CN104034974A - Complex power quality disturbance signal identification method - Google Patents
Complex power quality disturbance signal identification method Download PDFInfo
- Publication number
- CN104034974A CN104034974A CN201410193729.5A CN201410193729A CN104034974A CN 104034974 A CN104034974 A CN 104034974A CN 201410193729 A CN201410193729 A CN 201410193729A CN 104034974 A CN104034974 A CN 104034974A
- Authority
- CN
- China
- Prior art keywords
- frequency
- point
- disturbance
- amplitude
- signal
- 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.)
- Pending
Links
Abstract
The invention relates to the technical field of power quality analysis and monitoring, and mainly relates to high-identification rate complex power quality disturbance signal identification method based on multiple features. Power quality signals acquired by monitoring equipment of a power quality monitoring point is used to serve as a to-be-identified disturbance signal type, that is, input of an automatic identification system, and the type of the disturbance signals is outputted through automatic identification. According to the method, multiple kinds of single disturbance existing in the power system can be identified, multiple kinds of complex disturbance can be precisely identified, an auxiliary decision is provided for management and governance on the power quality, and important practical significance is provided.
Description
Technical field
The invention belongs to power quality analysis and monitoring technical field, relate in particular to a kind of recognition methods of complex electric energy quality disturbance signal.
Background technology
Develop rapidly along with industrial technology, such as non-linear, impact and uncompensated load the coming into operation in electric system in a large number such as electric furnace arrangement for producing steel, electric railway, power electronic equipment, cause power distribution network to occur the power quality problem that a series of need such as voltage dip, voltage swell, voltage fluctuation, harmonic wave, Voltage notches, pulse and vibration transient interference attract great attention.The quality of power supply is polluted and is easily caused system loss to increase; measuring instrument measuring error increases; protective relaying device misoperation; the compensation condenser life-span significantly reduces; computer corruption or loss of data; predictable undervoltage tripping etc. not, the economic loss of bringing to sensitive load user thus increases year by year.Therefore, power supply department and power consumer grow with each passing day to the degree of attentiveness of the quality of power supply, and power quality problem has become electric system problem demanding prompt solution.
Power quality problem can be divided into stationary power quality problem and transient power quality problem by generation and duration.Along with the quality of power supply strengthens with people the deepening continuously of quality of power supply research gradually on the impact of national economy, people have been not only every Index For Steady-states such as voltage, frequency and harmonic wave to the focus of electrical energy power quality disturbance, also comprise various transient state disturbances.The research of stationary power quality disturbance deeply, also has strict standard both at home and abroad, and various inspective regulation methods are also comparatively ripe.And a large amount of inputs of new type microprocessor equipment and various power electronic equipments make transient power quality problem constantly outstanding.Although the coverage of transient state disturbance is less than stable state disturbance, its consequence causing can not look down upon.To the solution of transient state disturbance, need to take to the fully realizing as prerequisite of transient power quality disturbance, the detection and Identification method tool of therefore studying electrical energy power quality disturbance, particularly transient power quality disturbance is of great significance.
In addition, situation more complicated during due to actual electric network generation disturbance, be all not that single disturbance occurs, also there is a strong possibility is the situation that multiple disturbance mixes, and the disturbance producing during as system generation singlephase earth fault may be the mixing of voltage dip and vibration transient state.Because complex electric energy quality is that multiple single disturbance interaction produces, in the process of feature extraction, can even lose efficacy because influencing each other of single disturbance causes the aliasing of feature, its identification problem is identified to the many of complexity than single disturbance.Although there are at present some complex electric energy quality disturbance recognition methodss, but these methods are also lower for the discrimination of many disturbing signals, therefore inventing a kind of complex electric energy quality disturbance recognition methods with higher discrimination has important practical significance.
Summary of the invention
The present invention is directed to the deficiency of existing electrical energy power quality disturbance recognition methods, a kind of recognition methods of complex electric energy quality disturbance signal is provided, described recognition methods is carried out based on many characteristic quantities.
A recognition methods for complex electric energy quality disturbance signal, concrete steps are as follows:
S1, carry out the collection of Power Quality Disturbance: adopt voltage transformer (VT) to gather the Power Quality Disturbance of electric energy quality monitoring point in electrical network, obtain thus the voltage signal U that monitoring point contains disturbance;
The proper vector of S2, extraction Power Quality Disturbance, comprising:
S21, adopt described in overall empirical mode decomposition method (Ensemble Empirical Mode Decomposition, EEMD) treatment S 1 containing the voltage signal U of disturbance, extract effective modal characteristics amount M that EEMD decomposes
1, M
2, wherein, M
1for EEMD decomposes the instantaneous amplitude maximal value characteristic quantity of first mode obtaining, M
2for EEMD decomposes the amplitude sum after first mode each point delivery obtaining;
Disturbance voltage signal U described in S22, employing Dynamic Measurement method treatment S 1, extracts signal frequency characteristic quantity N
f, N
1, N
2, N
h, wherein, N
ffor characterizing disturbing signal, be simple signal or the characteristic quantity of multiple-frequency signal, N
1for characterizing the characteristic quantity of harmonic frequency point, N
2for characterizing, whether contain integral multiple fundamental frequency point characteristic quantity, N
hfor characterizing frequency spectrum high band, whether contain the characteristic quantity of main frequency point, specific as follows:
S221, thereby disturbance voltage signal U described in S1 is carried out to the amplitude spectrum A (f) that Fast Fourier Transform (FFT) obtains Fourier transform;
S222, employing Dynamic Measurement method are found out main frequency point and in conjunction with extreme point envelope, from amplitude spectrum A (f), are extracted feature N
f, N
1, N
2, N
h, detailed process is as follows:
S2221, establish X
1and X
nfor two on curve f different points, on curve, the part between these 2 is denoted as path P (X
1, X
n), i.e. P (X
1, X
n)=(X
1, X
2..., X
n), wherein, N is non-vanishing natural number;
S2222, path P (X
1, X
n) difference in height of the upper highs and lows Dynamic Measurement that is path, that is, and D
yn[P (X
1, X
n)]=sup (| h
alt(X
i) |-| h
alt(X
j) |), wherein, X
i, X
jfor the upper X of curve f
1and X
nbetween point, and i, j ∈ [2, N-1], sup represents supremum, h
altrepresent height;
S2223, establish X
mfor a maximal point of curve f, on curve, exist and compare X
mduring higher maximal point, some X
mdynamic Measurement equal an X
mlead to path Dynamic Measurement minimum in all paths with its co-altitude point, i.e. D
yn[X
m]={ inf{D
yn[P (X
m, X
n)], wherein, h
alt(X
m)=h
alt(X
n), inf represents infimum, P (X
m, X
n) the upper point of expression curve f X
mwith an X
nbetween part;
S2224, extreme point largest enveloping is asked for to the Dynamic Measurement spectrum A (w) that obtains original signal after Dynamic Measurement, the oscillation frequency of vibration disturbance is higher, in Dynamic Measurement spectrum, is distributed in high band, if when finding main frequency point extreme point X
mdynamic Measurement D
yn[X
m] meet
extreme point X
mcorresponding frequency is main frequency point, and frequency analysis scope is that [a, b] is the high band of A (w), when high band exists main frequency point, N
h=1, when high band does not exist main frequency point, N
h=0, wherein, T
hrfor the threshold value of setting, a is the Frequency point on A (w), and b is the Frequency point on A (w), a < b, X
maxit is maximum extreme point;
S2225, while containing harmonic components in disturbing signal, the corresponding frequency of harmonic components will be read present in spectrum dynamically surveying, and in the Dynamic Measurement spectrum of disturbing signal, exists and meets
main frequency point place frequency values while being harmonic frequency and the described main frequency point respective frequencies odd-multiple that is fundamental frequency, work as N
1=1 there is harmonic components, works as N
1=0 there is not harmonic components;
In S2226, due to voltage spikes, contain a large amount of integer harmonics compositions, when Dynamic Measurement method is asked for main extreme point, as extreme point X
mdynamic Measurement D
yn[X
m] meet
extreme point X
mcorresponding frequency is main frequency point, because first-harmonic composition, harmonic components and the part corresponding frequency of transient state of vibrating is also main frequency point in Dynamic Measurement spectrum, therefore the impact that remove these main frequency points when main frequency point is counted, if disregard first-harmonic composition, harmonic components and part, vibrate that to still have a large amount of main frequency points place frequency in the situation of impact of the corresponding main frequency point of transient state be integral multiple fundamental frequency, work as N
2=1 there is integral multiple fundamental frequency point, works as N
2=0 there is not integral multiple fundamental frequency point;
S2227, on the basis of S2224, S2225 and S2226, determine characteristic quantity N
f, when simultaneously, meet N
1=0, N
2=0, N
h=0, use N
f=0 represents in the Dynamic Measurement spectrum of disturbing signal only containing the corresponding main frequency point of first-harmonic composition, otherwise uses N
f=1 represents;
Described in S23, employing S conversion process S1, disturbance voltage signal U obtains mould time-frequency matrix, extracts S transform characteristics amount S from described mould time-frequency matrix
m, S
av, S
min, S
max, S
std, wherein, S
mfor S transformation matrix HFS maximum amplitude characteristic quantity, S
avfor the average amount of the corresponding fundamental frequency amplitude of S transformation matrix, S
minfor the corresponding fundamental frequency amplitude of S transformation matrix minimal characteristic amount, S
maxfor the maximum characteristic quantity of the corresponding fundamental frequency amplitude of S transformation matrix, S
stdfor the corresponding fundamental frequency amplitude of S transformation matrix standard deviation characteristic quantity, specific as follows:
S231, the definition S conversion time dependent curve J of amplitude (l)=S corresponding to mould time-frequency matrix fundamental component place
j(l, f
b), S conversion mould time-frequency matrix fundamental frequency amplitude characteristics of mean is
wherein, l represents sampled point, and L is total sampling number, f
brepresent basic frequency, S
avthe amplitude situation of change of reaction signal fundamental component;
S232, voltage swell, fall and interrupt presenting temporarily rising or the reduction of amplitude, these disturbances are all fundamental frequency disturbance, and the signal amplitude reacting condition that it causes is at the fundamental frequency part of S conversion mould time-frequency matrix, fundamental frequency amplitude minimum value characteristic quantity S
minwith maximal value characteristic quantity S
maxcan characterize fundamental frequency amplitude intensity of variation, thereby whether reaction signal composition can contain voltage swell in assistant analysis voltage signal, will and interrupt temporarily;
S233, when fundamental frequency amplitude changes by fundamental frequency amplitude standard deviation characteristic quantity S
minwith maximal value characteristic quantity S
maxcan react the situation of change of fundamental frequency amplitude, prevent from having noise spot in fundamental curve and the minimum value characteristic quantity S that causes
minwith maximal value characteristic quantity S
maxmis-classification to signal;
S3, according to the feature of Power Quality Disturbance described in S2, carry out Power Quality Disturbance classification, specific as follows:
S31, build the sorter software module of rule-based base " IF-THEN ";
S32, by the proper vector T input sorter of Power Quality Disturbance described in S2, automatically identify the type of 24 kinds of Power Quality Disturbances, wherein, T=[M
1, M
2, N
f, N
1, N
2, N
h, S
m, S
av, S
min, S
max, S
std], 24 kinds of Power Quality Disturbances comprise 8 kinds of single electrical energy power quality disturbances and 16 kinds of complex electric energy quality disturbances, and described single electrical energy power quality disturbance comprises due to voltage spikes R
1, pulse transient state R
2, voltage interruption R
3, voltage dip R
4, voltage swell R
5, vibration transient state R
6, harmonic wave R
7with voltage fluctuation R
8, described complex electric energy quality disturbance is R
2aMP.AMp.Amp R
5, R
2aMP.AMp.Amp R
4, R
2aMP.AMp.Amp R
3, R
2aMP.AMp.Amp R
8, R
3aMP.AMp.Amp R
8, R
4aMP.AMp.Amp R
8, R
5aMP.AMp.Amp R
8, R
3aMP.AMp.Amp R
6, R
4aMP.AMp.Amp R
6, R
5aMP.AMp.Amp R
6, R
3aMP.AMp.Amp R
7, R
4aMP.AMp.Amp R
7, R
5aMP.AMp.Amp R
7, R
7aMP.AMp.Amp R
2, R
7aMP.AMp.Amp R
6, R
7aMP.AMp.Amp R
8.
Further, described in S1, electric energy quality monitoring point is arranged on generating plant bus, region transformer station, important branch road and key user's access point, and wherein important branch road and the definition of key user's access point are referring to < < design of civil buildings standard > >.
Further, described in S222 Dynamic Measurement method in conjunction with extreme point envelope to ask for process as follows: establish | F (m) | be monitoring point containing the mould of the discrete Fourier transform (DFT) result of the voltage signal U of disturbance, its total L maximum point, is designated as
if
between Frequency point number be X
i+1,
with
between X
i+1the maximum value envelope of individual Frequency point is
the maximum value envelope of maximum point is itself,
wherein, i=1,2,3 ..., L+1, j=1,2,3..., X
i+1.
The invention has the beneficial effects as follows:
The present invention has stronger Classification of Power Quality Disturbances ability, can be from identifying accurately 8 kinds of single disturbance types and 16 kinds of dual disturbance types containing the voltage signal of disturbance, not only can identify the multiple single disturbance existing in electric system, can also identify accurately multiple compound disturbance, and then for management and the improvement of the quality of power supply provides aid decision making, have important practical significance.The present invention adopts the sorter software module of rule-based base " IF-THEN " form, sorter is simple in structure, recognition correct rate is high, and under 40dB noise, the correct recognition rata of the method is up to 98.875%, and under higher 25dB noise, the correct recognition rata of the method is still up to 90.3%.In addition the present invention adopts the method that EEMD conversion, Dynamic Measurement method and S conversion combine, comprehensive multi-level extraction voltage disturbance signal characteristic information, the feature aliasing or the Problem of Failure that in compound disturbance, exist have been taken into full account, therefore, the present invention can accurately identify single disturbance and compound disturbance, and total accuracy is up to more than 95%.
Accompanying drawing explanation
Fig. 1 is that signal characteristic quantity of the present invention extracts process flow diagram.
Fig. 2 is the sorter software module of Power Quality Disturbance automatic recognition system of the present invention.
In Fig. 2, number concrete meaning as follows:
1.-N2=0, Nf=0; 2.-Nf=1; N2=1; 3.-other; 4.-other; 5.-M1>0.24, M2>8.1; 6.-Nh=1, N1=0; 7.-N1=1; 8.-Smax>0.51; 9.-other; 10.-Sav>0.514;
-Sav ∈ [0.495,0.514];
-Sav<0.495;
-Sav ∈ [0.495,0.55];
-Sav<0.495;
-Sav>0.505;
sav ∈ [0.39,0.495];
-Sav<0.39;
-Sav>0.51;
-Sav ∈ [0.495,0.51];
-Smin<0.05;
-Smin>=0.05;
-Sav<0.49;
-Sav ∈ [0.49,0.512];
-Sav>0.512;
-Smin>0.478;
-Smin≤0.478;
-Smin≤0.055;
-Smin>0.055;
-Sstd>=0.1783;
-Sstd<0.1783;
-Smin<0.495;
-Smin>=0.495;
-Sstd≤0.193;
-Sstd>=0.193;
smin<0.49;
-Smin>=0.49;
-Sav>0.501;
-Sh>=0.13;
– Sh<0.13.
Embodiment
Below in conjunction with embodiment and accompanying drawing, describe technical scheme of the present invention in detail.
In order to realize the automatic identification of complex electric energy quality disturbance signal, shown in the process flow diagram in 1, need take following concrete steps with reference to the accompanying drawings:
Step 1, Power Quality Disturbance collection:
Because the sampled signal classification of practical power systems is single, be also difficult for obtaining great amount of samples, can not embody the diversity of Power Quality Disturbance completely.Therefore use the above-mentioned 24 kinds of Power Quality Disturbances of the random generation of MATLAB software, every kind of disturbing signal is random generates 200 samples, and the superpose white Gaussian noise of 40dB of each sample, and sampling rate is 3.2kHz, sampling period is 20 cycles, and signal length is 1280 points.
Step 2, Power Quality Disturbance feature extraction:
Adopt the disturbance voltage signal U collecting in overall empirical mode decomposition method treatment step 1, the first mode of decomposition result is processed and obtained feature M
1, M
2.The disturbance voltage signal U collecting in step 1 is carried out to fast fourier transform (FFT) thus obtain the amplitude spectrum A (f) of Fourier transform, then adopt Dynamic Measurement method in conjunction with extreme point envelope extraction feature N
f, N
1, N
2, N
h.Adopt the disturbance voltage signal U collecting in S transform method treatment step 1 to obtain mould time-frequency matrix, from mould time-frequency matrix, extract feature S
m, S
av, S
min, S
max, S
std.
Such three kinds of methods 11 characteristic quantities of common extraction that combine, feature extraction idiographic flow is shown in a kind of feature extraction process flow diagram of accompanying drawing, feature extraction is described below:
1, the feature M that overall empirical mode decomposition (EEMD) extracts
1, M
2specific as follows:
(1) EEMD decomposes the instantaneous amplitude maximal value characteristic quantity M of first mode (IMF) obtaining
1, what comprise due to first mode is the highest composition of original signal medium frequency, if in signal containing pulse composition; the noise contribution that first mode is signal, the maximum amplitude of maximum amplitude when containing pulse.Whether this feature is used for describing disturbing signal is whether to contain pulse composition in pulse signal or compound disturbing signal, and its threshold value is taken as 0.24.
(2) EEMD decomposes the amplitude sum characteristic quantity M after first mode (IMF) each point delivery obtaining
2.The frequency of vibration transient state is higher, and while containing vibration transient state composition in signal, the oscillationg component of high frequency also can decompose first mode, because its duration is much larger than pulse transient state, so the characteristic quantity M of oscillationg component
2larger.This feature is as the auxiliary criterion of vibration transient state and pulse transient state, and its assistant criteria as pulse transient state is that threshold value gets 8.2, as the assistant criteria of vibration transient state, is that its threshold value gets 13.
2, the feature N that Dynamic Measurement method is extracted
f, N
1, N
2, N
hspecifically describe as follows:
(1) whether frequency spectrum high band contains the characteristic quantity N of main frequency point
h:
Extreme point largest enveloping is asked for to the Dynamic Measurement spectrum A (w) that obtains original signal after Dynamic Measurement, the oscillation frequency of vibration disturbance is higher, in Dynamic Measurement spectrum, is distributed in high band.When finding main frequency point, get T
hr=3%, if extreme point x
mdynamic Measurement D
yn[x
m] meet
extreme point x
mcorresponding frequency is main frequency point.Frequency analysis scope is 500Hz-1600Hz.If high band exists main frequency to put, N
h=1, otherwise N
h=0.
(2) characterize harmonic frequency point characteristic quantity N
1:
While containing harmonic components in disturbing signal, the corresponding frequency of harmonic components will be read to present in spectrum dynamic survey, when setting T
hrin the time of=3%, in the Dynamic Measurement spectrum of 55Hz-500Hz frequency band disturbing signal, exist and meet
main frequency point and these main frequency point respective frequencies odd-multiple that is fundamental frequency, work as N
1=1 there is harmonic components, works as N
1=0 there is not harmonic components.
(3) whether contain integral multiple fundamental frequency point feature N
2:
In due to voltage spikes, contain a large amount of integer harmonics compositions, when Dynamic Measurement method is asked for main extreme point, set T
hrif extreme point x in the time of=3%
mdynamic Measurement D
yn[x
m] meet
extreme point x
mcorresponding frequency is main frequency point, now for avoiding the frequency analysis scope that affects of fundamental frequency, is decided to be 60Hz-1600Hz, removes and still have another N of a large amount of qualified Frequency points after the main frequency point that harmonic wave and oscillationg component are corresponding in these main frequency points
2=1, otherwise N
2=0.This feature is used for characterizing in disturbing signal, whether to contain spike composition.
(4) characterizing disturbing signal is the characteristic quantity N of single-frequency or multiple-frequency signal
f:
Extracting characteristic quantity N
1, N
2, N
hbasis on determine characteristic quantity N
f, work as N
1, N
2, N
hfeature be 0, i.e. N during only containing the corresponding main frequency point of first-harmonic composition in the Dynamic Measurement of disturbing signal spectrum
f=0, otherwise N
f=1.
3, the characteristic quantity S that S conversion is extracted
m, S
av, S
min, S
maxand S
std, specifically describe as follows:
(1) S transformation matrix HFS maximum amplitude characteristic quantity S
m:
S conversion mould time-frequency matrix is the three-dimensional matrice of a sign amplitude, time, frequency relation, definition expression formula A
s=A (t, w), gets w > 50000 and obtains an A
ssubmatrix A '
sbe the HFS of mould time-frequency matrix, its maximum amplitude is S
m.S
mthe size of reaction disturbing signal radio-frequency component, is taken as 0.13 as its threshold value of assistant criteria of vibration transient state.
(2) the average amount S of the corresponding fundamental frequency amplitude of S transformation matrix
av:
S conversion mould time-frequency matrix fundamental frequency amplitude characteristics of mean is
f
brepresent basic frequency 50Hz, L is total sampling number 1280, S
avreacted the amplitude situation of change of signal fundamental component.
(3) the corresponding fundamental frequency amplitude of S transformation matrix minimum value characteristic quantity S
minwith maximal value characteristic quantity S
max:
Voltage swell, fall and interrupt presenting temporarily rising or the reduction of amplitude, these disturbances are all fundamental frequency disturbance, and the signal amplitude reacting condition that it causes is in the fundamental frequency part of S conversion mould time-frequency matrix.Fundamental frequency amplitude minimum value characteristic quantity S
minwith maximal value characteristic quantity S
maxcan characterize fundamental frequency amplitude intensity of variation, thereby whether reaction signal composition can contain voltage swell in assistant analysis voltage signal, will and interrupt temporarily.
(4) the corresponding fundamental frequency amplitude of S transformation matrix standard deviation characteristic quantity S
std:
When fundamental frequency amplitude changes, by fundamental frequency amplitude standard deviation, can react the situation of change of fundamental frequency amplitude, prevent from having noise spot in fundamental curve and the minimum value characteristic quantity S that causes
minwith maximal value characteristic quantity S
maxmis-classification to signal.
Step 3, Power Quality Disturbance classification:
Build the sorter software module of rule-based base " IF-THEN " form, as shown in Fig. 2 dotted line frame.By 11 kinds of required proper vector T of Power Quality Disturbance characteristic quantity composition and classification that extract in step 2, T=[M
1m
2n
fn
1n
2n
hs
ms
acs
mins
maxs
std], the input using proper vector as described sorter, just can identify the disturbance type of described signal automatically.Recognition result is as shown in table 1:
Table 1
Constructed sorter can accurately be identified the type of Power Quality Disturbance under certain signal to noise ratio (S/N ratio) condition as can be seen from Table 1, not only can identify very accurately 8 kinds of single Power Quality Disturbances but also can identify 16 kinds of complex electric energy quality disturbance signals.Sorter is simple in structure directly perceived in addition, is easy to safeguard and expand in utilization process, when the new electrical energy power quality disturbance type of needs analysis, adds new feature rule after only needing to extract new characteristic quantity.
In sum, the invention provides a kind of high discrimination complex electric energy quality disturbance signal recognition method based on many characteristic quantities, can identify 8 kinds of single electrical energy power quality disturbance types and 16 kinds of complex electric energy quality disturbance types, overcome the shortcoming that existing recognition methods discrimination is low, can not effectively identify multiple complex electric energy quality disturbance type.And the sorter building in recognition methods is simple in structure directly perceived, is easy to expansion and safeguards.
Claims (3)
1. a recognition methods for complex electric energy quality disturbance signal, is characterized in that, comprises the steps:
S1, carry out the collection of Power Quality Disturbance: adopt voltage transformer (VT) to gather the Power Quality Disturbance of electric energy quality monitoring point in electrical network, obtain thus the voltage signal U that monitoring point contains disturbance;
The proper vector of S2, extraction Power Quality Disturbance, comprising:
S21, adopt described in overall empirical mode decomposition method treatment S 1 containing the voltage signal U of disturbance, extract effective modal characteristics amount M that EEMD decomposes
1, M
2, wherein, M
1for EEMD decomposes the instantaneous amplitude maximal value characteristic quantity of first mode obtaining, M
2for EEMD decomposes the amplitude sum after first mode each point delivery obtaining;
Disturbance voltage signal U described in S22, employing Dynamic Measurement method treatment S 1, extracts signal frequency characteristic quantity N
f, N
1, N
2, N
h, wherein, N
ffor characterizing disturbing signal, be simple signal or the characteristic quantity of multiple-frequency signal, N
1for characterizing the characteristic quantity of harmonic frequency point, N
2for characterizing, whether contain integral multiple fundamental frequency point characteristic quantity, N
hfor characterizing frequency spectrum high band, whether contain the characteristic quantity of main frequency point, specific as follows:
S221, thereby disturbance voltage signal U described in S1 is carried out to the amplitude spectrum A (f) that Fast Fourier Transform (FFT) obtains Fourier transform;
S222, employing Dynamic Measurement method are found out main frequency point and in conjunction with extreme point envelope, from amplitude spectrum A (f), are extracted feature N
f, N
1, N
2, N
h, detailed process is as follows:
S2221, establish X
1and X
nfor two on curve f different points, on curve, the part between these 2 is denoted as path P (X
1, X
n), i.e. P (X
1, X
n)=(X
1, X
2..., X
n), wherein, N is non-vanishing natural number;
S2222, path P (X
1, X
n) difference in height of the upper highs and lows Dynamic Measurement that is path, that is, and D
yn[P (X
1, X
n)]=sup (| h
alt(X
i) |-| h
alt(X
j) |), wherein, X
i, X
jfor the upper X of curve f
1and X
nbetween point, and i, j ∈ [2, N-1], sup represents supremum, h
altrepresent height;
S2223, establish X
mfor a maximal point of curve f, on curve, exist and compare X
mduring higher maximal point, some X
mdynamic Measurement equal an X
mlead to path Dynamic Measurement minimum in all paths with its co-altitude point, i.e. D
yn[X
m]={ inf{D
yn[P (X
m, X
n)], wherein, h
alt(X
m)=h
alt(X
n), inf represents infimum, P (X
m, X
n) the upper point of expression curve f X
mwith an X
nbetween part;
S2224, extreme point largest enveloping is asked for to the Dynamic Measurement spectrum A (w) that obtains original signal after Dynamic Measurement, the oscillation frequency of vibration disturbance is higher, in Dynamic Measurement spectrum, is distributed in high band, if when finding main frequency point extreme point X
mdynamic Measurement D
yn[X
m] meet
extreme point X
mcorresponding frequency is main frequency point, and frequency analysis scope is that [a, b] is the high band of A (w), when high band exists main frequency point, N
h=1, when high band does not exist main frequency point, N
h=0, wherein, T
hrfor the threshold value of setting, a is the Frequency point on A (w), and b is the Frequency point on A (w), a < b, X
maxit is maximum extreme point;
S2225, while containing harmonic components in disturbing signal, the corresponding frequency of harmonic components will be read present in spectrum dynamically surveying, and in the Dynamic Measurement spectrum of disturbing signal, exists and meets
main frequency point place frequency values while being harmonic frequency and the described main frequency point respective frequencies odd-multiple that is fundamental frequency, work as N
1=1 there is harmonic components, works as N
1=0 there is not harmonic components;
In S2226, due to voltage spikes, contain a large amount of integer harmonics compositions, when Dynamic Measurement method is asked for main extreme point, as extreme point X
mdynamic Measurement D
yn[X
m] meet
extreme point X
mcorresponding frequency is main frequency point, because first-harmonic composition, harmonic components and the part corresponding frequency of transient state of vibrating is also main frequency point in Dynamic Measurement spectrum, therefore the impact that remove these main frequency points when main frequency point is counted, if disregard first-harmonic composition, harmonic components and part, vibrate that to still have a large amount of main frequency points place frequency in the situation of impact of the corresponding main frequency point of transient state be integral multiple fundamental frequency, work as N
2=1 there is integral multiple fundamental frequency point, works as N
2=0 there is not integral multiple fundamental frequency point;
S2227, on the basis of S2224, S2225 and S2226, determine characteristic quantity N
f, when simultaneously, meet N
1=0, N
2=0, N
h=0, use N
f=0 represents in the Dynamic Measurement spectrum of disturbing signal only containing the corresponding main frequency point of first-harmonic composition, otherwise uses N
f=1 represents;
Described in S23, employing S conversion process S1, disturbance voltage signal U obtains mould time-frequency matrix, extracts S transform characteristics amount S from described mould time-frequency matrix
m, S
av, S
min, S
max, S
std, wherein, S
mfor S transformation matrix HFS maximum amplitude characteristic quantity, S
avfor the average amount of the corresponding fundamental frequency amplitude of S transformation matrix, S
minfor the corresponding fundamental frequency amplitude of S transformation matrix minimal characteristic amount, S
maxfor the maximum characteristic quantity of the corresponding fundamental frequency amplitude of S transformation matrix, S
stdfor the corresponding fundamental frequency amplitude of S transformation matrix standard deviation characteristic quantity, specific as follows:
S231, the definition S conversion time dependent curve J of amplitude (l)=S corresponding to mould time-frequency matrix fundamental component place
j(l, f
b), S conversion mould time-frequency matrix fundamental frequency amplitude characteristics of mean is
wherein, l represents sampled point, and L is total sampling number, f
brepresent basic frequency, S
avthe amplitude situation of change of reaction signal fundamental component;
S232, voltage swell, fall and interrupt presenting temporarily rising or the reduction of amplitude, these disturbances are all fundamental frequency disturbance, and the signal amplitude reacting condition that it causes is at the fundamental frequency part of S conversion mould time-frequency matrix, fundamental frequency amplitude minimum value characteristic quantity S
minwith maximal value characteristic quantity S
maxcan characterize fundamental frequency amplitude intensity of variation, thereby whether reaction signal composition can contain voltage swell in assistant analysis voltage signal, will and interrupt temporarily;
S233, when fundamental frequency amplitude changes by fundamental frequency amplitude standard deviation characteristic quantity S
minwith maximal value characteristic quantity S
maxcan react the situation of change of fundamental frequency amplitude, prevent from having noise spot in fundamental curve and the minimum value characteristic quantity S that causes
minwith maximal value characteristic quantity S
maxmis-classification to signal;
S3, according to the feature of Power Quality Disturbance described in S2, carry out Power Quality Disturbance classification, specific as follows:
S31, build the sorter software module of rule-based base " IF-THEN ";
S32, by the proper vector T input sorter of Power Quality Disturbance described in S2, automatically identify the type of 24 kinds of Power Quality Disturbances, wherein, T=[M
1, M
2, N
f, N
1, N
2, N
h, S
m, S
av, S
min, S
max, S
std], 24 kinds of Power Quality Disturbances comprise 8 kinds of single electrical energy power quality disturbances and 16 kinds of complex electric energy quality disturbances, and described single electrical energy power quality disturbance comprises due to voltage spikes R
1, pulse transient state R
2, voltage interruption R
3, voltage dip R
4, voltage swell R
5, vibration transient state R
6, harmonic wave R
7with voltage fluctuation R
8, described complex electric energy quality disturbance is R
2aMP.AMp.Amp R
5, R
2aMP.AMp.Amp R
4, R
2aMP.AMp.Amp R
3, R
2aMP.AMp.Amp R
8, R
3aMP.AMp.Amp R
8, R
4aMP.AMp.Amp R
8, R
5aMP.AMp.Amp R
8, R
3aMP.AMp.Amp R
6, R
4aMP.AMp.Amp R
6, R
5aMP.AMp.Amp R
6, R
3aMP.AMp.Amp R
7, R
4aMP.AMp.Amp R
7, R
5aMP.AMp.Amp R
7, R
7aMP.AMp.Amp R
2, R
7aMP.AMp.Amp R
6, R
7aMP.AMp.Amp R
8.
2. the recognition methods of a kind of complex electric energy quality disturbance signal according to claim 1, it is characterized in that: described in S1, electric energy quality monitoring point is arranged on generating plant bus, region transformer station, important branch road and key user's access point, wherein important branch road and the definition of key user's access point are referring to < < design of civil buildings standard > >.
3. the recognition methods of a kind of complex electric energy quality disturbance signal according to claim 1, it is characterized in that: described in S222 Dynamic Measurement method in conjunction with extreme point envelope to ask for process as follows: establish | F (m) | be that monitoring point is containing the mould of the discrete Fourier transform (DFT) result of the voltage signal U of disturbance, its total L maximum point, is designated as
if
with
between Frequency point number be X
i+1,
with
between X
i+1the maximum value envelope of individual Frequency point is
The maximum value envelope of maximum point is itself,
Wherein, i=1,2,3 ..., L+1, j=1,2,3..., X
i+1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410193729.5A CN104034974A (en) | 2014-05-09 | 2014-05-09 | Complex power quality disturbance signal identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410193729.5A CN104034974A (en) | 2014-05-09 | 2014-05-09 | Complex power quality disturbance signal identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104034974A true CN104034974A (en) | 2014-09-10 |
Family
ID=51465816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410193729.5A Pending CN104034974A (en) | 2014-05-09 | 2014-05-09 | Complex power quality disturbance signal identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104034974A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104316844A (en) * | 2014-11-14 | 2015-01-28 | 广东电网有限责任公司佛山供电局 | Power distribution network fault type recognition method and device |
| CN104391207A (en) * | 2014-12-09 | 2015-03-04 | 湖南工业大学 | Voltage sag detection method adopting fundamental frequency single vector S transformation |
| CN104407198A (en) * | 2014-11-28 | 2015-03-11 | 深圳供电局有限公司 | Method and system for detecting SAG signal in DVR device |
| CN105510687A (en) * | 2015-12-24 | 2016-04-20 | 合肥工业大学 | Empirical mode decomposition-based voltage anomaly characteristic identification method |
| CN107315111A (en) * | 2017-07-17 | 2017-11-03 | 浙江群力电气有限公司 | A kind of Power Quality Disturbance Classification Method and system |
| CN107832777A (en) * | 2017-10-12 | 2018-03-23 | 吉林化工学院 | A kind of electrical energy power quality disturbance recognition methods using the quick S-transformation feature extraction of time domain data compression multiresolution |
| CN109034277A (en) * | 2018-09-17 | 2018-12-18 | 吉林大学 | Power Quality Disturbance Classification Method based on multi-feature fusion and system |
| CN111639583A (en) * | 2020-05-26 | 2020-09-08 | 国网湖南省电力有限公司 | Method and system for identifying power quality disturbance of power grid |
| CN113128396A (en) * | 2021-04-16 | 2021-07-16 | 东南大学 | Electric energy quality composite disturbance classification method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100092028A1 (en) * | 2008-10-10 | 2010-04-15 | National Central University | Data Decomposition Method and Computer System Therefrom |
| CN102608441A (en) * | 2011-01-25 | 2012-07-25 | 华北电力科学研究院有限责任公司 | S conversion based SDPQD (Short Duration Power Quality Disturbance) signal identifying method, device and system |
| CN102831433A (en) * | 2012-06-06 | 2012-12-19 | 西南交通大学 | Method for classifying electric energy quality mixing disturbances based on multi-feature quantity of time-frequency domain |
| CN103323702A (en) * | 2013-05-28 | 2013-09-25 | 西南交通大学 | Composite power quality disturbing signal identifying method |
-
2014
- 2014-05-09 CN CN201410193729.5A patent/CN104034974A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100092028A1 (en) * | 2008-10-10 | 2010-04-15 | National Central University | Data Decomposition Method and Computer System Therefrom |
| CN102608441A (en) * | 2011-01-25 | 2012-07-25 | 华北电力科学研究院有限责任公司 | S conversion based SDPQD (Short Duration Power Quality Disturbance) signal identifying method, device and system |
| CN102831433A (en) * | 2012-06-06 | 2012-12-19 | 西南交通大学 | Method for classifying electric energy quality mixing disturbances based on multi-feature quantity of time-frequency domain |
| CN103323702A (en) * | 2013-05-28 | 2013-09-25 | 西南交通大学 | Composite power quality disturbing signal identifying method |
Non-Patent Citations (1)
| Title |
|---|
| 刘德建，等: "一种基于多特征量的复合电能质量扰动自动识别方法", 《成都大学学报（自然科学版）》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104316844A (en) * | 2014-11-14 | 2015-01-28 | 广东电网有限责任公司佛山供电局 | Power distribution network fault type recognition method and device |
| CN104316844B (en) * | 2014-11-14 | 2016-05-11 | 广东电网有限责任公司佛山供电局 | Distribution network failure kind identification method and device |
| CN104407198A (en) * | 2014-11-28 | 2015-03-11 | 深圳供电局有限公司 | Method and system for detecting SAG signal in DVR device |
| CN104391207A (en) * | 2014-12-09 | 2015-03-04 | 湖南工业大学 | Voltage sag detection method adopting fundamental frequency single vector S transformation |
| CN105510687A (en) * | 2015-12-24 | 2016-04-20 | 合肥工业大学 | Empirical mode decomposition-based voltage anomaly characteristic identification method |
| CN107315111A (en) * | 2017-07-17 | 2017-11-03 | 浙江群力电气有限公司 | A kind of Power Quality Disturbance Classification Method and system |
| CN107832777A (en) * | 2017-10-12 | 2018-03-23 | 吉林化工学院 | A kind of electrical energy power quality disturbance recognition methods using the quick S-transformation feature extraction of time domain data compression multiresolution |
| CN107832777B (en) * | 2017-10-12 | 2021-01-26 | 吉林化工学院 | Electric energy quality disturbance identification method adopting time domain compression multi-resolution rapid S-transform feature extraction |
| CN109034277A (en) * | 2018-09-17 | 2018-12-18 | 吉林大学 | Power Quality Disturbance Classification Method based on multi-feature fusion and system |
| CN111639583A (en) * | 2020-05-26 | 2020-09-08 | 国网湖南省电力有限公司 | Method and system for identifying power quality disturbance of power grid |
| CN113128396A (en) * | 2021-04-16 | 2021-07-16 | 东南大学 | Electric energy quality composite disturbance classification method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104034974A (en) | Complex power quality disturbance signal identification method | |
| US9142968B2 (en) | Systems and methods for islanding detection | |
| CN103323702B (en) | Complex electric energy quality disturbance signal recognition method | |
| Jiang et al. | A novel dominant mode estimation method for analyzing inter-area oscillation in China southern power grid | |
| CN101339208A (en) | Voltage quality monitoring and perturb automatic classification method based on analysis in time-domain | |
| US11245260B2 (en) | Automatic discovery of electrical supply network topology and phase | |
| CN110716097B (en) | Power supply network topology identification method and device | |
| CN103198184A (en) | Low-frequency oscillation character noise-like identification method in electric power system | |
| CN101944776A (en) | Low-frequency oscillation monitoring system of electrical network | |
| CN107342628B (en) | New energy access broadband stability multistage early warning method integrating multi-source information | |
| Prince et al. | Targeted approach to apply masking signal-based empirical mode decomposition for mode identification from dynamic power system wide area measurement signal data | |
| CN103424620A (en) | 10kV distribution network harmonic source identification method based on time sequence relevancy | |
| CN104459398A (en) | Electric energy quality complex disturbance recognition method for lowering noise through two-dimensional morphology | |
| CN111103498B (en) | Low-voltage transformer area line-change user relationship identification system and method | |
| CN102798758B (en) | Method and system for measuring series reactance rate of shunt capacitor bank | |
| Alhelou et al. | Wide area power systems stability, protection, and security | |
| Donato et al. | Review of NILM applications in smart grids: Power quality assessment and assisted independent living | |
| CN105550450B (en) | Electric energy quality interference source characteristic harmonic modeling method | |
| CN104714156A (en) | GIS partial discharge positioning method based on ultrahigh frequency discharge online monitoring data | |
| Elombo et al. | Residential load variability and diversity at different sampling time and aggregation scales | |
| Li et al. | Islanding detection for microgrid based on frequency tracking using extended Kalman filter algorithm | |
| Naik et al. | Analysis of power quality disturbances using wavelet packet transform | |
| Godoy et al. | Multiple signal processing techniques based power quality disturbance detection, classification, and diagnostic software | |
| Qaisar et al. | Effective Power Quality Disturbances Identification Based on Event‐Driven Processing and Machine Learning | |
| Qing et al. | Application of HHT for oscillation mode analysis in power system based on PMU |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140910 |
|
| RJ01 | Rejection of invention patent application after publication |