CN110071483A - Bridge differential current protection faulty action preventing based on mutual approximate entropy algorithm is latched determination method - Google Patents
Bridge differential current protection faulty action preventing based on mutual approximate entropy algorithm is latched determination method Download PDFInfo
- Publication number
- CN110071483A CN110071483A CN201910399364.4A CN201910399364A CN110071483A CN 110071483 A CN110071483 A CN 110071483A CN 201910399364 A CN201910399364 A CN 201910399364A CN 110071483 A CN110071483 A CN 110071483A
- Authority
- CN
- China
- Prior art keywords
- sequence
- current
- criterion
- bridge
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 229960004402 tiopronin Drugs 0.000 claims abstract description 23
- YTGJWQPHMWSCST-UHFFFAOYSA-N Tiopronin Chemical compound CC(S)C(=O)NCC(O)=O YTGJWQPHMWSCST-UHFFFAOYSA-N 0.000 claims abstract description 18
- 239000013598 vector Substances 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 230000002265 prevention Effects 0.000 claims description 3
- 230000005284 excitation Effects 0.000 abstract description 5
- 238000005259 measurement Methods 0.000 abstract description 5
- 238000011084 recovery Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 6
- 229920006395 saturated elastomer Polymers 0.000 description 5
- 238000004804 winding Methods 0.000 description 5
- 230000005389 magnetism Effects 0.000 description 3
- 230000016507 interphase Effects 0.000 description 2
- 108010076504 Protein Sorting Signals Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
- H02H7/10—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for converters; for rectifiers
- H02H7/12—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for converters; for rectifiers for static converters or rectifiers
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Emergency Protection Circuit Devices (AREA)
- Protection Of Transformers (AREA)
Abstract
Bridge differential current protection faulty action preventing based on mutual approximate entropy algorithm is latched determination method, comprising: obtains the current sequence I that the Y/Y change of current becomes secondary side current mutual inductor TA measurement12a、I12bAnd I12cAnd the Y/ △ change of current becomes the current sequence I of secondary side TA measurement22a、I22bAnd I22c;Utilize Bridge differential current protection amplitude criterion formula △ IY=Iac‑IacYWith △ ID=Iac‑IacDForm Y bridge criterion sequence △ IYWith D bridge criterion sequence △ ID;Judge amplitude criterion △ IY>IsetOr △ ID>IsetIt is whether true;Judge whether it is idle-loaded switching-on;Calculate three groups of current sequence I that the Y/ △ change of current becomes secondary side TA measurement22a、I22bAnd I22cMutual approximate entropy CApEn (m, r, N) value between any two;The new action logic of Bridge differential current protection is made of amplitude criterion and locking criterion.The method of the present invention is while guaranteeing Bridge differential current protection action performance, Bridge differential current protection nonaction caused by the recovery inrush that the excitation surge current and Removal of external faults that generate by idle-loaded switching-on can be effectively prevent to generate.
Description
Technical Field
The invention relates to the technical field of converter station bridge difference protection, in particular to a bridge difference protection anti-misoperation lockout determination method based on a mutual approximation entropy algorithm.
Background
The converter bridge difference protection is backup protection for protecting a direct current system by tripping off a network side breaker when a converter bridge has a valve commutation fault or a trigger fault.
The converter transformer is easy to generate serious inrush current with unconventional characteristics in the no-load switching-on and external fault removal and power restoration processes, and the frequent occurrence and new characteristics of the serious inrush current can cause misoperation of bridge differential protection of the converter. When the direct current high slope converter station is over the direct current high slope, the bridge difference II section protection misoperation is triggered due to the standby switching locking operation of the converter transformer, the direct current single pole locking is caused, and direct impact is caused on a plurality of devices and even systems of the converter station.
At present, the countermeasure for preventing the bridge error protection misoperation is to use TA with better transient characteristic, optimize the protection fixed value and introduce the zero sequence circulation at the △ side of the Y/△ converter transformer into the starting criterion as the braking quantity.
The method has certain guiding significance for preventing the bridge difference protection misoperation, but still has a plurality of defects in practical application: the replacement of the TA increases the cost and is easily restricted by the installation location; increasing the protection rating may reduce the sensitivity and snap-action of converter bridge differential protection to in-zone faults.
The bridge difference protection criterion is analyzed, and it is found that both the excitation inrush current generated by the converter transformer during no-load closing and the recovery inrush current generated by the converter transformer during external fault removal generate a large zero-sequence circulating current in a ring at the side of the Y/△ converter transformer △, and when the zero-sequence circulating current flows through a certain phase winding TA with large remanence, the phase TA is saturated, so that TA transmission characteristics are degraded, the current amount adopted by protection is mistakenly measured, and bridge difference protection misoperation is caused.
The false action reason is analyzed, the current sequence of the TA saturated phase and the current sequence of other normal phases have obvious difference in similarity degree, and the mutual approximation entropy algorithm has obvious advantage in processing sequence similarity. Therefore, the similarity degree between the current sequences is calculated by utilizing a mutual approximation entropy algorithm, and a bridge difference protection anti-misoperation locking criterion is formed.
Disclosure of Invention
The invention provides a bridge difference protection misoperation-prevention lockout judgment method based on a mutual approximation entropy algorithm, which can effectively identify the difference between a TA saturated phase and a normal phase current sequence and can effectively solve the problem of bridge difference protection misoperation caused by TA saturation of a winding of a converter transformer.
The technical scheme adopted by the invention is as follows:
the bridge difference protection anti-misoperation lockout judgment method based on the mutual approximation entropy algorithm comprises the following steps:
step 1: under a certain sampling frequency, obtaining a current sequence I measured by a secondary side current transformer TA of a Y/Y converter transformer12a、I12bAnd I12cAnd a current sequence I measured by the secondary side TA of the Y/△ converter transformer22a、I22bAnd I22cThe formula △ I is determined by using bridge difference protection amplitudeY=Iac-IacYAnd △ ID=Iac-IacDForming a Y bridge criterion sequence △ IYAnd D bridge criterion sequence △ ID. Wherein, IacY=max{IaY,IbY,IcY}、IacD=max{IaD,IbD,IcD}、Iac=max{IacY,IacD};
IaY、IbYAnd IcYAre each equal to I12a、I12bAnd I12c;
IaD、IbDAnd IcDAre each equal to I22a-I22b、I22b-I22cAnd I22c-I22a。
Step 2, judging amplitude criterion △ IY>IsetOr △ ID>IsetIs established, wherein IsetIf the bridge difference protection action setting value is established, starting a misoperation prevention locking criterion to step 3; otherwise, returning to continuously judge the inequality.
And step 3: judging whether the switching-on is carried out in an idle load mode, and if the switching-on is carried out in the idle load mode, executing the step 4; otherwise, the current sequence is subjected to standardization processing.
Step 4, calculating three groups of current sequences I measured by TA at the secondary side of the Y/△ converter transformer22a、I22bAnd I22cMutually approximate entropy Capen (m, r, N) values between every two, and convert the three groups of results obtained by calculation into corresponding three groups of C values; the specific method comprises the following steps: by the formulaCalculating the value of the Capen (m, r, N) between the two sequences, wherein r is the similarity tolerance; b and A are respectively the similarity of two sequence sampling values under the tolerance r, and are marked as events A and B, Pi(B | A) is the probability of similarity in the sense of tolerance r; n is the sample length; m is the mode dimension, and then the calculated value of CApEn (m, r, N) is taken as C ═ e-CApEn(m,r,N)Conversion is performed so that the obtained result is in the interval (0, 1)]And (4) the following steps.
And 5: the amplitude criterion and the locking criterion form a new action logic of bridge difference protection, when the amplitude criterion is met, the action quantity output is 1, otherwise, the action quantity output is 0; when there are two or more groups C<Cset,CsetWhen the locking criterion setting value is set, the locking criterion action output quantity is 0, otherwise, the locking criterion action output quantity is 1;when the amplitude criterion action quantity and the locking criterion action quantity are both 1, the bridge difference protection acts; otherwise, locking is performed.
A bridge difference protection anti-misoperation lockout judgment method based on a mutual approximation entropy algorithm is characterized in that a mutual approximation entropy algorithm is adopted to calculate the value of a Capen (m, r, N) between every two phases of a current sequence of a Y/△ converter transformer three-phase TA secondary side, and the result is converted into a value of C.
The invention discloses a bridge difference protection anti-misoperation lockout judgment method based on a mutual approximation entropy algorithm, which has the following technical effects:
the method compares the similarity degree between every two phases of the three-phase current sequence on the secondary side of the Y/△ converter transformer TA, has better anti-noise capability and can deal with strong interference which is generated occasionally.
2: the invention does not need to describe the complete picture of the signal sequence, and only describes the similarity of the sequence from the statistical perspective, so that the data window is flexible to select, and a stable statistical value can be obtained by using less data.
3: the invention is used as the bridge difference protection locking criterion, can effectively prevent the protection maloperation caused by excitation surge current and recovery surge current when no-load switching-on and external fault removal, and has high reliability.
Drawings
FIG. 1(a) is a waveform diagram of primary side current of three-phase TA of a Y/△ converter during no-load closing;
FIG. 1(b) is a waveform diagram of the secondary side current of the Y/△ converter transformer three-phase TA when no load is applied.
FIG. 2(a) is a waveform diagram of primary side current of three-phase TA of Y/△ converter when external fault is removed;
FIG. 2(b) is a waveform diagram of secondary side current of three-phase TA of Y/△ converter when external fault is removed.
Fig. 3(a) is a C-value comparison diagram between two phases of three-phase current on the side of the Y/△ converter transformer △ during no-load closing of the present invention.
Fig. 3(b) is a diagram showing the output of the three-phase current bridge differential protection operation amount at the side of the Y/△ converter transformer △ during no-load closing of the present invention.
Fig. 4(a) is a C-value comparison graph between each two phases of three-phase current on the side of the Y/△ converter transformer △ when the external fault is removed.
Fig. 4(b) is a graph showing the output of the three-phase current bridge differential protection operation amount on the side of the Y/△ converter transformer △ when the external fault is removed.
Fig. 5(a) is a C-value comparison graph between each two phases of three-phase current at the side of the Y/△ converter transformer △ during internal fault of the invention.
FIG. 5(b) is a graph showing the output of the Y/△ converter transformer △ side three-phase current bridge differential protection operation amount at the time of an internal fault in the present invention.
FIG. 6 is a flow chart of a novel scheme of bridge difference protection anti-false-operation blocking based on mutual approximation entropy algorithm.
Detailed Description
As shown in fig. 6, the method for determining the bridge difference protection anti-false-operation locking based on the mutual approximation entropy algorithm includes the following steps:
step 1: at a certain sampling frequency, taking 4kHz as an example, obtaining a current sequence I measured by a secondary side TA of the Y/Y converter transformer12a、I12bAnd I12cAnd current sequence I measured by secondary side TA of Y/△ converter transformer22a、I22bAnd I22c。
Formula △ I using bridge difference protection amplitude criterionY=Iac-IacYAnd △ ID=Iac-IacDForming a Y bridge criterion sequence △ IYAnd D bridge criterion sequence △ ID。
Wherein, IacY=max{IaY,IbY,IcY}、IacD=max{IaD,IbD,IcD}、Iac=max{IacY,IacD};IaY、IbYAnd IcYAre each equal to I12a、I12bAnd I12c;IaD、IbDAnd IcDAre each equal to I22a-I22b、I22b-I22cAnd I22c-I22a。
Step 2, judging amplitude criterion △ IY>IsetOr △ ID>IsetIs established, wherein IsetAnd setting a bridge difference protection action value. If yes, starting the false action prevention locking criterion provided by the invention, and going to step 3; otherwise, returning to continuously judge the inequality.
And step 3: judging whether the switching-on is carried out in an idle load mode, and if the switching-on is carried out in the idle load mode, executing the step 4; otherwise, the current sequence is subjected to standardization processing.
Step 4, calculating three groups of current sequences I measured by TA at the secondary side of the Y/△ converter transformer22a、I22bAnd I22cAnd (3) mutually approximating entropy Capen (m, r, N) values between every two, and converting three groups of results obtained by calculation into corresponding three groups of C values.
The specific method comprises the following steps: by the formulaCalculating the value of the CApEn (m, r, N) between the two sequences,
wherein r is a similarity tolerance; b and A are respectively the similarity of two sequence sampling values under the tolerance r, and are marked as events A and B, Pi(B | A) is the probability of similarity in the sense of tolerance r; n is the sample length; m is the mode dimension. Then, the calculated value of CApEn (m, r, N) is set as C ═ e-CApEn(m,r,N)Conversion is performed so that the obtained result is in the interval (0, 1)]And (4) the following steps.
And 5: the amplitude criterion and the locking criterion form a new action logic of bridge difference protection. When the amplitude criterion is satisfied, the action quantity is outputIs 1, otherwise is 0; when there are 2 groups and above C<Cset,CsetAnd when the locking criterion setting value is the locking criterion setting value, the locking criterion action output quantity is 0, otherwise, the locking criterion action output quantity is 1. When the amplitude criterion action quantity and the locking criterion action quantity are both 1, the bridge difference protection acts; otherwise, locking is performed.
1. When no-load switching-on and external faults are removed, the current capacity characteristics of the primary side and the secondary side of the Y/△ converter transformer TA are as follows:
when no-load switching-on and external fault removal occur, the combined action of zero-sequence circulating current generated in △ side rings of the Y/△ converter transformer and residual magnetism of windings causes TA of a certain phase to be saturated, so that the measurement of protection electricity flow of the phase is wrong, and bridge difference protection is misoperated.
It is assumed that residual magnetism exists in a phase A in a winding TA at a side of a Y/△ converter transformer △, and the waveforms of current waveforms of a phase A in a three-phase TA of a Y/△ converter transformer at the time of no-load switching-on and a current waveform of a secondary side in a time of no-load switching-on, and switching-on at a time of t being 0.1s, and the waveforms of current waveforms of a phase A and a phase B in a time of external fault removal in a time of Y/△ converter transformer at the time of fig. 2(a) and fig. 2(b), when an external fault occurs at the time of t being 0.1s and a fault is removed at the time of t being 0.2s, zero-sequence currents generated by excitation inrush current and recovery inrush current and zero-sequence currents flow in a loop at a side of a Y/△ converter transformer △, the phase TA is saturated by the residual magnetism, so that a transmitted current waveform is distorted, and other two phases TA are not affected, so that a current waveform sequence of.
2. Mutual approximation entropy algorithm:
mutual approximation entropy is a quantity that represents the degree of complexity between two discrete sequences. It analyzes the sequence from the view point of sequence complexity to generate new pattern probability, and judges the similarity degree by calculating the conditional probability of the same pattern existing between two sequences.
The mutual approximate entropy method is used for expressing the similarity degree of two sequences, namely finding the similar mode of the two sequences, and the mutual approximate entropy value is defined as:
in the formula: r is a similarity tolerance; b and A are respectively the similarity of two series sampling values in a tolerance r, and are marked as events A and B, Pi(B | A) is the probability of similarity in the sense of tolerance r; n is the sample length, and the mutual approximation entropy can obtain a stable value only by comparing relatively few data points; m is the mode dimension, and in practical application, m is 2.
Calculating a mutual approximation entropy value according to the following steps:
(1) and sequentially carrying out vector reconstruction on the two current sequences to be compared to obtain vectors corresponding to the two currents. Firstly, a current sequence is divided into a plurality of one-dimensional time sequences with the length of N, ix={ix(n) } and iy={iy(N), wherein N is 1, … N. Reconstructing m-dimensional vectors according to the formula (2) and the formula (3), and respectively recording the two current sequences as Ix(j) And Iy(k)。
Ix(j)=[ix(j),…,ix(j+m-1)],j=1,…,N-m+1 (2)
Where N is the sample length, m is the mode dimension, typically 2, and j is the current sequence IxNumber of (1), ix(j) For a sequence of currents ix(n) } j-th point, ix(j + m-1) is a current sequence ix(n) } j + m-1 th point, Ix(j) Is the jth m-dimensional vector.
Iy(k)=[iy(k),…,iy(k+m-1)],k=1,…,N-m+1 (3)
Where N is the sample length, m is the mode dimension, typically 2, and k is the current sequence IyNumber of (1), iy(k) For a sequence of currents iy(n) } the k-th point, iy(k + m-1) is the current sequence iy(n) } the k + m-1 th point in the center, Iy(k) Is the k-th m-dimensional vector.
When current sequence ixAnd iyWhen the amplitude difference is large, the current sequence is generally standardized firstly, and the specific method is as follows:since the sequence has already been standardized, r may be 0.2. Wherein,for normalized current sequences ix(n),mean(ix) As a sequence of currents ixAverage value of ix(n) is a current sequence ixPoint n of middle, SD (i)x) As a sequence of currents ixThe standard deviation of (a) is determined,for normalized current sequences iyN-th point of (1), iy(n) is a current sequence iyThe nth point in middle, mean (i)y) As a sequence of currents iyAverage value of (d), SD (i)y) As a sequence of currents iyStandard deviation of (2).
(2) Calculating the similar tolerance according to the two groups of current sequences, wherein r is the current sequence ix(n) and iy(n) 0.2 times the covariance, i.e. r 0.2 × COV (i)x(n),iy(n)), wherein
(3) Defining the distance between two sequences as Ix(j) And Iy(k) Corresponding to the maximum value of the difference of the currents, the maximum difference expression is as follows:
wherein j is a current sequence IxK is the current sequence IyP is 0 and 1, m is the mode dimension, typically 2, Ix(j + p) is the (j + p) th current sequence, Iy(k + p) is the (k + p) th current sequence.
(4) And counting d (I) for each j and k according to the set value r of the similarity marginx(j),Iy(k) A number smaller than r) and calculating the ratio of the number of r to the total number of vectors N-m +1, denoted as Cm,rThe value is the approaching probability of the m-dimensional mode in the two sequences under the condition of the similar tolerance r, and the specific expression is as follows:
Cm,r=[d(Ix(j),Iy(k))<number of r]/(N-m+1) (5)
(5) To Cm,rObtaining I by calculating the logarithm and then the averagex(j) And Iy(k) Mutual similarity of (D) is denoted as Tm,r[d(Ix(j),Iy(k))]The specific expression is as follows:
(6) and when the embedding dimension is changed into m +1, repeating the steps (1) to (4) to obtain C with the dimension of m +1m+1,rAnd Tm+1,r[d(Ix(j),Iy(k))]Will Tm,r[d(Ix(j),Iy(k))]And Tm+1,r[d(Ix(j),Iy(k))]And (3) obtaining a final mutual approximation entropy value of the two currents by difference, wherein the expression is as follows:
CApEn(m,r,N)=Tm,r[d(Ix(j),Iy(k))]-Tm+1,r[d(Ix(j),Iy(k))](7)
according to the steps, the mutual approximate entropy value of the two sequences can be obtained.
Analyzing according to the steps, and finding that if the similarity degree of the two sequences is higher, the mutual approximation entropy value obtained according to the formula (7) is almost 0; if they are not similar, the mutual approximation entropy is larger. Therefore, the calculated mutual approximation entropy ranges from the value of Capen (m, r, N) E [0, + ∞). The range of the value is too wide, which is not beneficial to forming effective setting value to divide the similarity of the two waveform sequences. Therefore, further processing of the mutual approximation entropy values is required. According to y ═ e-xAs a result, when x ∈ [0 ] + ∞ ], y ∈ [0, 1 ]]Therefore, the value C is defined as the similarity discrimination value of the present invention, and the specific formula is:
C=e-CApEn(m,r,N)(8)
extracting a current sequence I measured by a secondary side three-phase TA of a Y/△ converter transformer22a、I22bAnd I22cThree sets of comparison sequences (I)22aAnd I22b、I22bAnd I22c、I22cAnd I22a)。
Substituting the two sequences to be compared into an algorithm calculation formula by utilizing a mutual approximation entropy algorithmxAnd IyAnd when the converter transformer is subjected to idle load switching, external fault removal and internal fault (A, B interphase fault) respectively in the steps of FIG. 3(a), FIG. 4(a) and FIG. 5(a), the C value between each two phases of the secondary side three-phase current of the Y/△ converter transformer is set to be 1 because the new anti-misoperation lockout criterion is started simultaneously with the bridge difference protection amplitude criterion, so that the C value before the bridge difference protection amplitude criterion is started is set to be 1.
As can be seen from fig. 3(a), fig. 4(a) and fig. 5(a), when no-load switching and external fault removal are performed, two groups of C values in the three groups of C values are smaller than the setting value of 0.65, and when an internal fault occurs, the three groups of C values are all above the setting value of 0.65.
3. A novel bridge difference protection anti-misoperation lockout scheme based on a mutual approximation entropy algorithm comprises the following steps:
obtaining secondary side TA measurement of Y/Y converter transformer under certain sampling frequencySequence of currents of magnitude I12a、I12bAnd I12cAnd current sequence I measured by secondary side TA of Y/△ converter transformer22a、I22bAnd I22cForming a Y-bridge criterion sequence △ I by using a bridge difference protection amplitude criterionYAnd D bridge criterion sequence △ IDJudgment △ IYOr △ IDAfter the new bridge difference protection misoperation-prevention locking scheme is started, firstly, whether the current sequence is in no-load closing is judged, if the current sequence is in no-load closing, the current sequence is directly calculated, otherwise, the current sequence is subjected to standardized processing, and then three groups of current sequences (I) measured by secondary side TA of the Y/△ converter transformer are calculated22a、I22bAnd I22c) The value of CApEn (m, r, N) between each two, and the 3 sets of results obtained by calculation are converted into the corresponding 3 sets of C values. The specific method comprises the following steps: by the formulaCalculating the value of the CApEn (m, r, N) between the two sequences, and then taking the calculated value of the CApEn (m, r, N) as C ═ e-CApEn(m,r,N)Conversion is performed so that the obtained result is in the interval (0, 1)]And (4) the following steps.
Through setting reasonable CsetAnd the internal fault can be effectively distinguished from other conditions. The amplitude criterion and the locking criterion form a new action logic of bridge difference protection. When the amplitude criterion is satisfied, the output of the motion quantity is 1, otherwise, the output is 0; when there are 2 groups and above C<Cset,CsetThe novel bridge difference protection misoperation-preventing lockout scheme based on the mutual approximation entropy algorithm can effectively prevent the phenomenon of bridge difference protection misoperation caused by excitation surge current generated by no-load closing and recovery surge current generated when external faults are removedThe flow chart of the scheme is shown in figure 6.
The new scheme is utilized to calculate the C value between each two phases of the secondary side three-phase current sequence of the Y/△ converter transformer TA when the converter transformer has no-load closing, external fault removal and internal fault (A, B interphase fault), and whether the bridge difference protection acts or not is judged according to new action logic, and the result is shown in fig. 3(b), fig. 4(b) and fig. 5(b), wherein C issetTake 0.65.
It can be clearly seen that:
(1): and when no-load switching-on and external fault removal are carried out, the locking criterion is started, the action output quantity of the amplitude criterion is 1, but two groups of C values in the three groups of C values obtained by calculation are less than 0.65 setting value, and the action output quantity of the C value criterion is 0. Therefore, the bridge difference protection action output quantity does not satisfy the action logic. The protection can be locked reliably at the moment, and misoperation cannot occur.
(2) When an internal fault occurs, the C value between every two phases of the secondary side three-phase current of the Y/△ converter transformer TA is greater than a setting value of 0.65, the amplitude criterion action output quantity and the C value criterion action output quantity are both 1 at the moment, a new criterion action logic is met, the internal fault can be reliably identified at the moment, and the protection is continuously opened.
The invention discloses a bridge difference protection anti-misoperation lockout judgment method based on a mutual approximation entropy algorithm, which can effectively prevent the misoperation problem caused by TA saturation of a winding while ensuring the performance of bridge difference protection actions.
Claims (4)
1. The bridge difference protection anti-misoperation lockout determination method based on the mutual approximation entropy algorithm is characterized by comprising the following steps of:
step 1: under a certain sampling frequency, obtaining a current sequence I measured by a secondary side current transformer TA of a Y/Y converter transformer12a、I12bAnd I12cAnd a current sequence I measured by the secondary side TA of the Y/△ converter transformer22a、I22bAnd I22cThe formula △ I is determined by using bridge difference protection amplitudeY=Iac-IacYAnd △ ID=Iac-IacDForming a Y bridge criterion sequence △ IYAnd D bridge criterion sequence △ ID;
Wherein, IacY=max{IaY,IbY,IcY}、IacD=max{IaD,IbD,IcD}、Iac=max{IacY,IacD};
IaY、IbYAnd IcYAre each equal to I12a、I12bAnd I12c;
IaD、IbDAnd IcDAre each equal to I22a-I22b、I22b-I22cAnd I22c-I22a;
Step 2, judging amplitude criterion △ IY>IsetOr △ ID>IsetIs established, wherein IsetIf the bridge difference protection action setting value is established, starting a misoperation prevention locking criterion to step 3; otherwise, returning to continuously judge the inequality;
and step 3: judging whether the switching-on is carried out in an idle load mode, and if the switching-on is carried out in the idle load mode, executing the step 4; otherwise, the current sequence is subjected to standardization treatment;
step 4, calculating three groups of current sequences I measured by TA at the secondary side of the Y/△ converter transformer22a、I22bAnd I22cMutually approximate entropy Capen (m, r, N) values between every two, and convert the three groups of results obtained by calculation into corresponding three groups of C values; the specific method comprises the following steps: by the formulaCalculating the value of the Capen (m, r, N) between the two sequences, wherein r is the similarity tolerance; b and A are respectively the similarity of two sequence sampling values under the tolerance r, and are marked as events A and B, Pi(B | A) is the probability of similarity in the sense of tolerance r; n is the sample length; m is the mode dimension, and then the calculated value of CApEn (m, r, N) is taken as C ═ e-CApEn(m,r,N)Conversion is performed so that the obtained result is in the interval (0, 1)]Internal;
and 5: the amplitude criterion and the locking criterion form a new action logic of bridge difference protectionWhen the amplitude criterion is satisfied, the output of the motion quantity is 1, otherwise, the output is 0; when there are two or more groups C<Cset,CsetWhen the locking criterion setting value is set, the locking criterion action output quantity is 0, otherwise, the locking criterion action output quantity is 1; when the amplitude criterion action quantity and the locking criterion action quantity are both 1, the bridge difference protection acts; otherwise, locking is performed.
2. The bridge difference protection anti-false-operation locking judgment method based on the mutual approximation entropy algorithm is characterized in that: in step 4, the mutual approximation entropy is calculated according to the following steps:
sequentially carrying out vector reconstruction on the two current sequences to be compared to obtain vectors corresponding to the two currents; firstly, a current sequence is divided into a plurality of one-dimensional time sequences with the length of N, ix={ix(n) } and iy={iy(N), wherein N is 1, … N; reconstructing m-dimensional vectors according to the formula (2) and the formula (3), and respectively recording the two current sequences as Ix(j) And Iy(k);
Ix(j)=[ix(j),…,ix(j+m-1)],j=1,…,N-m+1 (2)
Where N is the sample length, m is the mode dimension, typically 2, and j is the current sequence IxNumber of (1), ix(j) For a sequence of currents ix(n) } j-th point, ix(j + m-1) is a current sequence ix(n) } j + m-1 th point, Ix(j) Is the jth m-dimensional vector;
Iy(k)=[iy(k),…,iy(k+m-1)],k=1,…,N-m+1 (3)
where N is the sample length, m is the mode dimension, typically 2, and k is the current sequence IyNumber of (1), iy(k) For a sequence of currents iy(n) } the k-th point, iy(k + m-1) is the current sequence iy(n) } the k + m-1 th point in the center, Iy(k) Is the k-th m-dimensional vector;
when current sequence ixAnd iyWhen the amplitude difference is large, the current sequence is generally standardized firstly, and the specific method isSince the sequence has been standardized, r may be 0.2;
wherein,for normalized current sequences ix(n),mean(ix) As a sequence of currents ixAverage value of ix(n) is a current sequence ixPoint n of middle, SD (i)x) As a sequence of currents ixThe standard deviation of (a) is determined,for normalized current sequences iyN-th point of (1), iy(n) is a current sequence iyThe nth point in middle, mean (i)y) As a sequence of currents iyAverage value of (d), SD (i)y) As a sequence of currents iyStandard deviation of (d);
step (2), calculating a similar tolerance according to the two groups of current sequences, wherein r is a current sequence ix(n) and iy(n) 0.2 times the covariance, i.e. r 0.2 × COV (i)x(n),iy(n)), wherein,
step (3) defining the distance between the two sequences as Ix(j) And Iy(k) Corresponding to the maximum value of the difference of the currents, the maximum difference expression is as follows:
wherein j is a current sequence IxK is the current sequence IyP is 0 and 1, m is the mode dimension, typically 2, Ix(j + p) is the (j + p) th current sequence, Iy(k + p) is the (k + p) th current sequence;
step (4), according to the set value r of the similarity tolerance, d (I) is counted for each j and kx(j),Iy(k) A number smaller than r) and calculating the ratio of the number of r to the total number of vectors N-m +1, denoted as Cm,rThe value is the approaching probability of the m-dimensional mode in the two sequences under the condition of the similar tolerance r, and the specific expression is as follows:
Cm,r=[d(Ix(j),Iy(k))<number of r]/(N-m+1) (5)
Step (5), for Cm,rObtaining I by calculating the logarithm and then the averagex(j) And Iy(k) Mutual similarity of (D) is denoted as Tm,r[d(Ix(j),Iy(k))]The specific expression is as follows:
and (6) when the embedding dimension is changed into m +1, repeating the steps (1) to (4) to obtain C when the dimension of m +1 is changedm+1R and Tm+1,r[d(Ix(j),Iy(k))]Will Tm,r[d(Ix(j),Iy(k))]And Tm+1,r[d(Ix(j),Iy(k))]And (3) obtaining a final mutual approximation entropy value of the two currents by difference, wherein the expression is as follows:
CApEn(m,r,N)=Tm,r[d(Ix(j),Iy(k))]-Tm+1,r[d(Ix(j),Iy(k))](7)
according to the steps, the mutual approximate entropy value of the two sequences can be obtained.
3. A bridge difference protection anti-misoperation lockout judgment method based on a mutual approximation entropy algorithm is characterized in that the mutual approximation entropy algorithm is adopted to calculate the value of CApEn (m, r, N) between every two phases of a current sequence of a Y/△ converter three-phase TA secondary side, and the result is converted into a value of C.
4. The mutual approximation entropy algorithm is applied to bridge difference protection of the converter station.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910399364.4A CN110071483B (en) | 2019-05-14 | 2019-05-14 | Bridge difference protection anti-misoperation lockout judgment method based on mutual approximation entropy algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910399364.4A CN110071483B (en) | 2019-05-14 | 2019-05-14 | Bridge difference protection anti-misoperation lockout judgment method based on mutual approximation entropy algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110071483A true CN110071483A (en) | 2019-07-30 |
CN110071483B CN110071483B (en) | 2021-02-19 |
Family
ID=67370773
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910399364.4A Active CN110071483B (en) | 2019-05-14 | 2019-05-14 | Bridge difference protection anti-misoperation lockout judgment method based on mutual approximation entropy algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110071483B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112587091A (en) * | 2020-12-04 | 2021-04-02 | 中山大学 | Sleep apnea detection method and device based on heart rate variability angular field image |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5846189A (en) * | 1989-09-08 | 1998-12-08 | Pincus; Steven M. | System for quantifying asynchrony between signals |
CN104578009A (en) * | 2015-01-19 | 2015-04-29 | 国家电网公司 | Current converter bridge differential current protection maloperation prevention method |
US20170005569A1 (en) * | 2015-06-30 | 2017-01-05 | Xj Group Corporation | Differential protection method for bridge circuit in current converter control system |
-
2019
- 2019-05-14 CN CN201910399364.4A patent/CN110071483B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5846189A (en) * | 1989-09-08 | 1998-12-08 | Pincus; Steven M. | System for quantifying asynchrony between signals |
CN104578009A (en) * | 2015-01-19 | 2015-04-29 | 国家电网公司 | Current converter bridge differential current protection maloperation prevention method |
US20170005569A1 (en) * | 2015-06-30 | 2017-01-05 | Xj Group Corporation | Differential protection method for bridge circuit in current converter control system |
Non-Patent Citations (2)
Title |
---|
沈冰等: "基于互近似熵的微电网并网同步检测方法", 《电力系统保护与控制》 * |
翁汉琍等: "HVDC换流站换流器桥差保护动作行为的仿真分析", 《电力科学与技术学报》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112587091A (en) * | 2020-12-04 | 2021-04-02 | 中山大学 | Sleep apnea detection method and device based on heart rate variability angular field image |
Also Published As
Publication number | Publication date |
---|---|
CN110071483B (en) | 2021-02-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Weng et al. | A novel criterion applicable to transformer differential protection based on waveform sinusoidal similarity identification | |
Styvaktakis | Automating power quality analysis | |
US9742181B2 (en) | Method and control system for handling a reclosing operation in a power system | |
Styvaktakis et al. | Signatures of voltage dips: transformer saturation and multistage dips | |
Bera et al. | Discrimination of internal faults and other transients in an interconnected system with power transformers and phase angle regulators | |
Kasztenny et al. | Digital relays improve protection of large transformers | |
CN110071483B (en) | Bridge difference protection anti-misoperation lockout judgment method based on mutual approximation entropy algorithm | |
WO1997043811A1 (en) | Method for on-line dynamic contingency screening of electric power systems | |
Hosseini et al. | An overview of DC microgrid protection schemes and the factors involved | |
Chen | Feature selection for identification and classification of power quality disturbances | |
CN112202153A (en) | Multi-terminal direct current power transmission system locking control method, device, terminal and medium | |
CN109119972B (en) | Locking method and device for preventing differential misoperation caused by CT (current transformer) disconnection of balance winding | |
US8194374B2 (en) | Differential element with harmonic blocking and harmonic restraint operating in parallel | |
Hasheminejad et al. | Transient actions analysis of power transformers based on S‐transform and hidden Markov model | |
CN104977488A (en) | Differential current gradient angle approximate entropy-based transformer magnetizing inrush current identification method | |
CN111049104B (en) | Converter transformer sympathetic inrush current identification method based on positive sequence second harmonic | |
Bera et al. | Intelligent protection & classification of transients in two-core symmetric phase angle regulating transformers | |
CN106058810A (en) | Excitation inrush current identification method based on power fault component criterion | |
Sharygin et al. | The Effect of Combining Information of Relay Protection Fault Detectors1 | |
Niazazari | Robust Event Cause Analysis in Power Grids using Machine Learning Algorithms | |
Fani et al. | Waveform feature monitoring scheme for transformer differential protection | |
Wang | Power Quality Disturbances and Protective Relays: Component Switching and Frequency Deviation | |
Modaresi et al. | New method to determine optimum impedance of fault current limiters for symmetrical and/or asymmetrical faults in power systems | |
Umasankar et al. | Multi wavelet transform with radial basis function neural network for internal faults diagnosis in power transformer | |
Ahmed et al. | Power transformer protection using wavelets and pattern recognition—part I: Theoretical verification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
OL01 | Intention to license declared |