CN104810795B - A kind of modified excitation surge current discrimination method - Google Patents
A kind of modified excitation surge current discrimination method Download PDFInfo
- Publication number
- CN104810795B CN104810795B CN201510217672.2A CN201510217672A CN104810795B CN 104810795 B CN104810795 B CN 104810795B CN 201510217672 A CN201510217672 A CN 201510217672A CN 104810795 B CN104810795 B CN 104810795B
- Authority
- CN
- China
- Prior art keywords
- data window
- mrow
- point
- current
- variable data
- 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.)
- Active
Links
- 230000005284 excitation Effects 0.000 title claims abstract description 30
- 238000012850 discrimination method Methods 0.000 title claims abstract description 15
- 230000004224 protection Effects 0.000 claims abstract description 50
- 238000005070 sampling Methods 0.000 claims abstract description 23
- 230000000903 blocking effect Effects 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 20
- 238000010606 normalization Methods 0.000 claims description 9
- 230000004888 barrier function Effects 0.000 claims 1
- 238000005094 computer simulation Methods 0.000 description 3
- 230000002708 enhancing effect Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Abstract
Accelerate distinguished number the invention discloses a kind of modified excitation surge current discrimination method, including main variable data window distinguished number and auxiliary：Threshold value ± ε is set respectively under on a timeline, using threshold value and the extreme value of a certain section of sample rate current, the starting point and terminating point of variable data window is determined；And in the variable data window selected; according to sampling interval and data window length; fictionalize a standard sine curve backward forward with this extreme value, the normalized area index of difference area between standard of comparison sine curve and sample rate current, and then realize the locking and opening of protection.Inrush and fault current waveform can be distinguished to greatest extent, can be with fast and reliable Blocking Differential Protection when excitation surge current occurs for transformer；Blocking Differential Protection is not missed during failure internally；Differential protection can also be quickly opened in transformer belt failure idle-loaded switching-on；And with stronger anti-transformer saturability.
Description
Technical field
The present invention relates to a kind of modified excitation surge current discrimination method, belong to technical field of power systems.
Background technology
At present, existing power transformer is widely used in power plant as transformation of electrical energy and the critical elements of transmission
And transformer station.Differential protection is the main protection of transformer, ideally, and transformer differential protection is short in normal operation and outside
Lu Shi, the electric current for flowing into differential relay is zero, and protection device is reliably failure to actuate.But actually when transformer during no-load closing is produced
During excitation surge current, the size of excitation surge current is up to several times even more than ten times of transformer rated current；On the other hand, excitation is gushed
Stream flows only through the mains side of transformer, and load side does not have electric current because of open circuit, and excitation surge current will fully flow into longitudinal difference protection
Differential circuit, makes to generate a very big out-of-balance current in differential relay.So, differentiate excitation surge current and inside event
Hindering electric current turns into the key issue of enhancing protection reliability.
The content of the invention
Purpose：In order to overcome the deficiencies in the prior art, the present invention provides a kind of modified excitation surge current discriminating side
Method, by the determination of variable data window, this method can to a greater degree be distinguished and shoved compared with the existing discrimination method that shoves
Waveform and fault current waveform.
Technical scheme：In order to solve the above technical problems, the technical solution adopted by the present invention is：
Accelerate distinguished number including main variable data window distinguished number and auxiliary；
Main variable data window distinguished number：Threshold value ± ε is set respectively under on a timeline, sampled using threshold value and a certain section
The extreme value of electric current, extreme value is maximum value or minimum value, determines the starting point and terminating point of variable data window；And selecting
In variable data window, according to sampling interval and data window length, a standard sine curve is fictionalized backward forward with this extreme value,
Calculate the normalization index Z of difference area between standard sine curve and sample rate current；With the passage of sampled point, algorithm is formed
Multiple variable data windows, each data window can export one normalization index Z, algorithm in real time will normalization index Z with it is whole
Definite value ZsetIt is compared, and then realizes the locking and opening of protection；
Auxiliary accelerates distinguished number：Setting is after auxiliary accelerates distinguished number to start, if phase current does not have during a cycle
There is the output of Z values, then judge the phase current immediately to shove, this accelerates distinguished number for auxiliary, so auxiliary accelerates distinguished number
Most slowly also can the latch-up protection in a cycle.
Described modified excitation surge current discrimination method, it is characterised in that：Main variable data window distinguished number is specifically referred to：
Fictionalized first according to the extreme point in a certain section come a standard sine curve, fictionalize the standard sine come bent
Line can be divided into two sections：
First paragraph is extreme point in data window to the curved section Y between data window starting point1, it is formulated as：
Y1(k1)=Ilim*sin[π/2-2π(N1-k1)/N], k1=1,2 ... N1
Wherein k1To fictionalize the first paragraph standard sine curve Y come1Discrete point sequence number；IlimFor extreme value in data window
(maximum value or minimum value), N is a cycle total sampling number, N1For curved section Y1Interior total data points；
Second segment is extreme point in data window to the curved section Y between data window terminating point2, it is formulated as：
Y2(k2)=Ilim*sin[π/2-2πk2/ N], k2=1,2 ... N2
Wherein k2To fictionalize the second segment standard sine curve Y come2Discrete point sequence number, N2For curved section Y2Interior sum
Strong point number；
Then total standard sine curve can be defined as：
Y=[Y1 Y2]
Make S1For sample rate current curve and the area of time shaft, S2To fictionalize the standard sine curve come and time shaft
Area, then S1With S2It can be expressed as：
Wherein N3For Y1With Y2Data window total length altogether, j is N3The sampled point sequence number of actual current, k in data window
For Y1With Y2The discrete point sequence number of total standard sine curve y altogether, i (j) is j-th of sampled point in above-mentioned total data window
When sample rate current instantaneous value, y (k) be total standard sine curve (its length is exactly total data window length) in k-th of discrete point
When instantaneous value.
Make the sampled point sequence number of actual current in the data window corresponding with standard sine curve y discrete point sequence number, i.e. j
=k, then calculating difference area S between sample rate current and standard sine curve is：
Discriminant criterion is normalized, S is mademin=min (S1, S2)
Z=S/Smin
SminFor S1, S2Middle smaller, Z is the index that variable data window judges excitation surge current or fault current；ZsetFor Z
Setting valve；
With the passage of sampled point, algorithm forms multiple variable data windows, and each data window can export an index Z,
Algorithm is in real time by this index Z and certain certain value ZsetIt is compared, works as Z>During Zset, the current waveform in data window is sentenced
It Wei not shove；As Z≤Zset, the current waveform in data window is determined as fault current.
Wherein, the selection of variable data window is specifically referred to：First on a timeline under threshold value ± ε is set respectively, if protection
Sample rate f, sampling interval T=1/f, each cycle sampling number is N, is comprised the following steps：
1) determination of maximum value or minimum value：System is normally and before idle-loaded switching-on, and poor stream should be within threshold interval；It is on duty
When stream exceeds threshold interval, main variable data window distinguished number starts, and selection sample rate current passes sampled point during threshold interval
As the starting point of first extremum search, sampled point when selection sample rate current is again out the threshold interval is as first
The terminating point of extremum search；First positive pole is searched out in starting point to the data window of terminating point of first extremum search
Big value ImaxOr reverse minimum Imin, record the sampled value I corresponding to the extreme pointlimAnd sampled point sequence number n, similarly find
Second extreme point Ilim2；
2) determination of variable data window：With first extreme value IlimSearched for backward forward for basic point, provide search procedure forward
In passed using electric current from threshold region after first point of starting point as variable data window, the instantaneous value of starting point is I1,
Sampled point serial number a, search condition such as formula (1)；If any sampled point is not left in region, in search procedure forward with
First point of starting point I as variable data window after current crosses threshold region1, search condition such as formula (2) seeks data window
The algorithm of starting point is：
When
i(j+1)<=| ε | ＆i (j)>| ε | j=n-1, n-2, n-3 ... (1) or
When,
Then make a=j, I1=i (a)
In formula i be sample rate current instantaneous value, j be current sample point sequence number, it is determined that after data window starting point sampled point
Serial number a, the instantaneous value of data window starting point is I1, n is extreme value sampled point sequence number, i (n)=Ilim；
First point of terminating point as variable data window after being passed backward in search procedure using electric current from threshold region,
Its instantaneous value is I2, sampled point serial number b, search condition such as formula (3)；If any sampled point is not left in region, to
The terminating point I of variable data window is equally used as using first sampled point after current crosses threshold region in search procedure afterwards2, search
Rope condition such as formula (4), i.e.,
i(j-1)<=| ε | ＆i (j)>| ε | j=n+1, n+2, n+3 ... (3)
Or
When,
Make b=j, I2=i (b)；
In formula i be sample rate current instantaneous value, j be current sample point sequence number, it is determined that after data window terminating point sampled point
Serial number b, the instantaneous value of data window terminating point is I2, n is extreme value sampled point sequence number, i (n)=Ilim；
3) first extreme value I is calculatedlimAnd I1Between sampling number N1, include IlimAnd I1, fictionalizing standard just at 2 points
The first paragraph of chord curve, order
Y1(k1)=Ilim*sin[π/2-2π(N1-k1)/N], k1=1,2 ... N1
Calculate IlimAnd I2Between sampling number N2, not comprising IlimBut include I2Point, fictionalizes standard sine curve
Second segment, order
Y2(k2)=Ilim*sin[π/2-2πk2/ N], k2=1,2 ... N2
By sequence Y1With Y2Spliced, made
Y=[Y1 Y2]
4) I is calculated1, I2Between points N3, include I1And I2, make the sampled point sequence number and mark of actual current in the data window
Quasi sinusoid y discrete point sequence number is corresponding, i.e. j=k then calculates difference face between sample rate current and standard sine curve
Accumulating S is：
Described modified excitation surge current discrimination method, it is characterised in that：With the passage of data window, algorithm is continuously exported
The decision rule such as table 1 that Z values are exported under Z values, different situations：
The decision rule that Z values are exported under the different situations of table 1
In view of supporting with transformer differential protection starting current, threshold value ± ε is set, ε=0.18I is setN, INTo become
Depressor rated current.
Consider the reliability of protection, set Zset=0.55.
Beneficial effect：The modified excitation surge current discrimination method that the present invention is provided, including main variable data window distinguished number
And auxiliary accelerates distinguished number.This method main innovation point includes：1st, this method is ingenious sets threshold value using in zero crossings
Method, solve the problems, such as some inrush of oversize transformer be completely in the side of time shaft i.e. so-called not zero passage when it is variable
The On The Choice of data window, sets the threshold interval also to advise the selection of variable data window Algorithms of Selecting under different situations in addition
Then unitize.2nd, by the analysis to Current Waveform Characteristics under different situations, variable data window Algorithms of Selecting, the algorithm are designed
Variable data window can be made to include the interruption angle part shoved in excitation surge current, can so make inrush and fictionalize what is come
Area folded by standard sine waveform is tried one's best maximum；In case of a fault, make fault current waveform and fictionalize the standard sine come
Area folded by waveform is tried one's best minimum.3rd, the locking auxiliary that proposes to shove accelerates criterion, by the use of variable data window access time as gushing
One auxiliary of stream identification accelerates criterion, it is ensured that algorithm is most slow in the case of shoving also can the latch-up protection in a cycle.4th, lead to
The difference that variable data window successively exports Z values is crossed, with reference to split-phase locking principle, to transformer belt failure idle-loaded switching-on situation, energy
Enough not even lockings that quickly unblocks.5th, the algorithm has stronger anti-transformer saturability.
The validity and superiority of technical solution of the present invention are demonstrated by dynamic simulation test, in the turn-to-turn of transformer belt 1.7%
During short trouble idle-loaded switching-on, it is protected in 20ms after combined floodgate and opens differential protection, it is faster than secondary harmonic brake open and protection speed
23 times.And in idle-loaded switching-on transformer saturation, algorithm can still make the reliable locking of protection.
Brief description of the drawings
Fig. 1 is the searching figure of variable data window maximum value or minimum value；
Fig. 2 is the selection figure of the asymmetric variable data window that shoves；
Fig. 3 is the selection figure of variable data window of symmetrically shoving；
Fig. 4 is the selection figure of fault current variable data window；
Fig. 5 is the selection figure of tape jam idle-loaded switching-on variable data window；
Fig. 6 (a) is normal idle-loaded switching-on segregated phase differential current and Fig. 6 (b) Z values output figure；
Fig. 7 (a) is idle-loaded switching-on in 1.7% shorted-turn fault segregated phase differential current and Fig. 7 (b) Z values output figure；
Fig. 8 (a) is idle-loaded switching-on in internal ground fault segregated phase differential current and Fig. 8 (b) Z values output figure；
Fig. 9 (a) is Transformer internal ground fault segregated phase differential current and Fig. 9 (b) Z values output figure；
Segregated phase differential current and Figure 10 (b) Z values output figure when Figure 10 (a) is idle-loaded switching-on transformer saturation.
Embodiment
The present invention is further described with reference to specific embodiment.
As shown in Figures 1 to 10, a kind of modified excitation surge current discrimination method, including：
(1) main variable data window distinguished number：Threshold value ± ε is set respectively under on a timeline, threshold value and a certain section is utilized
The extreme value of sample rate current, extreme value is maximum value or minimum value, determines the starting point and terminating point of variable data window；And choosing
In the variable data window gone out, according to sampling interval and data window length, a standard sine is fictionalized backward forward with this extreme value
Curve, the normalized area index of difference area between standard of comparison sine curve and sample rate current, and then realize closing for protection
Lock and open.This method can distinguish inrush and fault current waveform to greatest extent, and occurring excitation in transformer gushes
, can be with fast and reliable Blocking Differential Protection during stream；Blocking Differential Protection is not missed during failure internally；Can also be in transformer belt event
When hindering idle-loaded switching-on, differential protection is quickly opened；And with stronger anti-transformer saturability.
In order to further lift the performance of excitation surge current discrimination method, the present invention also proposes that a kind of auxiliary accelerates to differentiate and calculated
Method, distinguished number is accelerated with reference to auxiliary, algorithm Blocking Differential Protection during a cycle after idle-loaded switching-on at the latest.
(2) auxiliary accelerates distinguished number：Due to shove and fault current different wave shape, to variable data window, window itself is long
Access time also can as differentiation of shoving factor, therefore can will choose the time used in variable data window as accelerating to shove locking
An assistant criteria；
In order to strengthen the reliability of inrush current distinguishing, above main variable data window distinguished number extract variable data window when meeting
As often as possible interval angle data are included, if electric current mutually shoves, interval angle is excessive, and variable data window is included
Sampling number certainly will increase, although so enhancing algorithm differentiates the ability shoved, and the time for choosing variable data window can be by
It is long in the more sampling number of wait, influence the output speed of algorithm.It is actual, when after auxiliary accelerates distinguished number to start
The situation of only excitation surge current is possible to make interval angle very big in section, variable data window access time is exceeded a cycle,
And cause there is no the output of Z values during a cycle；So setting herein is after auxiliary accelerates distinguished number to start, if during a cycle
Electric current mutually without the output of Z values, then immediately Cutoff current this mutually to shove, this accelerates distinguished number for auxiliary, so auxiliary adds
Fast distinguished number is most slow also can the latch-up protection in a cycle.
Therefore, with the passage of data window, algorithm continuously exports Z values, and various situations can be carried out in Taxonomic discussion, table 1
Provide the decision rule that Z values are exported under different situations：
The decision rule that Z values are exported under the different situations of table 1
The present invention combines main variable data window distinguished number and auxiliary accelerates two kinds of algorithms of distinguished number, can not only become
Quick and reliably Blocking Differential Protection during the normal idle-loaded switching-on of depressor, does not miss that locking is differential to be protected in power transformer interior fault
Shield, moreover it is possible to the quick open differential protection in transformer belt failure idle-loaded switching-on.
Embodiment 1：
A) the selection of variable data window：Variable data window selection process is combined into the identification algorithm that shoves to be introduced, first
Threshold value ± the ε of very little is set respectively under on a timeline, if protection sample rate f, sampling interval T=1/f, each cycle sampling number
For N；Its step is：
1. the determination of maximum value or minimum value：System is normally and before idle-loaded switching-on, and poor stream should be within threshold interval.It is on duty
When stream exceeds threshold interval, main variable data window distinguished number starts, as shown in figure 1, selection sample rate current passes threshold interval
When sampled point as the starting point of first extremum search, sampled point when selection sample rate current is again out the threshold interval
It is used as the terminating point of first extremum search；First positive maximum or reverse minimum are searched out in the segment data window,
Record the sampled value I corresponding to the extreme pointlim(the I in Fig. 1max) and sampled point sequence number n, it can similarly be sought according to above-mentioned rule
Find second extreme point Ilim2(the I in Fig. 1min)；
2. (this sentences the maximum I that first variable data window of Fig. 1 is searched out for the determination of variable data windowmaxExemplified by carry out
Illustrate)：With maximum ImaxSearched for backward forward for basic point, provide to pass from threshold region with electric current in search procedure forward
First point of starting point as variable data window afterwards, its instantaneous value is I1, sampled point serial number a, search condition such as formula (1)；
If any sampled point is not left in region, in search procedure forward using the first point after current crosses threshold region as
The starting point I of variable data window1, search condition such as formula (2), the algorithm for seeking data window starting point is：
When
i(j+1)<=| ε | ＆i (j)>| ε | j=n-1, n-2, n-3 ... (1)
Or
When,
Then make a=j, I1=i (a)
In formula i be sample rate current instantaneous value, j be current sample point sequence number, it is determined that after data window starting point sampled point
Serial number a, the instantaneous value of data window starting point is I1, n is extreme value sampled point sequence number, i (n)=Imax；
First point of terminating point as variable data window after being passed backward in search procedure using electric current from threshold region,
Its instantaneous value is I2, sampled point serial number b, search condition such as formula (3)；If any sampled point is not left in region, to
The terminating point I of variable data window is equally used as using first sampled point after current crosses threshold region in search procedure afterwards2, search
Rope condition such as formula (4), i.e.,
i(j-1)<=| ε | ＆i (j)>| ε | j=n+1, n+2, n+3 ... (3)
Or
When,
Make b=j, I2=i (b)；
In formula i be sample rate current instantaneous value, j be current sample point sequence number, it is determined that after data window terminating point sampled point
Serial number b, the instantaneous value of data window terminating point is I2, n is extreme value sampled point sequence number, i (n)=Imax；
3. calculate first extreme value ImaxAnd I1Between sampling number N1, include ImaxAnd I1, fictionalizing standard just at 2 points
The first paragraph of chord curve, order
Y1(k1)=Imax*sin[π/2-2π(N1-k1)/N], k1=1,2 ... N1
Calculate ImaxAnd I2Between sampling number N2, not comprising ImaxBut include I2Point, fictionalizes standard sine curve
Second segment, order
Y2(k2)=Imax*sin[π/2-2πk2/ N], k2=1,2 ... N2
By sequence Y1With Y2Spliced, made
Y=[Y1 Y2]
4. calculate I1, I2Between points N3, include I1And I2, as shown in Fig. 2 making the sampling of actual current in the data window
Point sequence number is corresponding with standard sine curve y discrete point sequence number, i.e. j=k then calculates sample rate current and standard sine curve
Between difference area S be：
As shown in Figures 2 and 3, it is asymmetric shove and symmetrically shove in the case of, by algorithm above choose data window, due to bag
Interruption angle part is contained, it is very big that each variable data window hatched area S can become, and algorithm continuously exports larger normalizing
Change area index Z.And for internal fault, as shown in figure 4, each variable data window hatched area S is smaller, algorithm
Continuously export less normalized area index Z.
For tape jam idle-loaded switching-on, from the point of view of the partial waveform of electric current, although in transformer saturation section, excitation surge current compared with
Greatly, the feature of fault current is masked, but when transformer exits saturation into unsaturation region, performance is out of order by electric current
The feature of electric current.And above-mentioned variable data window selection algorithm has the ability for choosing unsaturation section electric current automatically.Such as Fig. 5 institutes
Show, variable data window selection algorithm can alternately select saturation section electric current and unsaturation section in the period 1 after tape jam idle-loaded switching-on
Electric current, then it is continuous alternately to export larger and less normalized area index Z, therefore algorithm has the ability to know within a short period of time completely
Do not go out tape jam combined floodgate situation, allow protection to open as early as possible.
B) normalized area index and criterion are adjusted
Because in actual state, protection sample rate is possible to inconsistent and can cause number by above-mentioned data window selection algorithm
Different in size according to window, the size of different situations electric current is also different in addition, then weighs hatched area size under different situations
Index will be inconsistent.In order to solve this problem, discriminant criterion need to be normalized, if
Wherein N3For Y1With Y2Data window total length altogether, j is N3The sampled point sequence number of actual current, k in data window
For Y1With Y2The discrete point sequence number of total standard sine curve y altogether, i (j) is j-th of sampled point in above-mentioned total data window
When sample rate current instantaneous value, y (k) be total standard sine curve (its length is exactly total data window length) in k-th of discrete point
When instantaneous value.
Make the sampled point sequence number of actual current in the data window corresponding with standard sine curve y discrete point sequence number, i.e. j
=k, then calculating difference area S between sample rate current and standard sine curve is：
Discriminant criterion is normalized, S is mademin=min (S1, S2)
Z=S/Smin
SminFor S1, S2Middle smaller, Z is the index that variable data window judges excitation surge current or fault current；ZsetFor Z
Setting valve；
With the passage of sampled point, algorithm forms multiple variable data windows, and each data window can export a normalization
Area index Z, algorithm is in real time by this normalized area index Z and certain certain value ZsetIt is compared, works as Z>, will during Zset
Current waveform in data window is determined as shoving；As Z≤Zset, the current waveform in data window is determined as fault current.
The criterion for differentiating fault current and excitation surge current is：
C) auxiliary accelerates distinguished number
In order to strengthen the reliability of inrush current distinguishing, algorithm as far as possible many can will be interrupted angular data when extracting variable data window
(such as Fig. 2, Fig. 3) is included, interval angle is excessive if certain mutually shoves, and the sampling number that variable data window is included certainly will increase
Plus, although so enhancing the discriminating power shoved, the time for choosing variable data window can be due to waiting more sampling number
And it is long, influence the output speed of algorithm.Actual, after algorithm startup, the situation of this period interior only excitation surge current just has can
Interval angle can be made very big, variable data window access time is exceeded a cycle, and cause there is no the output of Z values during a cycle.
So setting herein is after algorithm starts, mutually this is then judged immediately mutually to shove, this is during a cycle without the output of Z values
Auxiliary accelerates distinguished number, also can the latch-up protection in a cycle so algorithm is most slow.
Preferably, it is contemplated that supporting with transformer differential protection starting current, set threshold value ± ε, wherein ε=
0.18IN, INFor transformer rated current.Consider the reliability of protection, set Zset=0.55.
Embodiment 2：
The present invention algorithm is verified with the mode of dynamic simulation test, dynamic simulation test demonstrate the program validity and
Superiority, in 1.7% shorted-turn fault idle-loaded switching-on of transformer belt, is protected in 20ms after combined floodgate and opens differential protection.Under
Face is action characteristic analysis of the present invention under each operating mode：
1st, the normal idle-loaded switching-on of transformer
Fig. 6 (a) is segregated phase differential current in the case of the normal idle-loaded switching-on of transformer, and three-phase is A phases, B phases and C phases, segregated phase differential current
IA、IBAnd ICRepresented respectively with solid line, dotted line and chain-dotted line.From Fig. 6 (a), idle-loaded switching-on transformer, three-phase in 0.02s
I in difference streamAAnd IBShoved for one side, there is obvious interval angle, and interval angle is larger；ICSymmetrically to shove, interval angle is smaller；IBBy
In the influence not zero crossing of DC component, therefore the selection of variable data window is realized with the method for given threshold herein.
Three-phase Z values are respectively ZA、ZBAnd ZC；From Fig. 6 (b), due to IAAnd IBShoved to be asymmetric, ICSymmetrically to gush
Stream, therefore in 5 cycles, according to variable data window Algorithms of Selecting, ZAAnd ZBOutput points are ZCExport the half of points.C phases exist
6ms exports first Z value more than Z after idle-loaded switching-onset, algorithmically main criterion needs the 24ms after idle-loaded switching-on respectively for A phases and B phases
Blocking Differential Protection, but according to auxiliary accelerate distinguished number, then respectively can after idle-loaded switching-on 20ms Blocking Differential Protections.
2nd, idle-loaded switching-on is in the slight shorted-turn fault of transformer
Fig. 7 (a) is the segregated phase differential current of the shorted-turn fault of idle-loaded switching-on transformer star side A phases 1.7%.Because A phases are sent out
Raw small turn ratio short trouble, fault current and presence, and fault current is slighter simultaneously of shoving, so difference stream is in transformer saturation
Section is mainly shown as form of shoving, and in transformer unsaturation section, then it is poor to flow the feature for showing as fault current completely.Using can
Changing data window distinguished number extracts the variable data window with shove feature and fault signature respectively, then three-phase Z values are exported such as
Shown in Fig. 7 (b).
From Fig. 7 (b), the form fluctuated above and below A phase Z values output presentation, this is due to that A difference streams are continuously replaced
It is divided into the data window for the feature that shows to shove and shows the data window of fault signature.In the data window of feature that shows to shove
Interior, the output of Z values is larger；And showing in the data window of fault signature, difference stream waveform and standard sine wave are closely similar, therefore
The output of Z values is smaller.First Z values point that A phases are exported after idle-loaded switching-on during 13ms, is judged to shove, A phases are closed rapidly
Lock；But in 20ms, second Z values point is exported, less than Zset, fault current is identified as, according to the decision rule of table 1, now A
The rapid open and protection of phase, due to using split-phase locking principle, differential protection can be acted rapidly.Under identical test current conditions,
Secondary harmonic brake scheme 460ms or so open and protections after idle-loaded switching-on.
3rd, idle-loaded switching-on is in inside transformer earth fault
Fig. 8 (a) is idle-loaded switching-on transformer star side A phase internal ground faults.Because more serious failure occurs for A phases, then
IAAnd ICPoor stream is mainly shown as fault current feature, and IBStill there is larger excitation surge current feature.
From Fig. 8 (b), due to being influenceed by fault current composition, ZAAnd ZCOutput is less than Zset.And IBDue to still table
Reveal feature of shoving, therefore ZBOutput is still more than Zset.And due to using split-phase latching logic, therefore A phases and C after idle-loaded switching-on
Mutually will not locking, differential protection can act rapidly.
4th, inside transformer earth fault in running
Fig. 9 (a) is transformer star side A phase earth faults in operation, and sinusoidal waveform feature is presented in segregated phase differential current.
From Fig. 9 (b), due to occurring internal ground fault, three-phase Z values are exported twice in a cycle, and defeated
Go out value and be much smaller than setting value Zset, protection will not locking.
5th, CT saturation during idle-loaded switching-on
Segregated phase differential current during saturation occurs for transformer under idle-loaded switching-on for Figure 10 (a), is shoved because B is symmetrical smaller, therefore
Saturation does not occur, and saturation occurs for A phases and C phases, shove declines rapidly when saturation occurs for transformer, and produces reverse charging
Electric current, interval angle disappears.Figure 10 (b) provides in this case three-phase Z values output, it is seen that A phases and C phases are due to saturation, and Z values compare Figure 10
(b) output in the case of transformer is unsaturated is small, but the output of three-phase Z values is in ZsetMore than, therefore protection being capable of reliably locking.
Described above is only the preferred embodiment of the present invention, it should be pointed out that：For the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (5)
1. a kind of modified excitation surge current discrimination method, it is characterised in that：Add including main variable data window distinguished number and auxiliary
Fast distinguished number；
Main variable data window distinguished number：Threshold value ± ε is set respectively under on a timeline, threshold value and a certain section of sample rate current is utilized
Extreme value, extreme value is maximum value or minimum value, determines the starting point and terminating point of variable data window；And it is variable what is selected
In data window, according to sampling interval and data window length, a standard sine curve is fictionalized backward forward with this extreme value, calculate
The normalization index Z of difference area between standard sine curve and sample rate current；With the passage of sampled point, algorithm forms multiple
Variable data window, each data window can export a normalization index Z, and algorithm in real time will normalization index Z and setting valve
ZsetIt is compared, and then realizes the locking and opening of protection；Main variable data window distinguished number is specifically referred to：
Fictionalized first according to the extreme point in a certain section come a standard sine curve, fictionalizing the standard sine curve come can
It is divided into two sections：
First paragraph is extreme point in data window to the curved section Y between data window starting point1, it is formulated as：
Y1(k1)=Ilim*sin[π/2-2π(N1-k1)/N], k1=1,2 ... N1
Wherein k1To fictionalize the first paragraph standard sine curve Y come1Discrete point sequence number；IlimFor extreme value in data window；N is one
Cycle total sampling number, N1For curved section Y1Interior total data points；
Second segment is extreme point in data window to the curved section Y between data window terminating point2, it is formulated as：
Y2(k2)=Ilim*sin[π/2-2πk2/ N], k2=1,2 ... N2
Wherein k2To fictionalize the second segment standard sine curve Y come2Discrete point sequence number, N2For curved section Y2Interior total strong point
Number；
Then total standard sine curve can be defined as：
Y=[Y1 Y2]
Make S1For sample rate current curve and the area of time shaft, S2To fictionalize the area of the standard sine curve come and time shaft,
Then S1With S2It can be expressed as：
<mrow>
<msub>
<mi>S</mi>
<mn>1</mn>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>N</mi>
<mn>3</mn>
</mrow>
</munderover>
<mi>i</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>S</mi>
<mn>2</mn>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>N</mi>
<mn>3</mn>
</mrow>
</munderover>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</mrow>
Wherein N3For Y1With Y2Data window total length altogether, j is the sampled point sequence number of actual current, and k is Y1With Y2Altogether
Total standard sine curve y discrete point sequence number, i (j) be above-mentioned data window total length in j-th of sampled point when sample rate current
Instantaneous value, y (k) is instantaneous value of total standard sine curve in k-th of discrete point；
Make N3The sampled point sequence number of actual current and standard sine curve y discrete point sequence number are corresponding in data window, i.e. j=k,
Then calculating difference area S between sample rate current and standard sine curve is：
<mrow>
<mi>S</mi>
<mo>=</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>N</mi>
<mn>3</mn>
</mrow>
</munderover>
<mo>|</mo>
<mi>i</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>|</mo>
</mrow>
Discriminant criterion is normalized, S is mademin=min (S1, S2)
Z=S/Smin
SminFor S1, S2Middle smaller, ZsetFor Z setting valve；
With the passage of sampled point, algorithm forms multiple variable data windows, and each data window can export a normalization index
Z, algorithm is in real time by this normalization index Z and ZsetIt is compared, works as Z>ZsetWhen, the current waveform in data window is differentiated
To shove；As Z≤ZsetWhen, the current waveform in data window is determined as fault current；
Auxiliary accelerates distinguished number：Setting is after auxiliary accelerates distinguished number to start, if phase current does not have Z values during a cycle
Output, then judge the phase current to shove immediately, and this is that auxiliary accelerates distinguished number, so auxiliary accelerates distinguished number most to exist slowly
Latch-up protection during a cycle.
2. modified excitation surge current discrimination method according to claim 1, it is characterised in that：The selection tool of variable data window
Body refers to：First on a timeline under threshold value ± ε is set respectively, if protection sample rate f, sampling interval T=1/f, each cycle adopts
Number of samples is N, is comprised the following steps：
1) determination of maximum value or minimum value：System is normally and before idle-loaded switching-on, and poor stream should be within threshold interval；Stream of being on duty is super
When going out threshold interval, main variable data window distinguished number starts, and selection sample rate current passes sampled point conduct during threshold interval
The starting point of first extremum search, sampled point when selection sample rate current is again out the threshold interval is as first extreme value
The terminating point of search；First positive maximum is searched out in starting point to the data window of terminating point of first extremum search
ImaxOr reverse minimum Imin, record the sampled value I corresponding to first extreme pointlimAnd extreme value sampled point sequence number n, similarly
Find second extreme point；
2) determination of variable data window：Searched for backward forward using first extreme point as basic point, regulation forward in search procedure with
Electric current passed from threshold region after first point of starting point as variable data window, the instantaneous value of starting point is I1, sampling
Point serial number a, search condition such as formula (1)；If any sampled point is not left in region, with electric current in search procedure forward
Across first point of starting point as variable data window after threshold region, search condition such as formula (2) seeks data window starting point
Algorithm be：
When
i(j+1)≤|ε|&i(j)>| ε | j=n-1, n-2, n-3 ... (1)
Or
When,
Then make a=j, I1=i (a)
In formula i be sample rate current instantaneous value, j be actual current sampled point sequence number, n be extreme value sampled point sequence number, i (n)=
Ilim；
First point of terminating point as variable data window after being passed backward in search procedure using electric current from threshold region, its wink
Duration is I2, sampled point serial number b, search condition such as formula (3)；If not leaving any sampled point in region, searching backward
The terminating point of variable data window, search condition are equally used as using first sampled point after current crosses threshold region during rope
Such as formula (4), i.e.,
i(j-1)≤|ε|&i(j)>| ε | j=n+1, n+2, n+3 ... (3)
Or
When,
Make b=j, I2=i (b)；
In formula i be sample rate current instantaneous value, j be actual current sampled point sequence number, n be extreme value sampled point sequence number, i (n)=
Ilim；
3) first extreme point and I are calculated1Between sampling number N1, include IlimAnd I1, fictionalize standard sine bent at 2 points
The first paragraph of line, order
Y1(k1)=Ilim*sin[π/2-2π(N1-k1)/N], k1=1,2 ... N1
Calculate first extreme point and I2Between sampling number N2, not comprising IlimBut include I2Point, fictionalizes standard sine
The second segment of curve, order
Y2(k2)=Ilim*sin[π/2-2πk2/ N], k2=1,2 ... N2
By sequence Y1With Y2Spliced, made
Y=[Y1 Y2]
4) I is calculated1, I2Between points N3, include I1And I2, make N3The sampled point sequence number of actual current and standard be just in data window
Chord curve y discrete point sequence number is corresponding, i.e. j=k, then calculates difference area S between sample rate current and standard sine curve
For：
<mrow>
<mi>S</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>N</mi>
<mn>3</mn>
</mrow>
</munderover>
<mo>|</mo>
<mi>i</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>|</mo>
<mo>.</mo>
</mrow>
3. modified excitation surge current discrimination method according to claim 1, it is characterised in that：With the passage of data window,
It is as follows that algorithm continuously exports the decision rule that Z values are exported under Z values, different situations：In one cycle, there be the output of Z values under different situations
Zero degree, once or twice；
(1) when Z values are output as zero degree in a cycle, then perform auxiliary accelerate distinguished number, judge such a situation as
Shove, locking is carried out during a cycle, no open hour, such a inrush shoves to be asymmetric；
(2) when Z values are output as once in a cycle, now, only Z>Zset, judge such a situation to shove, carry out
Locking, blocking time is less than a cycle, and no open hour, waveform shoves to be asymmetric；
(3) when Z values are output as twice in a cycle, with Z values, output is judged for the first time, determines whether protection closes
Whether lock ensures to open, not locking afterwards；Major punishment is carried out with second of the output of Z values to determine, and determines to protect when first time exporting Z values
Whether reopen and put after shield locking, as Z≤ZsetThen weigh open and protection；Work as Z>ZsetOr Z values are exported without second, then continue locking and protect
Shield；It is specifically divided into following 4 kinds of classifications：
The first is first to export a Z>Zset, locking is carried out, then export a Z>Zset, continue to keep blocking above,
It is now inrush, no open hour, inrush is symmetrically to shove；
Second is first to export a Z≤Zset, it is judged to ensureing to open, then export a Z≤Zset, major punishment is set to internal former
Barrier, now without locking and open hour；
The third is first to export a Z≤Zset, it is judged to ensureing to open, then export a Z>Zset, it is empty that major punishment is set to tape jam
Carry and close a floodgate, now without locking and open hour；
4th kind is first to export a Z>Zset, locking is carried out, then export a Z≤Zset, major punishment is set to tape jam idle-loaded switching-on,
Open and protection is exported, and now blocking time is the half period, when the open hour are a cycle.
4. modified excitation surge current discrimination method according to claim 1, it is characterised in that：Consider and transformer longitudinal
The supporting of starting current is protected, threshold value ± ε is set, ε=0.18I is setN, INFor transformer rated current.
5. modified excitation surge current discrimination method according to claim 1, it is characterised in that：Consider the reliability of protection,
Set Zset=0.55.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510217672.2A CN104810795B (en) | 2015-04-30 | 2015-04-30 | A kind of modified excitation surge current discrimination method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510217672.2A CN104810795B (en) | 2015-04-30 | 2015-04-30 | A kind of modified excitation surge current discrimination method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104810795A CN104810795A (en) | 2015-07-29 |
| CN104810795B true CN104810795B (en) | 2017-10-24 |
Family
ID=53695405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510217672.2A Active CN104810795B (en) | 2015-04-30 | 2015-04-30 | A kind of modified excitation surge current discrimination method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104810795B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106549360B (en) * | 2017-01-18 | 2018-09-25 | 南京国电南自电网自动化有限公司 | A kind of excitation flow recognition method based on zero-sequence current |
| CN108173240B (en) * | 2017-12-19 | 2019-06-28 | 华中科技大学 | A kind of transformer excitation flow recognition method and system based on compound time difference method |
| CN110581485B (en) * | 2018-06-07 | 2021-06-08 | 安科瑞电气股份有限公司 | Microcomputer protection method for preventing misoperation caused by electromagnetic interference |
| CN109446484A (en) * | 2018-10-12 | 2019-03-08 | 湖南磁浮技术研究中心有限公司 | Data discrimination method and system based on curve comparison |
| DE102019120400A1 (en) * | 2019-07-29 | 2021-02-04 | Friedrich-Alexander-Universität Erlangen-Nürnberg | Method for the detection of an inrush event, computer program and disconnection device |
| CN112526338A (en) * | 2020-11-24 | 2021-03-19 | 重庆理工大学 | Novel synchronous time window excitation power unit online fault diagnosis method |
| CN113991606B (en) * | 2021-10-18 | 2024-01-02 | 国电南瑞科技股份有限公司 | Excitation surge current misoperation prevention method and device for transformer |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101764391A (en) * | 2009-08-20 | 2010-06-30 | 北京四方继保自动化股份有限公司 | Method and system for braking magnetizing inrush current in differential protection of ultra-high voltage transformer |
| CN102255284A (en) * | 2011-07-26 | 2011-11-23 | 重庆电力高等专科学校 | Method for realizing protection of transformer by comparing polarities of instantaneous values of current fault components |
| WO2014032915A1 (en) * | 2012-08-31 | 2014-03-06 | Siemens Aktiengesellschaft | Magnetizing inrush current detection method, magnetizing inrush current restraint method and device |
-
2015
- 2015-04-30 CN CN201510217672.2A patent/CN104810795B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101764391A (en) * | 2009-08-20 | 2010-06-30 | 北京四方继保自动化股份有限公司 | Method and system for braking magnetizing inrush current in differential protection of ultra-high voltage transformer |
| CN102255284A (en) * | 2011-07-26 | 2011-11-23 | 重庆电力高等专科学校 | Method for realizing protection of transformer by comparing polarities of instantaneous values of current fault components |
| WO2014032915A1 (en) * | 2012-08-31 | 2014-03-06 | Siemens Aktiengesellschaft | Magnetizing inrush current detection method, magnetizing inrush current restraint method and device |
Non-Patent Citations (2)
| Title |
|---|
| 结合波形后半波2-范数平均标准差鉴别变压器励磁涌流新方法;张凯等;《华北电力大学学报》;20130531;第40卷(第3期);第36-42页 * |
| 采用自适应数据窗电流的励磁涌流鉴别方法;王业等;《中国电机工程学报》;20140205;第34卷(第4期);第702-711页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104810795A (en) | 2015-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104810795B (en) | A kind of modified excitation surge current discrimination method | |
| CN104466903B (en) | Current transformer saturation recognition method for differential protection | |
| CN104810796A (en) | Method of identifying excitation inrush current on basis of normalized area indexes of variable data window | |
| CN106932747B (en) | A kind of CT saturation detection method that global and local feature combines | |
| CN105262051B (en) | Transformer excitation flow discrimination method based on sample sequence absolute value partial velocities | |
| CN101295869A (en) | High-credibility excitation inrush current braking method of transformer device | |
| CN103986120B (en) | Transformer differential protection system and method based on magnetic flux brake trap combined judgment | |
| CN106505523A (en) | A kind of excitation flow recognition method suitable for Traction networks transformer | |
| CN110514895A (en) | A kind of excitation surge current method for quickly identifying for Yd wiring transformer | |
| CN113109671B (en) | Method for accelerating detection of small half-wave stage of fault current | |
| CN105954563B (en) | The method for identifying tail currents based on electric current derivation and difference current zero crossing | |
| CN108134373B (en) | Voltage-regulating compensation transformer excitation inrush current locking method | |
| CN107123974A (en) | A kind of bus troubles inside the sample space CT saturations based on saturation mechanism open trip method | |
| CN104410044B (en) | Identification method for excitation surge current of transformer based on kurtosis and skewness | |
| CN105866511A (en) | Method for recognizing trailing currents based on adjacent sampling point specific values of current wave forms | |
| CN108695823A (en) | A kind of differential protection blocking method of series transformer tail currents | |
| CN114844000A (en) | Inter-turn protection method and system for converter transformer | |
| CN107294062B (en) | A kind of failure decision method when transformer drops | |
| CN110531195A (en) | A method of identification transformer excitation flow and internal fault | |
| CN109714034B (en) | Automatic tracking silicon controlled rectifier quasi-current zero-crossing trigger control method and circuit | |
| CN109510177A (en) | A kind of voltage jump amount guard method of DC power transmission line | |
| CN107860963A (en) | A kind of AC fault arc method for measuring and its device | |
| CN107134763B (en) | A kind of method of discrimination that fault trip quickly returns | |
| CN106253245A (en) | A kind of substance accelerates the method for discrimination of open asymmetric open element during closing a floodgate | |
| CN202042889U (en) | Line micro computer protection device utilizing inrush current identification technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |