CN110609200A - Power distribution network earth fault protection method based on fuzzy metric fusion criterion - Google Patents
Power distribution network earth fault protection method based on fuzzy metric fusion criterion Download PDFInfo
- Publication number
- CN110609200A CN110609200A CN201910893332.XA CN201910893332A CN110609200A CN 110609200 A CN110609200 A CN 110609200A CN 201910893332 A CN201910893332 A CN 201910893332A CN 110609200 A CN110609200 A CN 110609200A
- Authority
- CN
- China
- Prior art keywords
- fault
- sample
- class
- historical
- representing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a power distribution network earth fault protection method based on fuzzy metric fusion criterion, which comprises the following steps: obtaining a plurality of fault characteristic quantities of a protected feeder line to form a historical sample, and obtaining a sample to be tested according to the same method; clustering n historical samples into a fault class and a non-fault class by adopting a clustering algorithm; taking the membership degree, the distance specific gravity measurement and the angle specific gravity measurement as similarity measurement criteria respectively, calculating a fault metric value and a non-fault metric value of a sample to be measured, and constructing a fuzzy measurement fusion criterion matrix of the sample to be measured; and taking the fuzzy measurement fusion criterion matrix as a judgment set of a judgment index system, constructing a factor set of the judgment index system by using all similarity measurement criteria, presetting weight coefficients of all elements in the factor set, and judging the fuzzy measurement fusion criterion matrix to obtain whether the power distribution network to be detected fails. The invention widens the fault judgment interval, effectively avoids misjudgment caused by factors such as power system oscillation and the like, and improves the robustness.
Description
Technical Field
The invention relates to the technical field of power systems, in particular to a power distribution network ground fault protection method based on fuzzy measurement fusion criteria.
Background
A6-66 kV medium-voltage distribution network in China generally adopts a mode that a neutral point is not grounded and is grounded through an arc suppression coil. When single-phase earth fault happens, the line voltage of the system keeps symmetrical, so that continuous power supply to the load is not influenced, and the system can still continuously operate for 1-2 hours. However, especially for a low-current grounding system, due to low voltage level, weak fault current, and being easily affected by factors such as unstable arc and harmonic interference of load, the grounding protection problem of the feeder line of the power distribution network is not effectively solved, so that the grounding fault detection protection in the low-current grounding system becomes a problem acknowledged in the industry.
As the number of cable lines of the power distribution network increases, the capacitance current gradually increases, and the fault is easy to develop into an interphase fault or a multi-point fault after long-time operation with the fault; the influence of interference factors on fault detection is further enhanced by the wide access of the nonlinear load and the power electronic equipment; arc grounding faults easily cause over-voltage of the whole system, so that a plurality of groups of transformers and switch cabinets are burnt, and equipment and personal safety are endangered. Therefore, a power distribution network ground fault protection method with high precision and strong robustness needs to be researched to ensure safe and reliable operation of a power system.
After long-term efforts of experts and scholars at home and abroad, various power distribution network grounding protection methods are proposed, which can be roughly divided into: a protection method based on a steady-state characteristic criterion, a protection method based on a transient characteristic criterion and an injection signal method. The protection criterion of the method is generated only based on the analysis of the single characteristic quantity, the operation mode of the power distribution network is complex and changeable, the fault condition cannot be predicted, and the single protection criterion cannot cover all grounding working conditions, so that the accuracy of the protection action is not high (only 20% -30%).
In recent years, the rise of the smart Distribution grid (sdg) has greatly promoted the development of advanced Distribution automation (ada). The earth fault protection method based on the intelligent algorithm becomes a research hotspot in the relay protection field, and the methods mainly comprise the following steps: the method comprises a neural network method, a Bayesian method, a genetic algorithm, an intelligent fault protection method based on a rough set theory method and the like. The methods improve the precision and the adaptability of the fault protection scheme to a certain extent by virtue of good data processing capacity, but the forming process of the protection criterion generally lacks a clear physical mechanism, and the fault judgment is completed only by training mass samples. The comprehensive description of the system running state can not be realized under the condition that the running mode is changed, the fault judgment result has one-sidedness, and the self-adaptability requirement of relay protection in the dynamic environment of the power distribution network can not be met.
Disclosure of Invention
Based on the technical problems in the prior art, the invention provides the power distribution network ground fault protection method based on the fuzzy measurement fusion criterion, which can widen the fault judgment interval, effectively avoid misjudgment caused by factors such as power system oscillation and the like, and improve the robustness.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a power distribution network earth fault protection method based on fuzzy metric fusion criterion comprises the following steps:
step 1, constructing a historical sample set, dividing the historical sample set into a fault class and a non-fault class and calculating a clustering center;
a1, acquiring various fault characteristic quantities of a protected feeder line in a known power distribution network operation state to form a history sample; repeating the steps until n historical samples are obtained, and forming a historical sample set by the n historical samples;
step A2, clustering the samples in the historical sample set into fault classes and non-fault classes by adopting a clustering algorithm, and calculating fault class centers and non-fault class centers in the historical sample set;
step 2, constructing an increment sample set, dividing the increment sample set into a fault class and a non-fault class and calculating a clustering center;
b1, acquiring various fault characteristic quantities of the protected feeder line in the running state to be tested according to the A1 to form a sample to be tested; forming an incremental sample set by the sample to be detected and the n historical samples;
step B2, clustering the samples in the incremental sample set into fault classes and non-fault classes by adopting a clustering algorithm, and calculating fault class centers and non-fault class centers in the incremental sample set;
step 3, constructing a fuzzy measurement fusion criterion matrix of the sample to be tested;
step C1, calculating the membership degrees of the samples to be detected relative to the fault class and the non-fault class respectively by using the fault class center and the non-fault class center of the incremental sample set; calculating distance specific gravity measurement of the sample to be measured relative to the fault class and the non-fault class and angle specific gravity measurement of the sample to be measured relative to the fault class and the non-fault class respectively by utilizing the fault class center and the non-fault class center of the historical sample set;
step C2, constructing a fuzzy metric fusion criterion matrix of the sample to be tested according to the membership degree, the distance specific gravity metric and the angle specific gravity metric of the sample to be tested relative to the fault class and the membership degree, the distance specific gravity metric and the angle specific gravity metric of the sample to be tested relative to the non-fault class;
step 4, judging the running state of the protected feeder line;
the fuzzy metric fusion criterion matrix is used as a judgment set of a judgment index system, a factor set of the judgment index system is constructed by using membership, distance proportion measurement and angle proportion measurement, weight coefficients of all elements in the factor set are preset, and the fuzzy metric fusion criterion matrix of a sample to be tested is judged by using the judgment index system to obtain whether the protected feeder line is in fault or not.
According to the scheme, a plurality of fault characteristic quantities of a protected feeder line are extracted to form samples, a large number of samples are clustered and divided, and the membership degree, the distance specific gravity measurement and the angle specific gravity measurement are respectively used as similarity measurement criteria, so that the fault measurement (namely the membership degree, the distance specific gravity measurement and the angle specific gravity measurement relative to the fault class) of 3 samples to be tested relative to the fault class and the non-fault measurement (namely the membership degree, the distance specific gravity measurement and the angle specific gravity measurement relative to the non-fault class) of 3 samples to be tested relative to the non-fault class are calculated, a fuzzy measurement fusion criterion matrix is constructed, and a judgment index system is further adopted to judge the fuzzy measurement fusion criterion matrix, so that whether the power distribution network is in fault or not is known.
Because the fuzzy measurement fusion criterion matrix of the sample to be detected is the fusion of 3 similarity measurement criteria of membership, distance specific gravity measurement and angle specific gravity measurement, the robustness of the power distribution network fault judgment method is improved.
The transverse comparison of the fault degrees of all the feeder lines in the existing power distribution network fault judgment method is converted into the longitudinal comparison that a certain feeder line belongs to a fault class or a non-fault class, and a fault fuzzy measurement fusion criterion is obtained, so that a fault judgment interval is widened, misjudgment caused by factors such as power system oscillation can be effectively avoided, and the robustness is improved.
Furthermore, the evaluation index system is a multi-level evaluation index system, and a factor set formed by membership, distance specific gravity measurement and angle specific gravity measurement is a final-stage factor set of the evaluation index system;
the evaluation rule is as follows: taking the fuzzy measurement fusion criterion matrix as a final-stage criterion matrix, sequentially obtaining the judgment matrix of each stage factor set to the judgment set according to the sequence from the final stage to the first stage, and taking the finally obtained first-stage judgment matrix as a fault measurement comprehensive judgment matrix B1＝(b1,b2) If b is1>b2The power distribution network to be distributed is in a fault state, otherwise the power distribution network to be distributed is in a non-fault state;
wherein, the evaluation matrix B output by the q levelqThe calculation method comprises the following steps: b isq＝Aq·RqAnd R isq-1＝Bq，AqA set of predetermined weighting factors, R, representing the input q-th level and corresponding to the q-th level factor setqMatrix of criteria representing input q-th level。
The multi-stage evaluation index system can weaken the non-fault measurement of the fault sample to be detected, and meanwhile, the fault measurement can be highlighted, so that the fault sample can be still accurately identified under the condition that a fault signal is weak and is influenced by strong interference.
Further, the plurality of fault characteristic quantities include 4 kinds of steady-state characteristic quantities and 3 kinds of transient characteristic quantities, where the 4 kinds of steady-state characteristic quantities are respectively: zero sequence impedance angle, negative sequence current, zero sequence current, earth fault resistance, 3 kinds of transient characteristic quantity are respectively: zero-sequence current db4 wavelet transform post-modulus maximum, cubic B-spline wavelet transform post-modulus maximum, and zero-sequence energy function value in half of power frequency cycle after fault initiation.
By extracting a plurality of fault characteristic quantities of the protected feeder line, the influence of the distortion condition of extracting partial characteristic quantities (caused by the influence of interference factors such as nonlinear load and the like and the distortion of fault signals) can be avoided, and therefore the fault judgment precision is improved.
Further, the evaluation index system is a two-level evaluation index system, and the first-level factor set includes 2 factors: a steady state characteristic quantity and a transient characteristic quantity.
Further, the specific process of step C1 is:
firstly, calculating clustering samples x in the incremental sample set according to the following formulakMembership mu to fault classk′1And membership μ 'to non-failure classes'k2：
In the formula, i represents the category of the cluster, c represents the number of categories and has a value of c-2, i-1 represents a fault class, and i-2 represents a non-fault class; mu's'kiRepresenting a cluster sample xkDegree of membership, p 'to the ith cluster category'1To representFault class hub of incremental sample set, p'2Representing a non-fault class center of the incremental sample set, wherein d is a weighted index;
then, the sample x to be measuredgMembership mu 'to failure class'g1As the first fault measure, the sample x to be measuredgMembership mu 'to non-failed classes'g1As a first non-failure metric.
Further, the specific process of step C2 is:
firstly, the Euclidean distance in the distance discrimination method is selected to measure the sample x to be measuredgWith class i centres piThe distance between:
in the formula: dg1Representing the sample x to be measuredgFault clustering center p with historical sample set1The distance between them; dg2Representing the sample x to be measuredgNon-fault cluster center p with historical sample set2Distance between, xgjRepresenting the sample x to be measuredgJ sample data, pijRepresenting class i centres piThe jth sample data of (1);
then, calculating a second fault metric of the sample to be tested relative to the fault class:and a second non-fault metric with respect to the non-fault class:
further, the specific process of step C3 is:
first, a sample x to be measured is calculatedgWith class i centres piCosine of the included angle therebetween:
in the formula: cos θg1Representing the sample x to be measuredgFault clustering center p with historical sample set1Cosine of the included angle of (c); cos θg2Representing the sample x to be measuredgNon-fault class center p with historical sample set2Cosine of (x)gjRepresenting the sample x to be measuredgJ sample data, pijRepresenting class i centres piThe jth sample data of (1);
then, calculating a third fault metric of the sample to be tested relative to the fault class:and a third non-fault metric with respect to the non-fault class:
further, the kth historical sample x 'obtained in the step 1'kExpressed as: x'k＝(x′k1,x′k2,…,x′ks) (ii) a Wherein, x'k1、…、x′ksIs the kth historical sample x'kS fault characteristic quantities, and the jth fault characteristic quantity is expressed as x'kj(ii) a The history sample set composed of n history samples is as follows: x '═ X'1,x′2,…,x′n}T；
Before step 2, the method further comprises the following steps of normalizing the historical sample set X':
in the formula, xkjThe sample data after normalization processing;the sample mean value of the jth fault characteristic quantity before normalization processing is obtained; s (x')jThe standard deviation is the sample standard deviation of the jth fault characteristic quantity before normalization processing;
after normalization preprocessing, the kth historical sample is obtained and is represented as: x is the number ofk＝(xk1,…,xkj,…,xks) The historical sample set is represented as: x ═ X1,x2,…,xn}T。
Further, step 2, clustering and dividing the n historical samples by adopting a fuzzy c-means clustering algorithm, wherein the specific method comprises the following steps: dynamically clustering all historical samples by means of balanced iterative equations (5) and (6) through an optimized objective function (4) as follows, and obtaining the clustering centers of fault classes and non-fault classes:
in the formula: c is the number of clustering categories, and c is 2; mu.ski∈[0,1]Representing a cluster sample xkMembership degree belonging to the ith cluster typepiIs the cluster center, expressed as: p is a radical ofi＝(pi1,pi2,…,pis) (ii) a Is a matrix norm characterizing the spatial distance between the clustered sample and the cluster center; d is a weighting index, and d is taken to be 2; u denotes the sample x of all clusterskDegree of membership mu ofkiForming a membership matrix: u ═ muki]n×c(ii) a P denotes the center of all clusterspiForming a cluster center matrix P ═ Pi]；JdRepresenting a clustering loss function; mfcRepresenting a cluster sample xkFuzzy c ofThe end conditions of the iterative process of optimizing the objective function (4) are as follows:w is the current iteration number, epsilon is an iteration stop threshold, and epsilon is 1.0 e-6.
Advantageous effects
According to the scheme, multiple fault characteristic quantities of a protected feeder line are extracted to form samples, a large number of samples are clustered and divided, and the membership degree, the distance specific gravity measurement and the angle specific gravity measurement are respectively used as similarity measurement criteria, so that the fault measurement of a sample to be tested 3 relative to a fault class and the non-fault measurement of the sample to be tested 3 relative to a non-fault class are calculated, a fuzzy measurement fusion criterion matrix is constructed, and a judgment index system is further adopted to judge the fuzzy measurement fusion criterion matrix to know whether a power distribution network fails or not.
Because the fuzzy measurement fusion criterion matrix of the sample to be detected is the fusion of 3 similarity measurement criteria, the advantages of different similarity measurement criteria under different fault conditions can be fully exerted, more fault conditions are covered, and the robustness of the power distribution network fault judgment method is improved.
The transverse comparison of the fault degrees of all the feeder lines in the existing power distribution network fault judgment method is converted into the longitudinal comparison that one feeder line belongs to a fault class or a non-fault class, and a fault fuzzy measurement fusion criterion is obtained, so that the advantages of various fault criteria can be effectively fused, the fault judgment interval is widened, misjudgment caused by factors such as power system oscillation can be effectively avoided, and the robustness is improved.
Drawings
FIG. 1 is a schematic diagram of a historical sample set construction method according to the present invention;
FIG. 2 is a schematic diagram of a two-level evaluation index system model according to the present invention;
FIG. 3 is a schematic diagram of a power distribution network ground protection model according to the present invention;
fig. 4 is a schematic diagram of a simulation model of a ground fault of the power distribution system according to the present invention.
Detailed Description
The following describes embodiments of the present invention in detail, which are developed based on the technical solutions of the present invention, and give detailed implementation manners and specific operation procedures to further explain the technical solutions of the present invention.
Step 1, extracting various characteristic quantities of different running states of a power distribution network, and constructing a historical sample set
Under different operation states (including a fault state and a non-fault state) of the power distribution network, by using a distribution automation terminal unit (FTU) installed at an outlet of a protected feeder line, as shown in fig. 4, a plurality of fault characteristic quantities representing the operation state of the feeder line are extracted. The multiple fault characteristic quantities comprise 4 steady-state fault characteristic quantities (respectively comprising a zero-sequence impedance angle, a negative-sequence current, a zero-sequence current and a ground fault resistor) and 3 transient fault characteristic quantities (respectively comprising a zero-sequence db4 wavelet transform post-modulus maximum, a cubic B-spline wavelet transform post-modulus maximum and a zero-sequence energy function value in a half power frequency cycle after the start of the fault). Defining s fault characteristic quantities measured in the kth running state to form a kth historical sample x'k：
x′k＝(x′k1,x′k2,…,x′ks)；
In the formula: x'k1、…、x′ksThe s fault characteristic quantity specific values extracted by the FTU are respectively.
As shown in fig. 1, n history samples are extracted in n operating states, respectively, and a history sample set X '═ { X'1,x′2,…,x′n}T。
Carrying out normalization preprocessing on the historical sample set to obtain:
in the formula, xkjThe jth historical characteristic quantity of the kth historical sample obtained by normalization preprocessing is obtained;the sample mean value of the jth historical characteristic quantity is obtained; s (x'j) Is the sample standard deviation of the jth historical characteristic quantity.
After normalization preprocessing, the kth history sample is represented as:
xk＝(xk1,…,xkj,…,xks)；
the feature index matrix containing n historical samples is represented as:
X＝{x1,x2,…,xn}T。
step 2, fuzzy clustering processing is carried out on the historical samples
And training the historical samples in the characteristic index matrix by adopting a fuzzy c-means clustering algorithm. The process is mathematically described as (X) for the normalized preprocessed feature index matrix Xk∈RsAnd k is 1, …, n), each historical sample is taken as a clustering sample through an optimization objective function (4), dynamic clustering of each historical sample is realized through balance iterative equations (5) and (6), and the clustering center of a fault class and a non-fault class is obtained.
In the formula: c is the number of clustering categories, and c is 2 in the text; mu.sgi∈[0,1]Representing a cluster sample xgMembership degree belonging to the ith cluster typepiIs the cluster center, expressed as: p is a radical ofi＝(pi1,pi2,…,pis) (ii) a I | · | is a matrix norm representing a spatial distance between the clustering samples and the clustering center; d is a weighting index, and d is taken to be 2. U denotes the sample x of all clusterskDegree of membership mu ofkiForming a membership matrix: u ═ muki]n×c(ii) a P denotes the center P of all clustersiForming a cluster center matrix P ═ Pi]；JdRepresenting a clustering loss function; mfcRepresenting a cluster sample xkFuzzy c ofThe end conditions of the iterative process of optimizing the objective function (4) are as follows:w is the current iteration number, epsilon is an iteration stop threshold, and epsilon is 1.0 e-6.
The characteristic quantity of the 2 clustering categories obtained by clustering is very obvious, and a person skilled in the art can directly judge which is the fault clustering center and which is the non-fault clustering center, so that the person skilled in the art can know which category is the fault category and which category is the non-fault category.
Step 3, extracting a sample to be detected
When the power distribution network has a fault, various fault characteristic quantities (4 steady-state fault characteristic quantities, namely zero-sequence impedance angle, negative-sequence current, zero-sequence current and ground fault resistance and 3 transient fault characteristics) are extracted by adopting various methodsQuantity: zero-sequence current db4 wavelet transform post-modulus maximum, cubic B-spline wavelet transform post-modulus maximum and 7 fault characteristic quantities in total of zero-sequence energy function values in half of power frequency cycle wave after fault initiation) to form a sample x to be detectedg。
Step 4, constructing a fuzzy measurement fusion criterion matrix of the sample to be measured according to the sample similarity measurement criterion
Judging whether the sample to be tested is a fault sample or a non-fault sample, quantifying the similarity degree between the sample to be tested and the historical sample, quantitatively describing the similarity relation between the sample to be tested and the historical sample, and obtaining reasonable fault measurement according to the affinity and the sparsity degree of the sample property to form a fuzzy measurement fusion criterion. The invention adopts 3 criteria of membership degree, distance specific gravity measurement and angle specific gravity measurement as fuzzy test fusion criteria.
Criterion one is as follows: membership criterion;
to-be-detected sample xgTaking the n +1 th historical sample, namely taking k-g-n +1, constructing an incremental sample set together with the previous n historical samples, dividing the incremental sample set into a fault class and a non-fault class by adopting the same clustering division method as the historical sample set, and calculating a fault class center p 'of the incremental sample set'1And non-failure class center p'2And calculating the membership degree of the to-be-detected sample relative to the incremental sample set in the fault class and the non-fault class.
Defining a sample x to be testedgDegree of membership to failure class mu'g1For the sample x to be measuredgOf the first failure metric of (2), degree of membership to non-failure class mu'g2For the sample x to be measuredgThe first non-fault metric of (a).
Distance specific gravity measurement criterion;
considering a sample as a defined point in a high-dimensional space, the degree of similarity between samples can be measured by the distance between two points in the high-dimensional space.
Calculating fault class center p of historical sample set X1And a non-fault class center p2Measuring the sample x to be measured by selecting Euclidean distance in distance discriminationgAnd a firstClass i center piThe distance between:
in the formula: dg1Representing the sample x to be measuredgDistance from the fault cluster center; dg2Representing the sample x to be measuredgDistance from non-fault cluster center.
Defining a sample x to be testedgThe second failure metric of (2) is:the second non-failure metric is:
the third criterion is an angle specific gravity measurement criterion;
any two samples xiAnd xjConsidering two vectors with the origin of coordinates as the starting point in the high-dimensional space, the cosine cos theta of the included angle between the vectorsijI.e. an angular similarity measure between samples.
Calculating a sample x to be measuredgVector of representation and class i center piCosine of angle between the represented vectors:
in the formula: cos θg1Representing the sample x to be measuredgCosine of an angle with a fault center; cos θg2Representing the sample x to be measuredgCosine of the angle with the center of the non-fault class.
Defining a sample x to be testedgThe third failure metric of (2) is:the third non-failure metric is:
then, combining 3 similarity measurement criteria of membership degree, distance specific gravity measurement and angle specific gravity measurement of the sample to be measured, and fault measurement values and non-fault measurement values which respectively correspond to the 3 similarity measurement criteria into a fuzzy measurement fusion criterion matrix.
Step 5, setting a multi-stage judgment index system, judging a fuzzy metric fusion criterion matrix, and acquiring whether the power distribution network is in fault;
and setting a multi-stage evaluation index system to evaluate the fuzzy measurement fusion criterion matrix and output a comprehensive criterion matrix, thereby realizing the longitudinal fusion of various fault test criteria of the sample to be tested and the transverse fusion of different types of fault characteristic quantities. In the present invention, a model in which a two-level evaluation index system is set is shown in fig. 2.
Wherein, the multi-level evaluation index system comprises: the system comprises a judgment set, a fault judgment factor set and a weight coefficient set.
Setting of a judgment set:
if V is { V ═ V1,V2Is a set of judgments, where V1Representing a fault metric, V2Representing a non-failure metric; the evaluation set is suitable for all the layers of fault judgment factor sets.
The fault judgment factor set is arranged according to hierarchical levels:
let U be the factor set containing all fault judgment factors, wherein the factors are divided into l groups:
U＝{U1,...,Ut,...,Ul}
in the formula of Ut＝{Ut1,Ut2,...,UtmAnd m represents the number of single factors contained in the t-th group of fault judgment factor sets.
Therefore, each group of the fault judgment factor set is divided into a plurality of layers: u is the highest level factor set, Ut(t 1, 2.., l) is the next highest order factor set, and so on. The number of actual grading layers depends on the particular situation.
The weight coefficient set is correspondingly set according to the hierarchical level of the fault judgment factor set:
let At＝{at1,at2,...,atmIs a set of next higher order factors UtThe weight coefficient set of each single factor pair evaluation set V, atk(t 1, 2.. said., m) is based on the set of next higher order factors UtThe importance degree of the corresponding single factor is distributed to meetA＝{a1,a2,...,amThe weight coefficient set of each single factor pair judgment set V in the U satisfies
Firstly, taking a fuzzy measurement fusion criterion matrix as a judgment set, and solving a comprehensive judgment result of a second-level factor by using a composite algorithm:
Bq＝Aq·Rq＝(bq1,bq2)；
wherein1≤q≤2。
Then, the first-level factor set is judged, and a judgment output matrix B of the second-level factor is utilizedqAnd (3) forming a single factor evaluation matrix R of the first-level factor set U:
R＝[B1,B2,...,BL]T；
therefore, the final comprehensive criterion matrix of the first-level factor set U is:
B＝A·R＝(b1,b2)；
in order to convert fuzzy metric fusion criteria in the fault metric comprehensive criterion matrix into actual protection judgment, the maximum membership judgment criterion is selected as a protection action criterion: b1>b2。
Example (b):
PSCAD/EMTDC simulation software is adopted to carry out simulation analysis on a 35kV power distribution system, and a simulation model is shown in figure 4. The system adopts a mode that a neutral point is grounded through an arc suppression coil, and a 110kV system supplies power to a bus through a delta/Y transformer.
Extracting 3 kinds of steady-state fault characteristic quantities: zero sequence impedance angle xk1Negative sequence current xk2Zero sequence current xk3And 3 kinds of transient fault characteristic quantities: zero sequence current db4 wavelet transform post-modulus maximum value xk4Cubic B-spline wavelet transform post-modulus maximum value xk5Zero sequence energy function value x in half power frequency cycle after fault initiationk6And 6, taking the fault characteristic quantity as the characteristic quantity of the fuzzy clustering analysis. And when the fault occurs on the feeder line 3 and the feeder line 4, the characteristic samples collected at the outlet of the feeder line 4 form a history sample set. For reasons of space, it is preferred that the most representative 16 sets of historical samples are listed in Table 1. Under the condition that the feeder 4 has a single-phase earth fault, a group of fault characteristic quantities are extracted as samples to be tested, and the samples are shown in a table 2.
TABLE 1 historical sample set
TABLE 2 sample of the characteristics of the fault to be tested
And respectively solving fault measurement and non-fault measurement of the similarity measurement criteria of various samples from the steady-state fault characteristic quantity and the transient-state fault characteristic quantity of the sample to be measured, and establishing a fuzzy measurement fusion criterion matrix. As can be seen from the data in table 3, the membership criterion, the angle specific gravity measurement criterion and the distance specific gravity criterion of the transient characteristic quantity all show that the fault measurement of the sample to be measured is greater than the non-fault measurement, and the sample to be measured is accurately judged as the fault sample; and the distance proportion criterion of the steady-state characteristic quantity shows that the sample to be detected is a non-fault sample. The reason is that the extraction process of the fault sample to be detected is influenced by interference factors such as nonlinear load and the like, the fault signal is distorted, the characteristic quantity extraction distortion is caused, and the fault judgment precision is influenced. If the traditional protection scheme is adopted to carry out protection identification on the sample to be detected, the protection judgment fails.
TABLE 3 two-stage index system for judging earth fault of distribution network
And establishing a two-stage evaluation index system to analyze the fuzzy measurement fusion criterion matrix of the fault sample to be tested, and referring to table 3. Weight coefficient sets A and A1The protection judgment accuracy of each level of criterion obtained by a large number of simulation calculations and field experiments is formulated, wherein A is (0.5 ), and A1＝A2＝(0.5,0.3,0.2)。
From the data in table 3, a fuzzy metric fusion criterion matrix can be obtained, which is:
calculate U1The comprehensive evaluation matrix of (1):
B1＝A1·R1＝(0.5145,0.4855)
in the same way, U can be obtained2The comprehensive evaluation matrix of (1):
B2＝A2·R2＝(0.6755,0.3245)
by means of B1、B2Constructing a first-level single-factor evaluation matrix: r ═ B1,B2]TAnd calculating a comprehensive judgment matrix B of the first-level factor set U, namely a fault measurement comprehensive judgment matrix:
B＝A·R＝(0.5950,0.4050)
b is formed by1>b2It can be seen that the fuzzy measurement fusion criterion matrix is corrected through the two-stage judgment index system, and the sample to be detected is accurately judged as a fault sample.
The calculation result shows that the two-stage judgment index system weakens the non-fault measurement of the fault sample to be detected, and simultaneously the fault measurement is highlighted. Under the condition that the fault signal is weak and is influenced by strong interference, the method can accurately identify the fault sample.
The above embodiments are preferred embodiments of the present application, and those skilled in the art can make various changes or modifications without departing from the general concept of the present application, and such changes or modifications should fall within the scope of the claims of the present application.
Claims (9)
1. A power distribution network ground fault protection method based on fuzzy metric fusion criterion is characterized by comprising the following steps:
step 1, constructing a historical sample set, dividing the historical sample set into a fault class and a non-fault class and calculating a clustering center;
a1, acquiring various fault characteristic quantities of a protected feeder line in a known power distribution network operation state to form a history sample; repeating the steps until n historical samples are obtained, and forming a historical sample set by the n historical samples;
step A2, clustering the samples in the historical sample set into fault classes and non-fault classes by adopting a clustering algorithm, and calculating fault class centers and non-fault class centers in the historical sample set;
step 2, constructing an increment sample set, dividing the increment sample set into a fault class and a non-fault class and calculating a clustering center;
b1, acquiring various fault characteristic quantities of the protected feeder line in the running state to be tested according to the A1 to form a sample to be tested; forming an incremental sample set by the sample to be detected and the n historical samples;
step B2, clustering the samples in the incremental sample set into fault classes and non-fault classes by adopting a clustering algorithm, and calculating fault class centers and non-fault class centers in the incremental sample set;
step 3, constructing a fuzzy measurement fusion criterion matrix of the sample to be tested;
step C1, calculating the membership degrees of the samples to be detected relative to the fault class and the non-fault class respectively by using the fault class center and the non-fault class center of the incremental sample set; calculating distance specific gravity measurement of the sample to be measured relative to the fault class and the non-fault class and angle specific gravity measurement of the sample to be measured relative to the fault class and the non-fault class respectively by utilizing the fault class center and the non-fault class center of the historical sample set;
step C2, constructing a fuzzy metric fusion criterion matrix of the sample to be tested according to the membership degree, the distance specific gravity metric and the angle specific gravity metric of the sample to be tested relative to the fault class and the membership degree, the distance specific gravity metric and the angle specific gravity metric of the sample to be tested relative to the non-fault class;
step 4, judging the running state of the protected feeder line;
the fuzzy metric fusion criterion matrix is used as a judgment set of a judgment index system, a factor set of the judgment index system is constructed by using membership, distance proportion measurement and angle proportion measurement, weight coefficients of all elements in the factor set are preset, and the fuzzy metric fusion criterion matrix of a sample to be tested is judged by using the judgment index system to obtain whether the protected feeder line is in fault or not.
2. The method according to claim 1, wherein the evaluation index system is a multi-level evaluation index system, and a factor set consisting of a membership degree, a distance weight metric and an angle weight metric is a final factor set of the evaluation index system;
the evaluation rule is as follows: taking the fuzzy measurement fusion criterion matrix as a final-stage criterion matrix, sequentially obtaining the judgment matrix of each stage factor set to the judgment set according to the sequence from the final stage to the first stage, and taking the finally obtained first-stage judgment matrix as a fault measurement comprehensive judgment matrix B1＝(b1,b2) If b is1>b2The power distribution network to be distributed is in a fault state, otherwise the power distribution network to be distributed is in a non-fault state;
wherein, the evaluation matrix B output by the q levelqThe calculation method comprises the following steps: b isq＝Aq·RqAnd R isq-1＝Bq，AqA set of predetermined weighting factors, R, representing the input q-th level and corresponding to the q-th level factor setqA matrix of criteria representing the input q-th level.
3. The method according to claim 2, wherein the plurality of fault characteristic quantities include 4 kinds of steady-state characteristic quantities and 3 kinds of transient characteristic quantities, and the 4 kinds of steady-state characteristic quantities are respectively: zero sequence impedance angle, negative sequence current, zero sequence current, earth fault resistance, 3 kinds of transient characteristic quantity are respectively: zero-sequence current db4 wavelet transform post-modulus maximum, cubic B-spline wavelet transform post-modulus maximum, and zero-sequence energy function value in half of power frequency cycle after fault initiation.
4. The method of claim 3, wherein the evaluation index system is a two-level evaluation index system, and the first-level factor set comprises 2 factors: a steady state characteristic quantity and a transient characteristic quantity.
5. The method according to claim 1, wherein the specific process of step C1 is as follows:
firstly, calculating clustering samples x in the incremental sample set according to the following formulakMembership mu 'to failure class'k1And membership μ 'to non-failure classes'k2：
In the formula, i represents the category of the cluster, c represents the number of categories and has a value of c-2, i-1 represents a fault class, and i-2 represents a non-fault class; mu's'kiRepresenting a cluster sample xkDegree of membership, p 'to the ith cluster category'1Fault class center, p ', representing incremental sample set'2Representing a non-fault class center of the incremental sample set, wherein d is a weighted index;
then, the sample x to be measuredgMembership mu 'to failure class'g1As the first fault measure, the sample x to be measuredgMembership mu 'to non-failed classes'g1As a first non-failure metric.
6. The method according to claim 1, wherein the specific process of step C2 is as follows:
firstly, the Euclidean distance in the distance discrimination method is selected to measure the sample x to be measuredgWith class i centres piThe distance between:
in the formula: dg1Representing the sample x to be measuredgFault clustering center p with historical sample set1The distance between them; dg2Representing the sample x to be measuredgNon-fault cluster center p with historical sample set2Distance between, xgjRepresenting the sample x to be measuredgJ sample data, pijRepresenting class i centres piThe jth sample data of (1);
then, calculating a second fault metric of the sample to be tested relative to the fault class:and a second non-fault metric with respect to the non-fault class:
7. the method according to claim 1, wherein the specific process of step C3 is as follows:
first, a sample x to be measured is calculatedgWith class i centres piCosine of the included angle therebetween:
in the formula: cos θg1Representing the sample x to be measuredgFault clustering center p with historical sample set1Cosine of the included angle of (c); cos θg2Representing the sample x to be measuredgNon-fault class center p with historical sample set2Cosine of (x)gjRepresenting the sample x to be measuredgJ sample data, pijRepresenting class i centres piThe jth sample data of (1);
then, calculating a third fault metric of the sample to be tested relative to the fault class:and a third non-fault metric with respect to the non-fault class:
8. the method of claim 1, wherein the kth historical sample x 'obtained in step 1'kExpressed as: x'k＝(x′k1,x′k2,…,x′ks) (ii) a Wherein, x'k1、…、x′ksIs the kth historical sample x'kS fault characteristic quantities, and the jth fault characteristic quantity is expressed as x'kj(ii) a The history sample set composed of n history samples is as follows: x '═ X'1,x′2,…,x′n}T；
Before step 2, the method further comprises the following steps of normalizing the historical sample set X':
in the formula, xkjThe sample data after normalization processing;the sample mean value of the jth fault characteristic quantity before normalization processing is obtained; s (x')jThe standard deviation is the sample standard deviation of the jth fault characteristic quantity before normalization processing;
after normalization preprocessing, the kth historical sample is obtained and is represented as: x is the number ofk＝(xk1,…,xkj,…,xks) The historical sample set is represented as: x ═ X1,x2,…,xn}T。
9. The method according to claim 8, wherein the step 2 adopts a fuzzy c-means clustering algorithm to cluster and divide the n historical samples, and the specific method is as follows: dynamically clustering all historical samples by means of balanced iterative equations (5) and (6) through an optimized objective function (4) as follows, and obtaining the clustering centers of fault classes and non-fault classes:
in the formula: c is the number of clustering categories, and c is 2; mu.ski∈[0,1]Representing a cluster sample xkMembership degree belonging to the ith cluster typepiIs the cluster center, expressed as: p is a radical ofi＝(pi1,pi2,…,pis) (ii) a I | · | is a matrix norm representing a spatial distance between the clustering samples and the clustering center; d is a weighting index, and d is taken to be 2; u denotes the sample x of all clusterskDegree of membership mu ofkiForming a membership matrix: u ═ muki]n×c(ii) a P denotes the center P of all clustersiForming a cluster center matrix P ═ Pi]；JdRepresenting a clustering loss function; mfcRepresenting a cluster sample xkFuzzy c ofThe end conditions of the iterative process of optimizing the objective function (4) are as follows:w is the current iteration number, epsilon is an iteration stop threshold, and epsilon is 1.0 e-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910893332.XA CN110609200B (en) | 2019-09-20 | 2019-09-20 | Power distribution network earth fault protection method based on fuzzy metric fusion criterion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910893332.XA CN110609200B (en) | 2019-09-20 | 2019-09-20 | Power distribution network earth fault protection method based on fuzzy metric fusion criterion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110609200A true CN110609200A (en) | 2019-12-24 |
| CN110609200B CN110609200B (en) | 2020-07-24 |
Family
ID=68891717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910893332.XA Active CN110609200B (en) | 2019-09-20 | 2019-09-20 | Power distribution network earth fault protection method based on fuzzy metric fusion criterion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110609200B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111551823A (en) * | 2020-05-14 | 2020-08-18 | 中国南方电网有限责任公司 | Power distribution network fault line selection method and system based on angle similarity |
| CN111596167A (en) * | 2020-05-14 | 2020-08-28 | 中国南方电网有限责任公司 | Fault section positioning method and device based on fuzzy c-means clustering algorithm |
| CN111639583A (en) * | 2020-05-26 | 2020-09-08 | 国网湖南省电力有限公司 | Method and system for identifying power quality disturbance of power grid |
| CN111769529A (en) * | 2020-07-28 | 2020-10-13 | 广东电网有限责任公司 | Grounding protection method |
| CN112684295A (en) * | 2020-12-31 | 2021-04-20 | 国网河南省电力公司电力科学研究院 | Power distribution network fault line selection method and system under high permeability based on similarity separation degree |
| CN112711905A (en) * | 2020-12-23 | 2021-04-27 | 长沙理工大学 | Wind power plant current collection line protection method based on clustering algorithm |
| CN113406537A (en) * | 2020-03-16 | 2021-09-17 | 上海长庚信息技术股份有限公司 | Quantitative evaluation method for fault degree of power equipment |
| CN113674106A (en) * | 2021-08-11 | 2021-11-19 | 国网山东省电力公司平阴县供电公司 | Combined positioning method for ground fault of medium-low voltage distribution network |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101201370A (en) * | 2006-12-13 | 2008-06-18 | 上海海事大学 | Fault diagnosis system adopting circuit information amalgamation and implementing method thereof |
| US8190299B2 (en) * | 2006-07-19 | 2012-05-29 | Rovnyak Steven M | Integrated and optimized distributed generation and interconnect system controller |
| US20130054603A1 (en) * | 2010-06-25 | 2013-02-28 | U.S. Govt. As Repr. By The Secretary Of The Army | Method and apparatus for classifying known specimens and media using spectral properties and identifying unknown specimens and media |
| CN103323749A (en) * | 2013-05-16 | 2013-09-25 | 上海交通大学 | Multi-classifier information fusion partial discharge diagnostic method |
| CN103941163A (en) * | 2014-05-12 | 2014-07-23 | 福州大学 | Resonant earthed system fault line selection method utilizing fuzzy K-means clustering |
| CN109142965A (en) * | 2018-06-29 | 2019-01-04 | 马瑞 | A kind of big data fusion distribution network failure is accurately positioned new method and its device |
| CN109856488A (en) * | 2019-03-15 | 2019-06-07 | 长沙理工大学 | A kind of Transformer State Assessment and fault detection method based on multisource data fusion |
| CN109902953A (en) * | 2019-02-27 | 2019-06-18 | 华北电力大学 | A kind of classification of power customers method based on adaptive population cluster |
-
2019
- 2019-09-20 CN CN201910893332.XA patent/CN110609200B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8190299B2 (en) * | 2006-07-19 | 2012-05-29 | Rovnyak Steven M | Integrated and optimized distributed generation and interconnect system controller |
| CN101201370A (en) * | 2006-12-13 | 2008-06-18 | 上海海事大学 | Fault diagnosis system adopting circuit information amalgamation and implementing method thereof |
| US20130054603A1 (en) * | 2010-06-25 | 2013-02-28 | U.S. Govt. As Repr. By The Secretary Of The Army | Method and apparatus for classifying known specimens and media using spectral properties and identifying unknown specimens and media |
| CN103323749A (en) * | 2013-05-16 | 2013-09-25 | 上海交通大学 | Multi-classifier information fusion partial discharge diagnostic method |
| CN103941163A (en) * | 2014-05-12 | 2014-07-23 | 福州大学 | Resonant earthed system fault line selection method utilizing fuzzy K-means clustering |
| CN109142965A (en) * | 2018-06-29 | 2019-01-04 | 马瑞 | A kind of big data fusion distribution network failure is accurately positioned new method and its device |
| CN109902953A (en) * | 2019-02-27 | 2019-06-18 | 华北电力大学 | A kind of classification of power customers method based on adaptive population cluster |
| CN109856488A (en) * | 2019-03-15 | 2019-06-07 | 长沙理工大学 | A kind of Transformer State Assessment and fault detection method based on multisource data fusion |
Non-Patent Citations (2)
| Title |
|---|
| KUN YU ET.AL: "a fault section location technique for distribution systems based on wide-area fault information fuzzy clustering analysis", 《2014 INTERNATIONAL CONFERENCE ON POWER SYSTEMS TECHNOLOGY》 * |
| 简金宝 等: "基于***聚类分析的馈线接地保护", 《电力***自动化》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113406537A (en) * | 2020-03-16 | 2021-09-17 | 上海长庚信息技术股份有限公司 | Quantitative evaluation method for fault degree of power equipment |
| CN113406537B (en) * | 2020-03-16 | 2024-05-07 | 上海长庚信息技术股份有限公司 | Quantitative evaluation method for fault degree of power equipment |
| CN111551823A (en) * | 2020-05-14 | 2020-08-18 | 中国南方电网有限责任公司 | Power distribution network fault line selection method and system based on angle similarity |
| CN111596167A (en) * | 2020-05-14 | 2020-08-28 | 中国南方电网有限责任公司 | Fault section positioning method and device based on fuzzy c-means clustering algorithm |
| CN111639583A (en) * | 2020-05-26 | 2020-09-08 | 国网湖南省电力有限公司 | Method and system for identifying power quality disturbance of power grid |
| CN111769529A (en) * | 2020-07-28 | 2020-10-13 | 广东电网有限责任公司 | Grounding protection method |
| CN112711905A (en) * | 2020-12-23 | 2021-04-27 | 长沙理工大学 | Wind power plant current collection line protection method based on clustering algorithm |
| CN112684295A (en) * | 2020-12-31 | 2021-04-20 | 国网河南省电力公司电力科学研究院 | Power distribution network fault line selection method and system under high permeability based on similarity separation degree |
| CN113674106A (en) * | 2021-08-11 | 2021-11-19 | 国网山东省电力公司平阴县供电公司 | Combined positioning method for ground fault of medium-low voltage distribution network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110609200B (en) | 2020-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110609200B (en) | Power distribution network earth fault protection method based on fuzzy metric fusion criterion | |
| CN110336534B (en) | Fault diagnosis method based on photovoltaic array electrical parameter time series feature extraction | |
| CN106055918B (en) | Method for identifying and correcting load data of power system | |
| CN112699913A (en) | Transformer area household variable relation abnormity diagnosis method and device | |
| CN111596167A (en) | Fault section positioning method and device based on fuzzy c-means clustering algorithm | |
| CN112149873B (en) | Low-voltage station line loss reasonable interval prediction method based on deep learning | |
| CN112964959A (en) | Power distribution network earth fault line selection method based on correlation coefficient measurement criterion | |
| CN109921415B (en) | Hybrid measurement-oriented online defense method for power grid malignant data injection attack | |
| CN112147462A (en) | Power transmission line fault identification method based on deep learning | |
| CN110927527A (en) | Transient fault sensing method based on transient energy double channels and information fusion | |
| CN112684295A (en) | Power distribution network fault line selection method and system under high permeability based on similarity separation degree | |
| CN111612326A (en) | Comprehensive evaluation method for power supply reliability of distribution transformer | |
| CN111654392A (en) | Low-voltage distribution network topology identification method and system based on mutual information | |
| CN113552443A (en) | Hybrid connection high-voltage line fault identification method based on alternating current-direct current pulse and random forest | |
| CN116011825A (en) | Multi-dimensional evaluation method for operation risk of distribution cable line | |
| CN111551823A (en) | Power distribution network fault line selection method and system based on angle similarity | |
| CN111062569A (en) | Low-current fault discrimination method based on BP neural network | |
| CN112508254B (en) | Method for determining investment prediction data of transformer substation engineering project | |
| CN116990633A (en) | Fault studying and judging method based on multiple characteristic quantities | |
| CN109784777B (en) | Power grid equipment state evaluation method based on time sequence information fragment cloud similarity measurement | |
| CN111275292A (en) | Ground grid state evaluation method based on fuzzy analytic hierarchy process | |
| CN112881827B (en) | Oil-immersed transformer fault diagnosis method based on improved grey correlation analysis | |
| Dehghani et al. | Distribution feeder classification based on self organized maps (case study: Lorestan province, Iran) | |
| CN115184728A (en) | Fault identification method for AC/DC hybrid power distribution network | |
| CN113092934B (en) | Single-phase earth fault judgment method and system based on clustering and LSTM |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230823 Address after: Room 201, Jinhongyuan, No. 229 Tongzipo West Road, Changsha High tech Development Zone, Changsha City, Hunan Province, 410221 Patentee after: Changsha Jingke Electric Technology Co.,Ltd. Address before: 410114 No. 960, Section 2, Wanjiali South Road, Tianxin District, Changsha City, Hunan Province Patentee before: CHANGSHA University OF SCIENCE AND TECHNOLOGY |