitude are um+1, the frequency of remaining CFIHs is about power frequency electric voltage-frequency
Rate fm+1Symmetrically, and symmetrical two CFIHs has approximately equal amplitude；
According to the frequency of CFIHs and its corresponding amplitude, the frequency and amplitude of m voltage fluctuation, calculation formula are calculated
It is as follows:
I=1,2 ..., m；
In above formula: the frequency of m voltage fluctuation isThe amplitude of m voltage fluctuation is ui+u2m+2-i。
Compared with existing wind power plant voltage fluctuation detection method, the wind power plant 3p of the present invention based on EIAMD
CFVFs detection method passes through the selection of reciprocity interval width and column reconstruct, row simplification and the instantaneous frequency of EIAMD Component Matrices
Amendment, alleviate the influence of end effect, improve the filtering performance of AMD, realize frequency and amplitude to 3p CFVFs
Effective detection, thus it has following advantage: first, the present invention is not compared with Fourier transformation, Short Time Fourier Transform
It is suitable only for the detection of single time varying signal, and can effectively detect the CFVFs being made of multiple voltage fluctuations.Second, with small
Wave conversion is compared with Hilbert-Huang transform, and the present invention can be realized no longer by frequency than the constraint less than 1.5 to smaller frequency
Effective detection of the CFVFs of rate difference, but need to be selected between the minimum frequency difference being able to detect that and amplitude error.Its
Three, compared with AMD, present invention improves its filtering performances, give equally spaced selection principle and preferably alleviate because adopting
The end effect problem that the sample time is shorter and generates, improves the accuracy of testing result.
The present invention efficiently solves existing wind power plant voltage fluctuation detection method and can not effectively detect by multiple voltage fluctuations
The problem of 3p CFVFs of composition, is suitable for wind power plant 3p CFVFs and detects.
Detailed description of the invention
Fig. 1 is overall flow schematic diagram of the invention.
Fig. 2 is the schematic diagram of sampled signal.
Fig. 3 is the schematic diagram of each IMF based on EMD.
Fig. 4 is the schematic diagram of the marginal spectrum based on EMD.
Fig. 5 is the schematic diagram of the calculated result of the S based on AMD.
Fig. 6 is the schematic diagram of each component amplitude based on AMD.
Fig. 7 is the schematic diagram of the instantaneous amplitude based on AMD.
Fig. 8 is the schematic diagram of the instantaneous frequency based on AMD.
Fig. 9 is the schematic diagram using column reconstruct and the simplified each component amplitude of row based on EIAMD.
Figure 10 is each frequency frequency of occurrence system using column reconstruct and row after simplified before instantaneous frequency amendment based on EIAMD
The schematic diagram of meter.
Figure 11 is the schematic diagram using column reconstruct, row simplification and the revised instantaneous frequency of instantaneous frequency based on EIAMD.
Figure 12 is the schematic diagram using column reconstruct, row simplification and the revised marginal spectrum of instantaneous frequency based on EIAMD.
Specific embodiment
Wind power plant 3p CFVFs detection method based on EIAMD, this method are realized using following steps:
Step S1: voltage signal u (t) acquisition；Voltage signal can both be expressed as power-frequency voltage and m CFVFs modulation
It forms, 2m+1 tight frequency m-Acetyl chlorophosphonazo (CFIHs) can also be expressed as and be added；When acquisition, sample frequency fsGreater than etc.
In 2 times of highest detection frequencies, frequency resolution is equal to the inverse of sampling time T, sample frequency fsIt is higher with frequency resolution, it adopts
Number of samples N is more, and calculation amount is bigger；
Step S2: width Delta f at equal intervalsbSelection；When selection, it then follows following principle:
1) it is equal to the integral multiple of frequency resolution；
2) it is less than or equal to the half of CFVFs minimum frequency difference；
3) take into account the error calculated of amplitude: interval selection is smaller, and the minimum frequency difference being capable of detecting when is smaller, but width
It is bigger to be worth error；Conversely, the minimum frequency difference being capable of detecting when is bigger, amplitude error is smaller；Generally, certain in order to guarantee
The minimum frequency being capable of detecting when is poor, at equal intervals width Delta fbIt can be suitably smaller；
Step S3: detection frequency separation [fmin,fmax] selection；fminFor the lower limit for detecting frequency, fmaxTo detect frequency
The upper limit, be typically chosen in the integral multiple of frequency resolution；
Step S4: each section cutoff frequency fbiSelection；Selection course is as follows:
According to width Delta f at equal intervalsbWith detection frequency separation [fmin,fmax], determine each section cutoff frequency fbi, determine public
Formula is as follows:
fbi=fmin+(i-1)·Δfb；
I=2,3 ... n；
fb1=fmin；
fb(n+1)=fmax；
Step S5: the selection in section；Selection course is as follows:
According to each section cutoff frequency fbi, determine that the n+2 section of EIAMD, n+2 section are expressed as: (0, fb1],
[fb1,fb2], [fb2,fb3] ..., [fb(n-1),fbn],[fbn,fb(n+1)], [fb(n+1),+∞]；
The formation of step S6:EIAMD Component Matrices；Forming process is as follows:
According to each section cutoff frequency fbi, calculate each component u in n+2 sectioni(t), calculation formula is as follows:
ui(t)=si(t)-si-1(t)；
si(t)=2 π f of sinbit·H[u(t)cos 2πfbit]-cos 2πfbit·H[u(t)sin 2πfbit]；
I=1,2 ... n+1；
s0(t)=0；
un+2(t)=u (t)-sn+1(t)；
In above formula: H [] is Hilbert transformation；
According to the component u in n+2 sectioni(t), the row of EIAMD Component Matrices is determined；According to sampling number N, determine
The column of EIAMD Component Matrices；EIAMD Component Matrices indicate are as follows: u(n+2)×N；
The column of step S7:EIAMD Component Matrices reconstruct；Column restructuring procedure is as follows:
Assuming that the intermediate data of each row of EIAMD Component MatricesAnd it uses
Intermediate data reconstitutes EIAMD Component Matrices after replacing two side datas；After reconstituting the matrix byA N0Composition, and
It indicates are as follows: N=[N0,N0,…,N0]；
The row of step S8:EIAMD Component Matrices simplifies；It is as follows that row simplifies process:
The 2i-1 row of EIAMD Component Matrices and 2i row are added, threshold value appropriate is selected, looks for
Each row maximum value is less than going and leaving out for threshold value in trip, finds out the row that each row maximum value in row is more than or equal to threshold value
And retain, the EIAMD decomposed component matrix after being thus simplified；Assuming that the row that maximum value is less than threshold value in row has l,
Then simplified EIAMD Component Matrices indicate are as follows:
Step S9: instantaneous amplitude ai(t) and instantaneous frequency fiThe calculating of ' (t)；Calculation formula is as follows:
Step S10: instantaneous frequency fiThe amendment of ' (t)；Makeover process is as follows:
Count each instantaneous frequency fiThe frequency that each frequency occurs in ' (t), and the highest frequency of frequency of occurrence is selected to replace
Thus all instantaneous frequencys obtain revised instantaneous frequency fi(t),
Step S11: marginal spectrum h (t) calculating；Calculation formula is as follows:
In marginal spectrum h (t), the frequency of amplitude non-zero is the m-Acetyl chlorophosphonazo frequency calculated, and amplitude is m-Acetyl chlorophosphonazo
Amplitude corresponding to frequency picks out the preceding 2m+1 amplitude and its corresponding frequency of amplitude maximum, the as amplitude of CHIHs
And frequency；
The frequency of S12:m voltage fluctuation of step and seeking for amplitude；Finding process is as follows:
The frequency of 2m+1 CFIHs obtained in step S11 is arranged according to ascending sequence, then between m+1 is a
The frequency f of harmonic wavem+1As power-frequency voltage frequency, corresponding amplitude are um+1, the frequency of remaining CFIHs is about power frequency electric voltage-frequency
Rate fm+1Symmetrically, and symmetrical two CFIHs has approximately equal amplitude；
According to the frequency of CFIHs and its corresponding amplitude, the frequency and amplitude of m voltage fluctuation, calculation formula are calculated
It is as follows:
I=1,2 ..., m；
In above formula: the frequency of m voltage fluctuation isThe amplitude of m voltage fluctuation is ui+u2m+2-i。
When it is implemented,
In the step S1, voltage signal u (t) is indicated are as follows:
U (t)=cos (2 π * 50t) [1+0.2cos (2 π * 0.023t)]=cos (2 π * 50t)+0.1cos (2 π *
49.977t)+0.1cos(2π*50.023t)；
Sample frequency fsIt is 100s for 256Hz, sampling time T, sampling number N is 25600；
In the step S2, width Delta f at equal intervalsbFor 0.01Hz；
In the step S3, frequency separation [f is detectedmin,fmax] it is [49,51]；
In the step S4, each section cutoff frequency
In the step S5,202 sections that resolution modalities decompose at equal intervals are expressed as: (0,49], [49,49.01],
[49.01,49.02] ... [50.98,50.99], [50.99,51], [51 ,+∞]；
In the step S6, resolution modalities decomposed component matrix is expressed as at equal intervals: [u]202×25600[u′]202×25600；
In the step S7, the intermediate data N of resolution modalities decomposed component matrix at equal intervals0=[12544,13055]；Weight
The decomposed component of the resolution modalities at equal intervals matrix newly constituted is by 50 N0Composition；
In the step S8, the simplified decomposed component matrix of resolution modalities at equal intervals is expressed as: [u]3×25600, such as Fig. 9
It is shown；
In the step S10, the highest frequency of frequency of occurrence is 49.98Hz, 50.00Hz, 50.02Hz, such as Figure 10, Figure 11
It is shown；
In the step S11, the number of amplitude corresponding to m-Acetyl chlorophosphonazo frequency and m-Acetyl chlorophosphonazo frequency is 3；It is humorous between 3
Wave frequency rate is respectively 49.98Hz, 50.00Hz, 50.02Hz；Amplitude corresponding to 3 m-Acetyl chlorophosphonazo frequencies is respectively 0.0944V,
0.9969V, 0.0945V, as shown in figure 12；
In the step S12, calculation formula is as follows:
U (t)=0.9969cos (2 π * 50t)+0.0944cos (2 π * 49.98t)+0.0945cos (2 π * 50.02t)
≈0.9969cos(2π*50t)[1+0.1895cos(2π*0.02t)]；
The frequency of voltage fluctuation is 0.02Hz, and the amplitude of voltage fluctuation is 0.1889V；The absolute error of frequency is
0.003Hz；The absolute error of amplitude is 0.0111V.
Claims (2)
1. a kind of wind power plant 3p CFVFs detection method based on EIAMD, wherein EIAMD represents resolution modalities at equal intervals and decomposes；
3p represents three times blower rotational frequency；CFVFs represents tight frequency voltage fluctuation；It is characterized by: this method is using as follows
What step was realized:
Step S1: voltage signal u (t) acquisition；Voltage signal is expressed as being modulated by power-frequency voltage and m CFVFs, also can
2m+1 tight frequency m-Acetyl chlorophosphonazo is expressed as to be added；When acquisition, sample frequency fsMore than or equal to voltage signal u (t) highest
2 times of frequency, frequency resolution are equal to the inverse of sampling time T, sample frequency fsIt is higher with frequency resolution, sampling number N
More, calculation amount is bigger；
Step S2: width Delta f at equal intervalsbSelection；When selection, it then follows following principle:
1) it is equal to the integral multiple of frequency resolution；
2) it is less than or equal to the half of CFVFs minimum frequency difference；
3) take into account the error calculated of amplitude: interval selection is smaller, and the minimum frequency difference being capable of detecting when is smaller, but amplitude is missed
Difference is bigger；Conversely, the minimum frequency difference being capable of detecting when is bigger, amplitude error is smaller；Generally, in order to guarantee it is certain can
The minimum frequency detected is poor, at equal intervals width Delta fbIt should be suitably smaller；
Step S3: detection frequency separation [fmin,fmax] selection；fminFor the lower limit for detecting frequency, fmaxFor the upper of detection frequency
Limit, is typically chosen in the integral multiple of frequency resolution；
Step S4: each section cutoff frequency fbiSelection；Selection course is as follows:
According to width Delta f at equal intervalsbWith detection frequency separation [fmin,fmax], determine each section cutoff frequency fbi, determine formula such as
Under:
fbi=fmin+(i-1)·Δfb；
I=2,3 ... n；
fb1=fmin；
fb(n+1)=fmax；
Step S5: the selection in section；Selection course is as follows:
According to each section cutoff frequency fbi, determine that the n+2 section of EIAMD, n+2 section are expressed as: (0, fb1], [fb1,
fb2], [fb2,fb3] ..., [fb(n-1),fbn],[fbn,fb(n+1)], [fb(n+1),+∞]；
The formation of step S6:EIAMD Component Matrices；Forming process is as follows:
According to each section cutoff frequency fbi, calculate each component u in n+2 sectioni(t), calculation formula is as follows:
ui(t)=si(t)-si-1(t)；
si(t)=sin2 π fbit·H[u(t)cos2πfbit]-cos2πfbit·H[u(t)sin2πfbit]；
I=1,2 ... n+1；
s0(t)=0；
un+2(t)=u (t)-sn+1(t)；
In above formula: H [] is Hilbert transformation；
According to the component u in n+2 sectioni(t), the row of EIAMD Component Matrices is determined；According to sampling number N, EIAMD component is determined
Matrix column；EIAMD Component Matrices indicate are as follows: u(n+2)×N；
The column of step S7:EIAMD Component Matrices reconstruct；Column restructuring procedure is as follows:
Assuming that the intermediate data of each row of EIAMD Component MatricesAnd in using
Between data replace two side datas after reconstitute EIAMD Component Matrices；After reconstituting the matrix byComposition, and table
It is shown as: N=[N0,N0,…,N0]；
The row of step S8:EIAMD Component Matrices simplifies；It is as follows that row simplifies process:
The 2i-1 row of EIAMD Component Matrices and 2i row are added,Threshold value appropriate is selected, is found out
Each row maximum value is less than going and leaving out for threshold value in row, finds outEach row maximum value is more than or equal to row and the guarantor of threshold value in row
It stays, the EIAMD decomposed component matrix after being thus simplified；Assuming thatThe row that maximum value is less than threshold value in row has l, then simple
EIAMD Component Matrices after change indicate are as follows:
Step S9: instantaneous amplitude ai(t) and instantaneous frequency f 'i(t) calculating；Calculation formula is as follows:
Step S10: instantaneous frequency f 'i(t) amendment；Makeover process is as follows:
Count each instantaneous frequency f 'i(t) frequency that each frequency occurs in, and it is all to select the highest frequency of frequency of occurrence to replace
Thus instantaneous frequency obtains revised instantaneous frequency fi(t),
Step S11: marginal spectrum h (t) calculating；Calculation formula is as follows:
In formula: Re is to take real part, and j is imaginary unit；
In marginal spectrum h (t), the frequency of amplitude non-zero is the m-Acetyl chlorophosphonazo frequency calculated, and amplitude is m-Acetyl chlorophosphonazo frequency
Corresponding amplitude picks out the preceding 2m+1 amplitude and its corresponding frequency of amplitude maximum, as tight frequency m-Acetyl chlorophosphonazo
Amplitude and frequency；
The frequency of S12:m voltage fluctuation of step and seeking for amplitude；Finding process is as follows:
The frequency of 2m+1 tight frequency m-Acetyl chlorophosphonazo obtained in step S11 is arranged according to ascending sequence, then m+1
The frequency f of a m-Acetyl chlorophosphonazom+1As power-frequency voltage frequency, corresponding amplitude are um+1, the frequency of remaining tight frequency m-Acetyl chlorophosphonazo
About power-frequency voltage frequency fm+1Symmetrically, and symmetrical two tight frequencies m-Acetyl chlorophosphonazo has approximately equal amplitude；
According to the frequency of tight frequency m-Acetyl chlorophosphonazo and its corresponding amplitude, the frequency and amplitude of m voltage fluctuation are calculated, is calculated
Formula is as follows:
I=1,2 ..., m；
In above formula: the frequency of m voltage fluctuation isThe amplitude of m voltage fluctuation is ui+u2m+2-i。
2. the wind power plant 3p CFVFs detection method according to claim 1 based on EIAMD, it is characterised in that:
In the step S1, voltage signal u (t) is indicated are as follows:
U (t)=cos (2 π * 50t) [1+0.2cos (2 π * 0.023t)]=cos (2 π * 50t)+0.1cos (2 π * 49.977t)+
0.1cos(2π*50.023t)；
Sample frequency fsIt is 100s for 256Hz, sampling time T, sampling number N is 25600；
In the step S2, width Delta f at equal intervalsbFor 0.01Hz；
In the step S3, frequency separation [f is detectedmin,fmax] it is [49,51]；
In the step S4, each section cutoff frequency
In the step S5,202 sections that resolution modalities decompose at equal intervals are expressed as: (0,49], [49,49.01],
[49.01,49.02] ... [50.98,50.99], [50.99,51], [51 ,+∞]；
In the step S6, resolution modalities decomposed component matrix is expressed as at equal intervals: [u]202×25600；
In the step S7, the intermediate data N of resolution modalities decomposed component matrix at equal intervals0=[12544,13055]；Again structure
At the decomposed component of resolution modalities at equal intervals matrix by 50 N0Composition；
In the step S8, the simplified decomposed component matrix of resolution modalities at equal intervals is expressed as: [u]3×25600；
In the step S10, the highest frequency of frequency of occurrence is 49.98Hz, 50.00Hz, 50.02Hz；
In the step S11, the number of amplitude corresponding to m-Acetyl chlorophosphonazo frequency and m-Acetyl chlorophosphonazo frequency is 3；3 m-Acetyl chlorophosphonazo frequencies
Rate is respectively 49.98Hz, 50.00Hz, 50.02Hz；Amplitude corresponding to 3 m-Acetyl chlorophosphonazo frequencies is respectively 0.0944V,
0.9969V, 0.0945V；
In the step S12, calculation formula is as follows:
The frequency of voltage fluctuation is 0.02Hz, and the amplitude of voltage fluctuation is 0.1889V；The absolute error of frequency is 0.003Hz；Width
The absolute error of value is 0.0111V.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511031105.4A CN105652084B (en) | 2015-12-31 | 2015-12-31 | Wind power plant 3p CFVFs detection method based on EIAMD |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511031105.4A CN105652084B (en) | 2015-12-31 | 2015-12-31 | Wind power plant 3p CFVFs detection method based on EIAMD |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105652084A CN105652084A (en) | 2016-06-08 |
| CN105652084B true CN105652084B (en) | 2019-01-15 |
Family
ID=56491270
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511031105.4A Expired - Fee Related CN105652084B (en) | 2015-12-31 | 2015-12-31 | Wind power plant 3p CFVFs detection method based on EIAMD |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105652084B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108508268A (en) * | 2018-04-02 | 2018-09-07 | 山西大学 | A kind of novel digital type flickermeter decomposed based on time-frequency domain |
| CN108871742B (en) * | 2018-05-03 | 2021-08-13 | 西安交通大学 | An Improved Feature Order Extraction Method for Keyless Phase Faults |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6631325B1 (en) * | 1996-08-12 | 2003-10-07 | The United States As Represented By The Administrator Of The National Aeronautics And Space Administration | Computer implemented empirical mode decomposition method apparatus, and article of manufacture utilizing curvature extrema |
| CN103163372A (en) * | 2013-03-26 | 2013-06-19 | 山西省电力公司长治供电分公司 | Method for analyzing harmonics of power system by adopting Hilbert-Huang transform (HHT) |
| CN104330624A (en) * | 2014-09-15 | 2015-02-04 | 燕山大学 | Method for detecting closely spaced frequency components of non-stationary signals |
| CN104567705A (en) * | 2014-12-19 | 2015-04-29 | 南京航空航天大学 | Strain and temperature aliasing signal decoupling method for optical fiber grating sensor under dynamic load |
| CN104677580A (en) * | 2015-01-30 | 2015-06-03 | 燕山大学 | Rotary mechanical fault diagnosis method based on analytical modal decomposition |
-
2015
- 2015-12-31 CN CN201511031105.4A patent/CN105652084B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6631325B1 (en) * | 1996-08-12 | 2003-10-07 | The United States As Represented By The Administrator Of The National Aeronautics And Space Administration | Computer implemented empirical mode decomposition method apparatus, and article of manufacture utilizing curvature extrema |
| CN103163372A (en) * | 2013-03-26 | 2013-06-19 | 山西省电力公司长治供电分公司 | Method for analyzing harmonics of power system by adopting Hilbert-Huang transform (HHT) |
| CN104330624A (en) * | 2014-09-15 | 2015-02-04 | 燕山大学 | Method for detecting closely spaced frequency components of non-stationary signals |
| CN104567705A (en) * | 2014-12-19 | 2015-04-29 | 南京航空航天大学 | Strain and temperature aliasing signal decoupling method for optical fiber grating sensor under dynamic load |
| CN104677580A (en) * | 2015-01-30 | 2015-06-03 | 燕山大学 | Rotary mechanical fault diagnosis method based on analytical modal decomposition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105652084A (en) | 2016-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106840637B (en) | Time-Frequency Analysis Method of GIS Mechanical Vibration Signal Based on Improved HHT Algorithm | |
| CN104502700B (en) | Sinusoidal parameter measurement method and system of power signal | |
| CN101587146B (en) | A Detection Method of Harmonic and Interharmonic Parameters | |
| CN104865474B (en) | A kind of method for monitoring low-frequency oscillation source in real time based on PMU data | |
| CN102496926B (en) | Method for judging and processing wind farm power prediction input data | |
| CN102435845A (en) | A Nuttall self-convolution window weighted Fourier transform method for harmonic energy measurement | |
| CN105652084B (en) | Wind power plant 3p CFVFs detection method based on EIAMD | |
| CN105004939A (en) | Composite electric energy quality disturbance signal quantitative analysis method | |
| CN103675447B (en) | A kind of high-precision real time-harmonic wave analysis method of electric railway | |
| CN104406558A (en) | Method for monitoring power transmission line sag | |
| CN107589299A (en) | Electric power signal synchronous phasor measuring method based on multi-frequency the measures model | |
| CN107064630A (en) | A kind of power system frequency measuring method and device | |
| CN105891625A (en) | Power system subsynchronous oscillation disturbance source identification method based on energy flow | |
| CN104407198A (en) | Method and system for detecting SAG signal in DVR device | |
| CN105223403A (en) | The wavelet packet extracting method of a kind of combining inverter net side resonance current information | |
| CN106300387A (en) | Given feedforward control method for dynamically inhibiting low-frequency oscillation of power grid based on SVG | |
| CN107017636A (en) | A kind of charging pile stress_responsive genes and filtering method | |
| Sun et al. | Parameter estimation of sub-/super-synchronous oscillation based on interpolated all-phase fast fourier transform with optimized window function | |
| CN104502674B (en) | A kind of three-phase voltage RMS to DC method and detection means | |
| CN103683230A (en) | Method and structure for achieving distance protection of power distribution network of power system | |
| CN105004920B (en) | Fourier's correction factor frequency measurement method | |
| Zaman et al. | Multimode synchronous resonance detection in converter dominated power system using synchro-waveforms | |
| CN103884910A (en) | Electric system phasor calculating method suitable for frequency deviation | |
| CN102707210A (en) | Lightning protection evaluation method and device for high-voltage overhead transmission line | |
| CN102269780A (en) | Method for rapid three-phase voltage phase-locked loop |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190115 Termination date: 20211231 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |