CN116008721A - Power distribution network fault positioning method and terminal - Google Patents
Power distribution network fault positioning method and terminal Download PDFInfo
- Publication number
- CN116008721A CN116008721A CN202211566956.9A CN202211566956A CN116008721A CN 116008721 A CN116008721 A CN 116008721A CN 202211566956 A CN202211566956 A CN 202211566956A CN 116008721 A CN116008721 A CN 116008721A
- Authority
- CN
- China
- Prior art keywords
- signal
- distribution network
- eigenmode
- power distribution
- fault
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Locating Faults (AREA)
Abstract
The invention discloses a power distribution network fault positioning method and a terminal, wherein a power distribution network system topological structure is constructed, and current signals among all sub-node terminals are collected according to the power distribution network system topological structure; the current signal is compressed and collected to obtain a compressed signal; reconstructing the compressed signal to obtain a reconstructed compressed signal, and decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigenmode functions; the fuzzy C-means clustering algorithm is used for carrying out clustering analysis based on the series of eigen mode functions to obtain fault point position information, signal sampling can be completed at a lower sampling rate, the effect of useful information of signals is fully exerted, redundancy of the information is reduced, better decomposition effect can be obtained by using CEEMDAN for signal decomposition, a fault line can be rapidly and accurately positioned by using the fuzzy C-means clustering algorithm, and therefore fault point positioning is accurately and efficiently achieved.
Description
Technical Field
The invention relates to the technical field of power distribution network fault detection, in particular to a power distribution network fault positioning method and a terminal.
Background
In recent years, with the change of ecological environment and the need of the overall layout of ecological civilization construction of the country, the development of new energy has become a focus of attention. The distribution network is an important component of the power grid, is a junction for connecting energy production and consumption, and is also a key component for constructing a novel power system. The large-scale distributed new energy, different energy storage and other devices are connected into the power grid, so that the power supply performance of the power grid is improved, meanwhile, the production operation mode of the power distribution network is deeply influenced, and the traditional distribution network protection mode is further improved. The distribution network is directly connected with the power grid terminal of a user and is an important public infrastructure for service folks, so that the requirements on the aspects of electric energy quality, power supply safety and the like are more strict. According to statistics, more than 80% of user power failures are caused by distribution network faults, wherein single-phase earth faults account for more than 80%, for a medium-voltage distribution network, if single-phase earth faults occur, a system can still keep normal operation in a short time, but in a novel distribution network, fault current is possibly increased due to the introduction of distributed new energy, if fault points are not isolated as soon as possible, fault expansion is easily caused, therefore, the position of the fault points is quickly, accurately and reliably determined, and fault isolation and fault restoration sections are extremely important to recover normal power supply as soon as possible.
At present, the method for locating the fault point of the power distribution network mainly comprises a direct method and an indirect method, the direct algorithm mainly comprises a matrix algorithm and the like, the fault location of the power distribution network is carried out by adopting the matrix algorithm, a description matrix representing a power distribution network model is formed by utilizing a power distribution network topological structure, a fault position discrimination matrix is obtained after a series of operations, and the point location of the fault point is obtained, but whether the algorithm is successful in location or not is directly related to the acquisition information of a sensor, and the fault tolerance capability on the fault information is poor. In the indirect algorithm application of power distribution network fault location, an improved bionic electromagnetic method is adopted, a solution model is built based on a switching function, and a fault section is used as a solution space to obtain an optimal solution. The method has the advantages of good fault tolerance, various types and high updating speed, but has the defects of low operation speed, easy sinking into local optimum and the like. In addition, the convolutional neural network is adopted to train the fault data of the power distribution network, the neural network has strong operation capability, the data can be processed when more and more complex data are obtained, but when the convolutional neural network is applied to fault location of the power distribution network with the distributed power supply, the algorithm is not easy to converge due to the characteristic that the output of the distributed power supply is not fixed and can be thrown and withdrawn at any time, and a great amount of data training is needed in the early stage of fault location of the power distribution network based on the neural network, so that the calculation amount is large. In the fault point detection method based on current signal processing, the existing signal acquisition process mainly comprises four processes of sampling, compressing, transmitting and decompressing, the sampling process must follow the Shannon-Nyquist sampling theorem, if the process is adopted for fault current data of the power distribution network, a large number of data stacks can be generated, the redundancy of information is increased, the processing time is prolonged, the transmission bandwidth and the storage space of a circuit are occupied, and high-efficiency data utilization and high-speed transmission can be severely restricted.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the power distribution network fault positioning method and the terminal can accurately and efficiently position fault points.
In order to solve the technical problems, the invention adopts a technical scheme that:
a power distribution network fault positioning method comprises the following steps:
constructing a power distribution network system topological structure, and collecting current signals among all child node terminals according to the power distribution network system topological structure;
the current signal is compressed and collected to obtain a compressed signal;
reconstructing the compressed signal to obtain a reconstructed compressed signal, and decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigenmode functions;
and carrying out cluster analysis by using a fuzzy C-means clustering algorithm based on the series of eigen mode functions to obtain fault point position information.
In order to solve the technical problems, the invention adopts another technical scheme that:
a power distribution network fault location terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
constructing a power distribution network system topological structure, and collecting current signals among all child node terminals according to the power distribution network system topological structure;
the current signal is compressed and collected to obtain a compressed signal;
reconstructing the compressed signal to obtain a reconstructed compressed signal, and decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigenmode functions;
and carrying out cluster analysis by using a fuzzy C-means clustering algorithm based on the series of eigen mode functions to obtain fault point position information.
The invention has the beneficial effects that: the current signals are compressed and collected to obtain compressed signals, the compressed signals are reconstructed, the reconstructed compressed signals are decomposed by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigen mode functions, a fuzzy C-means clustering algorithm is used for clustering analysis based on the series of eigen mode functions to obtain fault point position information, the current signals are sampled by using a compressed sampling method, the signal sampling can be completed at a lower sampling rate, the effect of useful information of the signals is fully exerted, the redundancy of the information is reduced, the modal aliasing problem existing in EMD can be restrained by using an adaptive noise complete set empirical mode decomposition algorithm (CEEMDAN), a better decomposition effect is obtained, and finally a fault line can be rapidly and accurately positioned by using the fuzzy C-means clustering algorithm, so that the positioning of the fault point is accurately and efficiently realized.
Drawings
FIG. 1 is a flowchart illustrating steps of a method for locating a fault in a power distribution network according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a fault location terminal of a power distribution network according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a network architecture of a power distribution network in a power distribution network fault location method according to an embodiment of the present invention;
fig. 4 is a schematic diagram of fault identification and fault isolation corresponding to a regional network architecture in a fault location method of a power distribution network according to an embodiment of the present invention.
Detailed Description
In order to describe the technical contents, the achieved objects and effects of the present invention in detail, the following description will be made with reference to the embodiments in conjunction with the accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides a method for locating a fault of a power distribution network, including the steps of:
constructing a power distribution network system topological structure, and collecting current signals among all child node terminals according to the power distribution network system topological structure;
the current signal is compressed and collected to obtain a compressed signal;
reconstructing the compressed signal to obtain a reconstructed compressed signal, and decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigenmode functions;
and carrying out cluster analysis by using a fuzzy C-means clustering algorithm based on the series of eigen mode functions to obtain fault point position information.
From the above description, the beneficial effects of the invention are as follows: the current signals are compressed and collected to obtain compressed signals, the compressed signals are reconstructed, the reconstructed compressed signals are decomposed by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigen mode functions, a fuzzy C-means clustering algorithm is used for clustering analysis based on the series of eigen mode functions to obtain fault point position information, the current signals are sampled by using a compressed sampling method, the signal sampling can be completed at a lower sampling rate, the effect of useful information of the signals is fully exerted, the redundancy of the information is reduced, the modal aliasing problem existing in EMD can be restrained by using an adaptive noise complete set empirical mode decomposition algorithm (CEEMDAN), a better decomposition effect is obtained, and finally a fault line can be rapidly and accurately positioned by using the fuzzy C-means clustering algorithm, so that the positioning of the fault point is accurately and efficiently realized.
Further, the compressing and collecting the current signal to obtain a compressed signal includes:
determining a discrete cosine transform basis as a sparse transform basis matrix, and initializing a sparse transform dictionary;
according to the current signal, solving sparse representation by using an orthogonal matching pursuit algorithm to obtain a sparse representation coefficient;
iteratively updating the sparse conversion dictionary to obtain an updated sparse conversion dictionary;
performing sparse transformation on the current signals according to the updated sparse transformation dictionary to obtain sparse signals;
determining a matrix which is uncorrelated with all column vectors in the sparse conversion base matrix and accords with Gaussian distribution as a measurement matrix;
and obtaining a compressed signal according to the measurement matrix and the sparse signal.
As can be seen from the above description, the compressed acquisition is a concept of low sampling rate data acquisition, and the sampling is completed at a frequency far lower than the nyquist sampling theorem, so that the information redundancy and the network load are reduced, and the efficiency of fault point positioning is improved.
Further, the sparse representation is solved by using an orthogonal matching pursuit algorithm according to the current signal, so as to obtain a sparse representation coefficient
Comprising the following steps:
I j,t ＝{I j,1 ,I j,2 ,K,I j,N }；
wherein I is j,t The current signal acquired by the terminal j at the time t is represented, D represents a sparse conversion dictionary, L represents the number of the terminals, and N represents the acquisition time length;
the current signal is subjected to sparse transformation according to the updated sparse transformation dictionary to obtain a sparse signal I s,t The method comprises the following steps:
the compressed signal R is obtained according to the measurement matrix and the sparse signal, and is:
R＝Φ*I s,t ；
where Φ represents the measurement matrix.
From the above description, since the current signal does not have sparsity or is not obvious in sparsity in the time domain, sparse transformation is required to be performed on the original current signal, and the signal after sparse transformation can be conveniently observed subsequently after being processed by the measurement matrix, so that the fault point is positioned.
Further, the decomposing the reconstructed compressed signal by using the adaptive noise complete set empirical mode decomposition algorithm to obtain an eigenmode function includes:
adding Gaussian white noise into the reconstructed compressed signal and performing adaptive noise complete set empirical mode decomposition to obtain a series of initial eigenmode functions;
calculating arithmetic average of the series of initial eigenmode functions to obtain a first eigenmode function;
removing the first eigenmode function from the current signal to obtain a latest signal, and returning to execute the step of adding Gaussian white noise into the reconstructed compressed signal according to the latest signal and performing adaptive noise complete set empirical mode decomposition until a unique residual error and a series of eigenmode functions which do not meet the decomposition conditions are obtained.
According to the description, based on the characteristic that fault point currents do not have similarity, the reconstructed compressed signals are decomposed by using an adaptive noise complete set empirical mode decomposition algorithm, so that the signal characteristics can be enhanced, and the accuracy of subsequent fault point positioning is improved.
Further, the Gaussian white noise is added to the reconstructed compressed signal and adaptive noise complete set empirical mode decomposition is performed, so that a series of initial eigenmode functions are obtained as follows:
in the method, in the process of the invention,
representing the reconstructed compressed signal, ε k Represents the standard deviation of Gaussian white noise, N i (t) Gaussian white noise with unit variance mean of 0, < >>
Representing a series of initial eigenmode functions, +.>
Representing signal residuals, m representing the number of groups of gaussian white noise, q representing the number of initial eigenmode functions;
the arithmetic average is calculated on the series of initial eigenmode functions, and a first eigenmode function is obtained as follows:
in the IMF 1 Representing a first eigenmode function.
As can be seen from the above description, the method can be used to decompose the signals, so as to effectively extract useful information in the signals, and facilitate the judgment of subsequent fault points.
Further, performing cluster analysis by using a fuzzy C-means clustering algorithm based on the series of eigen-mode functions, and obtaining fault point location information includes:
removing false modes in the series of eigenmode functions by using normalized energy entropy to obtain recombined eigenmode functions;
and performing cluster analysis by using a fuzzy C-means clustering algorithm according to the recombined eigenmode function to obtain fault point position information.
As can be seen from the description, due to the problems of a large amount of noise and false modes generated by early decomposition after the CEEMDAN algorithm is decomposed, the false modes are further removed by adopting the normalized energy entropy, so that the information contained in the signal modes is more accurate, and the accuracy of fault point positioning is improved.
Further, the removing the false modes in the series of eigenmode functions by using the normalized energy entropy, and obtaining the recombined eigenmode functions includes:
calculating the energy of the series of eigenmode functions;
normalizing the energy to obtain normalized energy;
calculating normalized energy entropy values corresponding to the series of eigen-mode functions according to the normalized energy;
and determining an eigenmode function with the normalized energy entropy value smaller than a first preset threshold value in the series of eigenmode functions as a false mode, and removing the false mode to obtain a recombined eigenmode function.
From the above description, the false modes can be removed quickly and effectively by using the normalized energy entropy, so that the reliability of the data is improved.
Further, performing cluster analysis by using a fuzzy C-means clustering algorithm according to the recombined eigenmode function, and obtaining fault point position information includes:
determining the recombined eigen mode function as a sample, and initializing a membership matrix;
determining a clustering center of each sample subset in the sample;
determining an objective function of a fuzzy C-means clustering algorithm and a constraint condition corresponding to the objective function based on the clustering center of each sample subset;
obtaining the iteration change amount of the objective function, and comparing the iteration change amount with a second preset threshold value;
if the iteration change amount is greater than or equal to the second preset threshold value, updating the membership matrix to obtain an updated membership matrix, and returning to execute the step of determining the clustering center of each sample subset according to the updated membership matrix;
if the iteration change amount is smaller than the second preset threshold value, acquiring the class with the largest membership degree, and classifying the recombined eigen-mode function into the class with the largest membership degree;
determining an upstream detection point and a downstream detection point adjacent to the fault point according to the membership matrix;
calculating the difference value of adjacent membership according to the upstream detection point and the downstream detection point;
and judging whether the difference value is larger than a third preset threshold value, if so, determining that the section between the detection points corresponding to the adjacent membership degrees is the position of the fault point.
The above description shows that the fault point can be positioned efficiently and accurately, so that the distribution network substation can cut off the fault line timely and accurately, and the line safety of the distribution network is protected.
Further, the determining a cluster center of each subset of the samples is:
wherein, c i Represents the cluster center of the ith sample, n represents the total number of samples, X j Representing the recombined eigenmode functions,
representing a sample x calculated using a membership index h j Membership belonging to class i;
the target function of the fuzzy C-means clustering algorithm and the constraint conditions corresponding to the target function are determined based on the clustering centers of the sample subsets:
wherein q represents the number of cluster centers, x j Representing the recombined eigenmode function X j The j-th sample in (b), u ij Representing sample x j Belonging to class i membership.
According to the description, the position of the fault point can be accurately determined based on the objective function and the constraint condition thereof, and a worker can conveniently execute closing and opening operations on the sectionalizer of the fault upstream node according to the position information, so that the safety of the whole power distribution network is protected, and the working stability of the power distribution network is improved.
Referring to fig. 2, another embodiment of the present invention provides a fault location terminal for a power distribution network, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements each step in the fault location method for the power distribution network when executing the computer program.
The power distribution network fault positioning method and the terminal can be suitable for the scene of fault detection of the power distribution network, and the power distribution network fault positioning method and the terminal are described in the following specific embodiments:
example 1
Referring to fig. 1 and fig. 3-4, a fault location method for a power distribution network in this embodiment includes the steps of:
s1, constructing a power distribution network system topological structure, and collecting current signals among all child node terminals according to the power distribution network system topological structure;
the topology structure of the power distribution network system defines the upper and lower relationship between the power distribution network substation and the terminal.
S2, compressing and collecting the current signal to obtain a compressed signal, which specifically comprises the following steps:
s21, determining a discrete cosine transform basis as a sparse transform basis matrix, and initializing a sparse transform dictionary;
s22, solving sparse representation by using an orthogonal matching pursuit algorithm (Orthogonal Matching Pursuit, OMP) according to the current signal to obtain a sparse representation coefficient
Specific:
I j,t ＝{I j,1 ,I j,2 ,K,I j,N }；
wherein I is j,t The current signal acquired by the terminal j at the time t is represented, D represents a sparse conversion dictionary, L represents the number of the terminals, and N represents the acquisition time length;
s23, carrying out iterative updating on the sparse conversion dictionary to obtain an updated sparse conversion dictionary;
specifically, each atom in the sparse conversion dictionary is subjected to iterative updating, and the updated sparse conversion dictionary can be obtained after k times of iterative solving.
S24, performing sparse transformation on the current signals according to the updated sparse transformation dictionary to obtain sparse signals I s,t Specific:
s25, determining a matrix which is irrelevant to all column vectors in the sparse conversion base matrix and accords with Gaussian distribution as a measurement matrix;
the sparse transform basis matrix ψ is:
r∈{0,...,M-1}；
wherein R is M×M Representing the base matrix of MxM order,
representing each index +.>
I representing the rows of the sparse transform basis matrix, r representing the columns of the sparse transform basis matrix, M representing the dimension of the M x M-order sparse transform basis matrix, C representing the coefficients;
wherein, when i=0,
when i+.0, +.>
In an alternative embodiment, the validity of the measurement matrix is also checked by using an equidistant condition, so as to judge whether the original signal can be effectively reconstructed after the generated random Gaussian matrix is used as the measurement matrix for projection. If the condition is satisfied, regenerating a random matrix satisfying the Gaussian distribution. The reason for using a random gaussian matrix instead of directly using a gaussian matrix is a certain randomness and a certain adaptation. S26, obtaining a compressed signal R according to the measurement matrix and the sparse signal, and specifically:
R＝Φ*I s,t ；
where Φ represents the measurement matrix.
S3, reconstructing the compressed signal to obtain a reconstructed compressed signal, and decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigenmode functions, wherein the method specifically comprises the following steps of:
s31, reconstructing the compressed signal to obtain a reconstructed compressed signal;
the reconstructed compressed signal
The method comprises the following steps:
s32, adding Gaussian white noise into the reconstructed compressed signal and performing adaptive noise complete set empirical mode decomposition to obtain a series of initial eigenmode functions, wherein the specific steps are as follows:
in the method, in the process of the invention,
representing the reconstructed compressed signal, ε k Represents the standard deviation of Gaussian white noise, N i (t) Gaussian white noise with unit variance mean of 0, < >>
Representing a series of initial eigenmode functions, +.>
Representing signal residuals, m representing the number of groups of gaussian white noise, q representing the number of initial eigenmode functions;
s33, calculating arithmetic average of the initial eigenmode functions to obtain a first eigenmode function, and specifically:
in the IMF 1 Representing a first eigenmode function.
S34, removing the first eigenmode function from the current signal to obtain a latest signal, and returning to execute S32-S33 according to the latest signal until a unique residual error and a series of eigenmode functions which do not meet the decomposition conditions are obtained.
Specifically, from the current signal I j,t Removing IMF in 1 Part, get the latest signal v, namely order
And then returning to S32-S33 until a unique residual R (t) and a series of eigen-mode functions which do not meet the decomposition conditions are obtained. />
For example, after obtaining the first eigenmode function, the signal residual r is calculated 1 (t) is:
then at r 1 (t) adding Gaussian white noise to perform S32-S33, then calculating the second residual error, and so on k And r k (t) until the decomposition condition cannot be met, terminating the decomposition to obtain a residual R (t), and writing a corresponding signal:
the decomposition is finished to obtain n m-dimensional modal energy sequences, which are expressed as: x is X n (t)＝{IMF 1 ,IMF 2 ,...,IMF N }；
Wherein the decomposition conditions are as follows:
wherein N is z Represents the number of extreme points, N e Representing the number of zeros, fmax (t) representing the upper envelope of the signal, fmin (t) representing the lower envelope of the signal;
specifically, curve fitting is carried out on maximum value and minimum value points respectively by adopting a cubic spline difference algorithm, an upper envelope corresponding to the maximum value points and a lower envelope corresponding to the minimum value points are obtained, an average value of the maximum value points and the lower envelope corresponding to the minimum value points is calculated and recorded as m (t), y (t) is the difference between a signal and m (t), and whether y (t) meets the decomposition condition is judged.
S4, performing cluster analysis by using a fuzzy C-means clustering algorithm based on the series of eigen mode functions to obtain fault point position information, wherein the method specifically comprises the following steps:
s41, eliminating false modes in the series of eigenmode functions by using normalized energy entropy to obtain recombined eigenmode functions, wherein the method specifically comprises the following steps of:
s411, calculating energy E (X) of the series of eigenmode functions, specifically:
wherein X is n (t) represents the series of eigenmode functions.
S412, carrying out normalization processing on the energy to obtain normalized energy p (n), and specifically:
EN n ＝-p(n)log 2 (p (n)); s414, determining the eigenmode function with the normalized energy entropy value smaller than a first preset threshold value in the series of eigenmode functions as a false mode, and eliminating the false mode to obtain the recombined eigenmode function.
The energy spectrum of the signal can represent the relative relation of the energy occupied by each state variable in the whole system, the influence of noise interference in the signal is small, the eigenmode function sensitive to the characteristic information of the original signal occupies main energy, and the energy occupation proportion of the false mode is small. Wherein the first preset threshold is 0.1;
the recombined eigenmode function X j The method comprises the following steps:
X j ＝{IMF 1 ,IMF 2 ,...,IMF Z }；
s42, performing cluster analysis by using a fuzzy C-means clustering algorithm according to the recombined eigen mode function to obtain fault point position information, wherein the method specifically comprises the following steps:
s421, determining the recombined eigen mode function as a sample, and initializing a membership matrix;
wherein, the sum of membership degrees of the satisfied sample set is 1.
S422, determining a clustering center of each sample subset in the samples, and specifically:
wherein, c i Represents the cluster center of the ith sample, n represents the total number of samples, X j Representing the recombined eigenmode functions,
representing a sample x calculated using a membership index h j Membership belonging to class i;
s423, determining an objective function of a fuzzy C-means clustering algorithm and a constraint condition corresponding to the objective function based on the clustering center of each sample subset, and specifically:
wherein q represents the number of cluster centers, x j Representing the recombined eigenmode function X j The j-th sample in (b), u ij Representing sample x j Belonging to class i membership.
Wherein u is obtained by Lagrangian multiplier method ij And c i Is interrelated, and u is obtained by iterative operation ij And c i Values of (2)。
S424, acquiring the iteration change amount of the objective function, and comparing the iteration change amount with a second preset threshold;
s425, if the iteration change amount is greater than or equal to the second preset threshold value, updating the membership matrix to obtain an updated membership matrix, and returning to execute the step S422 according to the updated membership matrix;
wherein, the updated membership matrix U is:
s426, if the iteration change amount is smaller than the second preset threshold value, acquiring a class with the largest membership degree, and classifying the recombined eigen-mode function into the class with the largest membership degree;
s427, determining an upstream detection point and a downstream detection point adjacent to the fault point according to the membership matrix;
s428, calculating the difference value of adjacent membership degrees according to the upstream detection point and the downstream detection point;
s429, judging whether the difference value is larger than a third preset threshold value, if so, determining that the section between the detection points corresponding to the adjacent membership degrees is the fault point position, otherwise, not the fault point.
Wherein the third preset threshold is 0.1;
the distribution substation can execute closing and opening operations on the sectionalizer of the node upstream of the fault point according to the position of the fault point.
As shown in fig. 3, fig. 3 is a schematic diagram of a network architecture of a distribution network in a certain area. The architecture is characterized in that a distribution network area is formed by a plurality of distribution rooms/switching stations which are directly and electrically connected, a distribution network substation is deployed in the area and used for realizing the control function of a master station, and the rest distribution rooms/switching stations are used as intelligent terminal service objects.
As shown in fig. 4, fig. 4 is a diagram illustrating the corresponding fault recognition and fault isolation under the regional network architecture. And collecting current signals among the terminal nodes by using a current sensor among the regional centralized distribution network nodes for compressive sampling, transmission reconstruction and cluster analysis. When the line breaks down, the distribution network substation receives the clustering analysis result, controls the recloser of the current stage to be disconnected, simultaneously sends a blocking signal of the sectionalizer to an upper node where the fault occurs, the node which receives the blocking signal controls the sectionalizer of the current stage to be blocked, and the node which does not receive the blocking signal controls the sectionalizer of the current stage to be disconnected, and the recloser is closed after time delay, so that the power supply of a non-fault area is quickly restored.
The relationship between the nodes shown in fig. 4 is: the distribution substation has 8 terminal nodes which are subordinate to T1-T8, and the point B between T3-T4 is a section with faults. After the distribution network substation receives the result of the fault signal, firstly, the T1 controls the K1-1 recloser to be disconnected, the distribution network substation issues a node sectionalizer locking signal, after receiving the sectionalizer locking signal, the T1 and T2 nodes control the sectionalizers K1-2 and K2-2 to complete locking operation, and the T3 controls the sectionalizer K3-1 to be disconnected, and after time delay, the recloser is closed, so that the power supply of a non-fault area is quickly restored.
Example two
Referring to fig. 2, a fault location terminal for a power distribution network in this embodiment includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements each step in the fault location method for the power distribution network in the first embodiment when executing the computer program.
In summary, according to the power distribution network fault positioning method and the terminal provided by the invention, a power distribution network system topology structure is constructed, and current signals among all sub-node terminals are collected according to the power distribution network system topology structure; the current signal is compressed and collected to obtain a compressed signal; reconstructing the compressed signal to obtain a reconstructed compressed signal, and decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigenmode functions; the method comprises the steps of carrying out clustering analysis by using a fuzzy C-means clustering algorithm based on a series of eigen mode functions to obtain fault point position information, specifically, carrying out signal decomposition by using a self-adaptive noise complete set empirical mode decomposition algorithm to inhibit the mode aliasing problem existing in EMD (empirical mode decomposition) to obtain better decomposition effect, and finally, rapidly and accurately positioning a fault line by using the fuzzy C-means clustering algorithm to accurately and efficiently realize the positioning of fault points.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent changes made by the specification and drawings of the present invention, or direct or indirect application in the relevant art, are included in the scope of the present invention.
Claims (10)
1. The power distribution network fault positioning method is characterized by comprising the following steps:
constructing a power distribution network system topological structure, and collecting current signals among all child node terminals according to the power distribution network system topological structure;
the current signal is compressed and collected to obtain a compressed signal;
reconstructing the compressed signal to obtain a reconstructed compressed signal, and decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain a series of eigenmode functions;
and carrying out cluster analysis by using a fuzzy C-means clustering algorithm based on the series of eigen mode functions to obtain fault point position information.
2. The method for locating a fault in a power distribution network according to claim 1, wherein the compressing and collecting the current signal to obtain a compressed signal includes:
determining a discrete cosine transform basis as a sparse transform basis matrix, and initializing a sparse transform dictionary;
according to the current signal, solving sparse representation by using an orthogonal matching pursuit algorithm to obtain a sparse representation coefficient;
iteratively updating the sparse conversion dictionary to obtain an updated sparse conversion dictionary;
performing sparse transformation on the current signals according to the updated sparse transformation dictionary to obtain sparse signals;
determining a matrix which is uncorrelated with all column vectors in the sparse conversion base matrix and accords with Gaussian distribution as a measurement matrix;
and obtaining a compressed signal according to the measurement matrix and the sparse signal.
3. The power distribution network fault location method according to claim 2, wherein the sparse representation is solved by using an orthogonal matching pursuit algorithm according to the current signal, so as to obtain a sparse representation coefficient
Comprising the following steps:
I j,t ＝{I j,1 ,I j,2 ,K,I j,N }；
wherein I is j,t The current signal acquired by the terminal j at the time t is represented, D represents a sparse conversion dictionary, L represents the number of the terminals, and N represents the acquisition time length;
the current signal is subjected to sparse transformation according to the updated sparse transformation dictionary to obtain a sparse signal I s,t The method comprises the following steps:
the compressed signal R is obtained according to the measurement matrix and the sparse signal, and is:
R＝Φ*I s,t ；
where Φ represents the measurement matrix.
4. The method for locating a fault in a power distribution network according to claim 1, wherein the decomposing the reconstructed compressed signal by using an adaptive noise complete set empirical mode decomposition algorithm to obtain an eigenmode function includes:
adding Gaussian white noise into the reconstructed compressed signal and performing adaptive noise complete set empirical mode decomposition to obtain a series of initial eigenmode functions;
calculating arithmetic average of the series of initial eigenmode functions to obtain a first eigenmode function;
removing the first eigenmode function from the current signal to obtain a latest signal, and returning to execute the step of adding Gaussian white noise into the reconstructed compressed signal according to the latest signal and performing adaptive noise complete set empirical mode decomposition until a unique residual error and a series of eigenmode functions which do not meet the decomposition conditions are obtained.
5. The method for locating a fault in a power distribution network according to claim 4, wherein the adding gaussian white noise to the reconstructed compressed signal and performing adaptive noise complete set empirical mode decomposition to obtain a series of initial eigenmode functions is as follows:
in the method, in the process of the invention,
representing the reconstructed compressed signal, ε k Represents the standard deviation of Gaussian white noise, N i (t) represents the mean of unit variancesGaussian white noise of 0, +.>
Representing a series of initial eigenmode functions, +.>
Representing signal residuals, m representing the number of groups of gaussian white noise, q representing the number of initial eigenmode functions;
the arithmetic average is calculated on the series of initial eigenmode functions, and a first eigenmode function is obtained as follows:
in the IMF 1 Representing a first eigenmode function.
6. The method for locating faults in a power distribution network according to claim 1, wherein the step of performing cluster analysis by using a fuzzy C-means clustering algorithm based on the series of eigen-mode functions to obtain fault point location information includes:
removing false modes in the series of eigenmode functions by using normalized energy entropy to obtain recombined eigenmode functions;
and performing cluster analysis by using a fuzzy C-means clustering algorithm according to the recombined eigenmode function to obtain fault point position information.
7. The method for locating a fault in a power distribution network according to claim 6, wherein said removing false modes in said series of eigenmode functions using normalized energy entropy, and obtaining the recombined eigenmode functions comprises:
calculating the energy of the series of eigenmode functions;
normalizing the energy to obtain normalized energy;
calculating normalized energy entropy values corresponding to the series of eigen-mode functions according to the normalized energy;
and determining an eigenmode function with the normalized energy entropy value smaller than a first preset threshold value in the series of eigenmode functions as a false mode, and removing the false mode to obtain a recombined eigenmode function.
8. The method for locating faults in a power distribution network according to claim 6, wherein the step of performing cluster analysis by using a fuzzy C-means clustering algorithm according to the recombined eigenmode functions to obtain fault point location information comprises the following steps:
determining the recombined eigen mode function as a sample, and initializing a membership matrix;
determining a clustering center of each sample subset in the sample;
determining an objective function of a fuzzy C-means clustering algorithm and a constraint condition corresponding to the objective function based on the clustering center of each sample subset;
obtaining the iteration change amount of the objective function, and comparing the iteration change amount with a second preset threshold value;
if the iteration change amount is greater than or equal to the second preset threshold value, updating the membership matrix to obtain an updated membership matrix, and returning to execute the step of determining the clustering center of each sample subset according to the updated membership matrix;
if the iteration change amount is smaller than the second preset threshold value, acquiring the class with the largest membership degree, and classifying the recombined eigen-mode function into the class with the largest membership degree;
determining an upstream detection point and a downstream detection point adjacent to the fault point according to the membership matrix;
calculating the difference value of adjacent membership according to the upstream detection point and the downstream detection point;
and judging whether the difference value is larger than a third preset threshold value, if so, determining that the section between the detection points corresponding to the adjacent membership degrees is the position of the fault point.
9. The method for locating faults in a power distribution network of claim 8, wherein said determining a cluster center for each subset of samples is:
wherein, c i Represents the cluster center of the ith sample, n represents the total number of samples, X j Representing the recombined eigenmode functions,
representing a sample x calculated using a membership index h j Membership belonging to class i;
the target function of the fuzzy C-means clustering algorithm and the constraint conditions corresponding to the target function are determined based on the clustering centers of the sample subsets:
wherein q represents the number of cluster centers, x j Representing the recombined eigenmode function X j The j-th sample in (b), u ij Representing sample x j Belonging to class i membership.
10. A power distribution network fault location terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of a power distribution network fault location method as claimed in any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211566956.9A CN116008721A (en) | 2022-12-07 | 2022-12-07 | Power distribution network fault positioning method and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211566956.9A CN116008721A (en) | 2022-12-07 | 2022-12-07 | Power distribution network fault positioning method and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116008721A true CN116008721A (en) | 2023-04-25 |
Family
ID=86033160
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211566956.9A Pending CN116008721A (en) | 2022-12-07 | 2022-12-07 | Power distribution network fault positioning method and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116008721A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117031194A (en) * | 2023-07-27 | 2023-11-10 | 中电鼎润(广州)科技发展有限公司 | Ultrasonic hidden danger detection method and system for power distribution network |
| CN117349735A (en) * | 2023-12-05 | 2024-01-05 | 国家电投集团云南国际电力投资有限公司 | Fault detection method, device and equipment for direct-current micro-grid and storage medium |
-
2022
- 2022-12-07 CN CN202211566956.9A patent/CN116008721A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117031194A (en) * | 2023-07-27 | 2023-11-10 | 中电鼎润(广州)科技发展有限公司 | Ultrasonic hidden danger detection method and system for power distribution network |
| CN117031194B (en) * | 2023-07-27 | 2024-04-09 | 中电鼎润(广州)科技发展有限公司 | Ultrasonic hidden danger detection method and system for power distribution network |
| CN117349735A (en) * | 2023-12-05 | 2024-01-05 | 国家电投集团云南国际电力投资有限公司 | Fault detection method, device and equipment for direct-current micro-grid and storage medium |
| CN117349735B (en) * | 2023-12-05 | 2024-03-26 | 国家电投集团云南国际电力投资有限公司 | Fault detection method, device and equipment for direct-current micro-grid and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116008721A (en) | Power distribution network fault positioning method and terminal | |
| CN109635928B (en) | Voltage sag reason identification method based on deep learning model fusion | |
| Li et al. | Location identification of power line outages using PMU measurements with bad data | |
| CN111080105A (en) | Transformer area user-to-user relationship identification method and system based on voltage time sequence data | |
| Kezunovic | Translational knowledge: From collecting data to making decisions in a smart grid | |
| CN111598166A (en) | Single-phase earth fault classification method and system based on principal component analysis and Softmax function | |
| Zhao et al. | An improved complex ICA based method for wind farm harmonic emission levels evaluation | |
| CN113219328B (en) | Intelligent fault diagnosis method for circuit breaker operating mechanism based on information fusion | |
| CN111123028A (en) | Active low-voltage distribution network fault location optimization method based on hierarchical model | |
| CN112289329A (en) | High-voltage circuit breaker fault diagnosis method based on GWO-KFCM | |
| CN115219838A (en) | Power distribution network fault studying and judging method based on multi-source information fusion | |
| CN112819649A (en) | Method and device for determining station area subscriber change relationship | |
| CN117110798B (en) | Fault detection method and system for intelligent power distribution network | |
| Sima et al. | Edge–cloud collaboration detection approach for small-sample imbalanced faults in power lines | |
| CN115877068A (en) | Voltage sag propagation track identification method of regional power grid based on deep learning | |
| CN111222800A (en) | Intelligent power system anomaly detection method | |
| Bhuiyan et al. | A deep learning through DBN enabled transmission line fault transient classification framework for multimachine microgrid systems | |
| CN116298699A (en) | Power distribution network fault identification method based on wavelet optimization and graph convolution neural network | |
| CN107241740B (en) | Island mobile electric power system health state monitoring sensor deployment method | |
| CN113988205B (en) | Method and system for judging electric precipitation working condition | |
| CN116388148A (en) | Wind power prediction method, device, equipment and storage medium | |
| CN113537330B (en) | Non-invasive load identification method | |
| CN113435351A (en) | Method, system and equipment for diagnosing mechanical fault of vacuum switch and readable storage medium | |
| CN106019043B (en) | Power grid fault diagnosis method based on fuzzy integral multi-source information fusion theory | |
| CN115456013A (en) | Wind turbine generator converter power module fault diagnosis method based on operation data |
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 |