CN112163012A - Power grid line fault phase selection method and device, electronic equipment and storage medium - Google Patents

Power grid line fault phase selection method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN112163012A
CN112163012A CN202010928620.7A CN202010928620A CN112163012A CN 112163012 A CN112163012 A CN 112163012A CN 202010928620 A CN202010928620 A CN 202010928620A CN 112163012 A CN112163012 A CN 112163012A
Authority
CN
China
Prior art keywords
fault
grid line
gradient
phase selection
information sequences
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
Application number
CN202010928620.7A
Other languages
Chinese (zh)
Other versions
CN112163012B (en
Inventor
韩佶
苗世洪
牛荣泽
孙芊
李丰君
徐恒博
李宗峰
郭祥富
张建宾
郭舒毓
殷浩然
王子欣
赵健
谢芮芮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
State Grid Corp of China SGCC
State Grid Henan Electric Power Co Ltd
Electric Power Research Institute of State Grid Henan Electric Power Co Ltd
Original Assignee
Huazhong University of Science and Technology
State Grid Corp of China SGCC
State Grid Henan Electric Power Co Ltd
Electric Power Research Institute of State Grid Henan Electric Power Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology, State Grid Corp of China SGCC, State Grid Henan Electric Power Co Ltd, Electric Power Research Institute of State Grid Henan Electric Power Co Ltd filed Critical Huazhong University of Science and Technology
Priority to CN202010928620.7A priority Critical patent/CN112163012B/en
Publication of CN112163012A publication Critical patent/CN112163012A/en
Application granted granted Critical
Publication of CN112163012B publication Critical patent/CN112163012B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2474—Sequence data queries, e.g. querying versioned data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/26—Visual data mining; Browsing structured data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Abstract

The invention discloses a power grid line fault phase selection method, a device, electronic equipment and a storage medium, belonging to the field of power grid fault diagnosis, wherein the method comprises the following steps: s1, acquiring a plurality of fault information sequences corresponding to the power grid line during the fault period, wherein the fault information sequences comprise a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence; s2, acquiring gradient similarity of sampling points in every two fault information sequences, and constructing gradient similarity matrixes corresponding to the fault information sequences; s3, carrying out visualization processing on the gradient similarity matrix to construct a visualization graph corresponding to the plurality of fault information sequences; and S4, inputting the visual graph into a convolutional neural network model to perform fault phase selection on the power grid line during the fault. The method omits the step of marking the occurrence time of the fault, simplifies the phase selection process of the fault of the power grid line, and simultaneously improves the accuracy of the phase selection of the fault of the power grid line.

Description

Power grid line fault phase selection method and device, electronic equipment and storage medium
Technical Field
The invention belongs to the field of power grid fault diagnosis, and particularly relates to a power grid line fault phase selection method, a device, electronic equipment and a storage medium.
Background
After the short-circuit fault occurs to the power grid line, the fault phase is timely and accurately judged, and the method has important significance for timely clearing the fault and guaranteeing the safe and stable operation of the power grid.
In recent years, deep learning techniques, particularly Convolutional Neural Network (CNN) techniques, have been developed rapidly and are widely used in the fields of power grid fault diagnosis, optimized operation, and the like.
However, in the prior art, when a neural network model is used for phase selection of a power grid line fault, the fault occurrence time must be accurately marked, and once an error exists in the mark of the fault occurrence time, a fault phase judgment error easily occurs.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides a power grid line fault phase selection method, a device, electronic equipment and a storage medium, and aims to simplify the fault identification process and improve the probability of fault phase identification, so that the technical problems of low accuracy and complex operation flow of power grid line fault phase selection by using a neural network model in the prior art are solved.
To achieve the above object, according to one aspect of the present invention, a method, an apparatus, an electronic device and a storage medium for selecting a phase of a grid line fault are provided.
A grid line fault phase selection method comprises the following steps:
s1, acquiring a plurality of fault information sequences corresponding to the power grid line during the fault period, wherein the fault information sequences comprise a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence;
s2, acquiring gradient similarity of sampling points in every two fault information sequences, and constructing gradient similarity matrixes corresponding to the fault information sequences;
s3, carrying out visualization processing on the gradient similarity matrix to construct a visualization graph corresponding to the plurality of fault information sequences;
and S4, inputting the visual graph into a convolutional neural network model to perform fault phase selection on the power grid line during the fault.
In one embodiment, the step S2 specifically includes:
s201, acquiring gradient vectors of sampling points in every two fault information sequences;
s202, acquiring the Manhattan distance between every two gradient vectors to identify the similarity of every two gradient vectors; and constructing the gradient similarity matrix by taking the Manhattan distance of each gradient vector as an element in the gradient similarity matrix.
In one embodiment, the step S201 specifically includes:
any two fault information sequences thetaiAnd thetaj，i,j＝1,2,…,m，θiAnd thetajThe corresponding time series are respectively recorded as
Figure BDA0002669374380000021
And
Figure BDA0002669374380000022
obtained by
Figure BDA0002669374380000023
And
Figure BDA0002669374380000024
the gradient vectors of the middle sampling points are respectively:
Figure BDA00026693743800000216
wherein the content of the first and second substances,
Figure BDA0002669374380000025
is thetaiCorresponding to
Figure BDA0002669374380000026
The (n) th sampling point of (c),
Figure BDA0002669374380000027
is thetajCorresponding to
Figure BDA0002669374380000028
N is 1, …, N;
Figure BDA0002669374380000029
is composed of
Figure BDA00026693743800000210
Gradient of the nth sample point;
Figure BDA00026693743800000211
is composed of
Figure BDA00026693743800000212
Gradient of the nth sample point, [ theta ]1-θmThe sampling interval of (a) is delta t, and m is the number of the fault information sequences;
the step S202 specifically includes:
manhattan distance MD for obtaining every two gradient vectorsi,jDescription of the invention
Figure BDA00026693743800000213
And
Figure BDA00026693743800000214
similarity between:
Figure BDA00026693743800000215
constructing the gradient similarity matrix GS-matrix by using the acquired respective Manhattan distances:
Figure BDA0002669374380000031
in one embodiment, the step S3 specifically includes:
s301, mapping each Manhattan distance in the gradient similarity matrix into a corresponding color level;
s302, constructing the visual graph by using each color level corresponding to the gradient similarity matrix.
In one embodiment, the step S301 specifically includes:
mapping each Manhattan distance into a corresponding color level according to a preset relationship, wherein the preset relationship is as follows:
Figure BDA0002669374380000032
wherein the content of the first and second substances,
Figure BDA0002669374380000033
is MDi,jThe color rank of (d); INT is a rounded symbol;
Figure BDA0002669374380000034
and
Figure BDA0002669374380000035
the maximum and minimum values of all elements in the GS-matrix, respectively.
In one embodiment, before the step S4, the method for selecting a phase of the grid line fault further includes:
setting the input layer structure of the convolutional neural network model to be the same as the size structure of the visual graph;
and setting an output layer structure of the convolutional neural network model according to the number of the fault phases.
In one embodiment, the step S1 specifically includes:
s101, acquiring a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence of the head end and the tail end of each phase on the power grid line during the fault period;
and S102, taking the voltage signal time series, the current signal time series, the active power time series and the reactive power time series of the head end and the tail end of each phase as the plurality of fault information series.
A grid line fault phase selection apparatus comprising:
the system comprises a sequence acquisition module, a fault detection module and a fault detection module, wherein the sequence acquisition module is used for acquiring a plurality of fault information sequences corresponding to a power grid line during a fault period, and the fault information sequences comprise a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence;
the matrix construction module is used for acquiring the gradient similarity of sampling points in every two fault information sequences and constructing gradient similarity matrixes corresponding to the fault information sequences;
the graph construction module is used for carrying out visualization processing on the gradient similarity matrix so as to construct a visualization graph corresponding to the plurality of fault information sequences;
and the fault phase selection module is used for inputting the visual graph into a convolutional neural network model so as to perform fault phase selection on the power grid line during the fault period.
An electronic device comprising a memory storing a computer program and a processor implementing the steps of the method described above when the processor executes the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
Generally speaking, compared with the prior art, the power grid line fault phase selection method provided by the invention can accurately perform fault phase selection on the power grid line without marking the fault occurrence time. The method and the device have the advantages that the step of marking the occurrence time of the fault is omitted, the phase selection process of the power grid line fault is simplified, meanwhile, the accuracy of the phase selection of the power grid line fault is improved, and the follow-up processing of the power grid line fault is facilitated.
Drawings
Fig. 1 is a flowchart of a phase selection method for a grid line fault in an embodiment of the present application;
FIG. 2 is a schematic diagram of a grid line in an embodiment of the present application;
FIG. 3 shows an embodiment of the present application under fault
Figure BDA0002669374380000041
(Δ T ═ 0.3s) schematic;
FIG. 4 is a schematic diagram of a GS-image construction process according to an embodiment of the present application;
FIG. 5 is a block diagram of a convolutional neural network model in an embodiment of the present application;
FIG. 6 is a graph comparing the accuracy of the phase selection method for line faults of the power grid provided by the present application with the conventional CNN method;
FIG. 7 is a flowchart of step S2 in an embodiment of the present application;
FIG. 8 is a flowchart of step S3 in an embodiment of the present application;
fig. 9 is a schematic structural diagram of a phase selection device for a grid line fault in an embodiment of the present application;
fig. 10 is a schematic diagram of an internal structure of an electronic device in an embodiment of the present application.
The same reference numbers will be used throughout the drawings to refer to the same or like elements or structures, wherein:
a sequence acquisition module 901, a matrix construction module 902, a graph construction module 903 and a fault phase selection module 904.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
The application provides a power grid line fault phase selection method, as shown in fig. 1, the power grid line fault phase selection method includes: step S1 to step S4.
And S1, acquiring a plurality of fault information sequences corresponding to the power grid lines during the fault period, wherein the fault information sequences comprise a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence. In one embodiment, the head end voltage signal time series, the current signal time series, the active power time series and the reactive power time series of each phase on the grid line during the fault are collected respectively, and the tail end voltage signal time series, the current signal time series, the active power time series and the reactive power time series are also collected respectively. For example, referring to fig. 2, the grid line has two ends, namely a head end and a tail end, and the time series of voltage, current, active power and reactive power at the head end and the tail end of the grid line during the fault are respectively obtained, and the time series of 4 signals is obtained in total, so as to perform the following visualization processing and fault phase selection. Each grid line has 3 phases, so the head end of the grid line can acquire a time series of 3 × 4 ═ 12 signals, the tail end of the grid line can acquire a time series of 3 × 4 ═ 12 signals, and a total of 12+12 ═ 24 signals. And taking the voltage signal time series, the current signal time series, the active power time series and the reactive power time series of the head end and the tail end of each phase as a plurality of fault information series. Note that these 24 signals are theta1-θ24And is recorded by theta1-θ24The corresponding fault information sequence is
Figure BDA0002669374380000061
Wherein N is
Figure BDA0002669374380000062
The number of sampling points of (a) and, at the same time,
Figure BDA0002669374380000063
all correspond to each otherThe inter-length deltat.
And S2, acquiring the gradient similarity of the sampling points in each two fault information sequences, and constructing a gradient similarity matrix corresponding to the plurality of fault information sequences.
Specifically, the gradient vectors of the sampling points in each fault information sequence may be obtained to obtain a plurality of gradient vectors, and then the manhattan distance between every two gradient vectors is calculated to identify the similarity between every two gradient vectors; and constructing a gradient similarity matrix by taking the Manhattan distance of each gradient vector as an element in the gradient similarity matrix.
Taking 24 fault information sequences as an example for description, first, any two θ are pointed outiAnd thetaj(i, j-1, 2, …,24) in time series
Figure BDA0002669374380000064
And
Figure BDA0002669374380000065
let θ be1-θ24Is Δ t, then
Figure BDA0002669374380000066
And
Figure BDA0002669374380000067
the gradient at each sample point in (a) is:
Figure BDA0002669374380000068
in the above formula, the first and second carbon atoms are,
Figure BDA0002669374380000069
and
Figure BDA00026693743800000610
are respectively as
Figure BDA00026693743800000611
And
Figure BDA00026693743800000612
the gradient of each of the sampling points in (c),
Figure BDA00026693743800000613
and
Figure BDA00026693743800000614
are all vectors;
Figure BDA00026693743800000615
is composed of
Figure BDA00026693743800000616
N is 1, …, N;
Figure BDA00026693743800000617
is composed of
Figure BDA00026693743800000618
The (n) th sampling point of (c),
Figure BDA00026693743800000619
is composed of
Figure BDA00026693743800000620
Gradient of the nth sample point;
Figure BDA00026693743800000621
is composed of
Figure BDA00026693743800000622
Gradient of the nth sample point. Wherein N is
Figure BDA00026693743800000623
The number of sampling points of (a) and, at the same time,
Figure BDA00026693743800000624
corresponds to a time length deltat. Taking Δ T to 0.3s, since the sampling frequency is 1000Hz in the embodiment, N to 0.3 × 1000 to 300 is obtained, as shown in fig. 3, which shows that a failure occurs at a certain time
Figure BDA00026693743800000625
In which the part enclosed by the black frame is Δ T0.3 s
Figure BDA00026693743800000626
Further, Manhattan distance description is used
Figure BDA00026693743800000627
And
Figure BDA00026693743800000628
is similar to that of
Figure BDA0002669374380000071
Wherein, MDi,jDescribed by using Manhattan distance
Figure BDA0002669374380000072
And
Figure BDA0002669374380000073
the similarity between them. Further, a gradient similarity matrix GS-matrix is constructed, i.e.
Figure BDA0002669374380000074
And S3, performing visualization processing on the gradient similarity matrix to construct a visualization graph corresponding to the plurality of fault information sequences.
Specifically, the GS-matrix is visualized, i.e., the GS-image is constructed. In GS-image, the gradation of a pixel in a pattern is an integer in the range of 0 to 255, for a total of 256 gradations. Defining a mapping relation between elements in the GS-matrix and pixel tone levels in the GS-image, e.g.
Figure BDA0002669374380000075
In the above formula, the first and second carbon atoms are,
Figure BDA0002669374380000076
is MDi,jThe color rank of (d); INT is a rounded symbol;
Figure BDA0002669374380000077
and
Figure BDA0002669374380000078
the maximum and minimum values of all elements in the GS-matrix, respectively. And finally, the construction of the gradient similarity visualization graph GS-image is realized. As shown in fig. 4, the faults in all 3 subgraphs occur at 0s, and the time window Δ T of the three subgraphs is 0.3 s. It can be seen that the start points of the three sub-graph time windows are different, but the generated GS-images are very similar. Therefore, the proposed method does not need to mark the time of occurrence of the fault, as long as the time window can wrap the electrical signal for a period of time after the fault occurs, a similar GS-image can be generated and used for subsequent phase selection based on the CNN fault.
And S4, inputting the visual graph into the convolutional neural network model to perform fault phase selection on the power grid line during the fault.
Specifically, the visual pattern is input into a trained convolutional neural network model, and the convolutional neural network model can identify the visual pattern and then output an identification result, namely a fault phase selection result. The fault phases are various, and the convolutional neural network model is used for identifying the visual graph and selecting one fault phase from the various fault phases. For example, the fault phase may be of class 10, such as a-phase short-circuit ground, B-phase short-circuit ground, C-phase short-circuit ground, AB-phase short-circuit ground, AC-phase short-circuit ground, BC-phase short-circuit ground, AB-phase short-circuit, AC-phase short-circuit, BC-phase short-circuit, ABC-phase short-circuit ground.
In one embodiment, before step S4, the method for selecting a phase of a grid line fault further includes: the input layer structure of the convolutional neural network model is set to be the same as the size structure of the visual graph. And setting an output layer structure of the convolutional neural network model according to the number of the fault phases. If the size of the GS-image is larger than other parameters such as M × M, the structure of the input layer of the convolutional neural network model is set to be M × M, for example, for the structure of the input layer of the convolutional neural network model, the size of the GS-image is 24 × 24, so the structure of the input layer of the CNN is set to be 24 × 24. The structure of the output layer is set according to the number of categories of the faulty phase selection, for example, the structure of the CNN output layer is set to 10 × 1 because there are 10 categories in the faulty phase selection problem. In the present embodiment, the internal structure of CNN is shown in fig. 5. In addition, fig. 5 is a schematic structural diagram of the CNN in this embodiment.
Parameter(s) Numerical value
Convolution kernel size in first convolutional layer 7×7
Pooled core size in first pooled layer 2×2
Number of convolution/pooling kernels in first convolution/pooling layer 3
Convolution kernel size in the second convolutional layer 4×4
Pooled core size in second pooled layer 2×2
Number of convolution/pooling kernels in second convolution/pooling layer 6
TABLE 1 CNN internal structural parameters
And further, taking GS-image as the input of the CNN, wherein the output of the CNN is the fault phase selection result of the power grid line. As shown in fig. 6, when the error of the time stamp of the fault is 0, both the proposed method and the conventional CNN method obtain a more accurate phase selection result of the fault; when a fault time marking error occurs, only the method is accurate; furthermore, the larger the error of the time stamp of the failure, the lower the accuracy of the conventional method, while the proposed method is substantially unaffected.
In one embodiment, as shown in fig. 7, step S2 specifically includes: s201, acquiring gradient vectors of sampling points in each fault information sequence; s202, acquiring the Manhattan distance between every two gradient vectors to identify the similarity of every two gradient vectors; and constructing a gradient similarity matrix by taking the Manhattan distance of each gradient vector as an element in the gradient similarity matrix.
In one embodiment, step S201 specifically includes:
any two fault information sequences thetaiAnd thetaj(i,j＝1,2,…,m)，θiAnd thetajThe corresponding time series are respectively recorded as
Figure BDA0002669374380000091
And
Figure BDA0002669374380000092
obtained by
Figure BDA0002669374380000093
And
Figure BDA0002669374380000094
the gradient vectors of the middle sampling points are respectively:
Figure BDA0002669374380000095
wherein the content of the first and second substances,
Figure BDA0002669374380000096
is thetaiCorresponding to
Figure BDA0002669374380000097
N is 1, …, N;
Figure BDA0002669374380000098
is thetajCorresponding to
Figure BDA0002669374380000099
(ii) is sampled at the nth sampling point (N-1, …, N);
Figure BDA00026693743800000910
is composed of
Figure BDA00026693743800000911
Gradient of the nth sample point;
Figure BDA00026693743800000912
is composed of
Figure BDA00026693743800000913
Gradient of the nth sample point, [ theta ]1-θmIs Δ t, and m is the number of fault information sequences.
Step S202 specifically includes:
manhattan distance MD for obtaining any two gradient vectorsi,jFor the description of
Figure BDA00026693743800000914
And
Figure BDA00026693743800000915
similarity between:
Figure BDA00026693743800000916
and then constructing a gradient similarity matrix GS-matrix by using the acquired Manhattan distances:
Figure BDA00026693743800000917
in one embodiment, as shown in fig. 8, step S3 specifically includes: s301, mapping each Manhattan distance in the gradient similarity matrix into a corresponding color level. And S302, constructing a visual graph by using each color level corresponding to the gradient similarity matrix.
Specifically, the gradient similarity matrix is visualized by mapping each element in the matrix to one of the color levels 0 to 255, and finally mapping the matrix to a graph, that is, converting the gradient similarity matrix to a visualized graph, wherein the visualized graph carries fault state information, and identifying the visualized graph can obtain a fault phase.
In one embodiment, step S301 specifically includes: mapping each Manhattan distance into a corresponding color level according to a preset relationship, wherein the preset relationship is as follows:
Figure BDA0002669374380000101
wherein the content of the first and second substances,
Figure BDA0002669374380000102
is MDi,jThe color rank of (d); INT is a rounded symbol;
Figure BDA0002669374380000103
and
Figure BDA0002669374380000104
the maximum and minimum values of all elements in the GS-matrix, respectively.
Specifically, the GS-matrix is visualized, i.e., the GS-image is constructed. In the GS-image, the gradation of pixels in the graphic is in the range of 0 to 255For a total of 256 levels. Defining the mapping relation between the elements in the GS-matrix and the pixel tone scale in the GS-image, i.e. defining
Figure BDA0002669374380000105
Wherein the content of the first and second substances,
Figure BDA0002669374380000106
is MDi,jThe color rank of (d); INT is a rounded symbol;
Figure BDA0002669374380000107
and
Figure BDA0002669374380000108
the maximum and minimum values of all elements in the GS-matrix, respectively. The above equation finally realizes the construction of the gradient similarity visualization graph GS-image.
An embodiment of the present application further provides a power grid line fault phase selection device, as shown in fig. 9, the power grid line fault phase selection device includes: a sequence acquisition module 901, a matrix construction module 902, a graph construction module 903 and a fault phase selection module 904. The sequence obtaining module 901 is configured to obtain a plurality of fault information sequences corresponding to a power grid line during a fault period, where the fault information sequences include a voltage signal time sequence, a current signal time sequence, an active power time sequence, and a reactive power time sequence. The matrix construction module 902 is configured to obtain gradient similarity between sampling points in each two fault information sequences, and construct a gradient similarity matrix corresponding to the multiple fault information sequences. And the shape construction module 903 is configured to perform visualization processing on the gradient similarity matrix to construct a visualization graph corresponding to the multiple fault information sequences. And a fault phase selection module 904, configured to input the visualization graph into the convolutional neural network model to perform fault phase selection on the power grid line during the fault.
In one embodiment, the matrix construction module 902 is configured to obtain a gradient vector of a sampling point in each fault information sequence; acquiring the Manhattan distance of every two gradient vectors to identify the similarity of every two gradient vectors; and constructing a gradient similarity matrix by taking the Manhattan distance of each gradient vector as an element in the gradient similarity matrix.
In one embodiment, the graph building module 903 is configured to map each manhattan distance in the gradient similarity matrix to a corresponding tone scale; and constructing a visual graph by using each color level corresponding to the gradient similarity matrix.
The division of each module in the grid line fault phase selection device is only used for illustration, and in other embodiments, the grid line fault phase selection device may be divided into different modules as needed to complete all or part of the functions of the grid line fault phase selection device.
For specific limitations of the phase selection device for the grid line fault, reference may be made to the above limitations of the phase selection method for the grid line fault, and details are not described here. All or part of each module in the grid line fault phase selection device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the electronic device, and can also be stored in a memory of the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Fig. 10 is a schematic diagram of an internal structure of an electronic device in one embodiment. As shown in fig. 10, the electronic device includes a processor and a memory connected by a system bus. Wherein, the processor is used for providing calculation and control capability and supporting the operation of the whole electronic equipment. The memory may include a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The computer program is executable by a processor for implementing a method for grid line fault phase selection provided in the following embodiments. The internal memory provides a cached execution environment for the operating system computer programs in the non-volatile storage medium.
The implementation of each module in the grid line fault phase selection device provided in the embodiment of the present application may be in the form of a computer program. The computer program may be run on a terminal or a server. Program modules constituted by such computer programs may be stored on the memory of the electronic device. Which when executed by a processor, performs the steps of the method described in the embodiments of the present application.
The embodiment of the application also provides a computer readable storage medium. One or more non-transitory computer-readable storage media containing computer-executable instructions that, when executed by one or more processors, cause the processors to perform the steps of the grid line fault phase selection method.
A computer program product containing instructions which, when run on a computer, cause the computer to perform a grid line fault phase selection method.
Any reference to memory, storage, database, or other medium used herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM), which acts as external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (10)

1. A grid line fault phase selection method is characterized by comprising the following steps:
s1, acquiring a plurality of fault information sequences corresponding to the power grid line during the fault period, wherein the fault information sequences comprise a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence;
s2, acquiring gradient similarity of sampling points in every two fault information sequences, and constructing gradient similarity matrixes corresponding to the fault information sequences;
s3, carrying out visualization processing on the gradient similarity matrix to construct a visualization graph corresponding to the plurality of fault information sequences;
and S4, inputting the visual graph into a convolutional neural network model to perform fault phase selection on the power grid line during the fault.
2. The grid line fault phase selection method according to claim 1, wherein the step S2 specifically includes:
s201, acquiring gradient vectors of sampling points in each fault information sequence;
s202, acquiring the Manhattan distance between every two gradient vectors to identify the similarity of every two gradient vectors; and constructing the gradient similarity matrix by taking the Manhattan distance of each gradient vector as an element in the gradient similarity matrix.
3. The grid line fault phase selection method according to claim 2, wherein the step S201 specifically includes:
any two fault information sequences thetaiAnd thetaj，i,j＝1,2,…,m，θiAnd thetajThe corresponding time series are respectively recorded as
Figure FDA0002669374370000011
And
Figure FDA0002669374370000012
obtained by
Figure FDA0002669374370000013
And
Figure FDA0002669374370000014
the gradient vectors of the middle sampling points are respectively:
Figure FDA0002669374370000015
wherein the content of the first and second substances,
Figure FDA0002669374370000016
is thetaiCorresponding to
Figure FDA0002669374370000017
The (n) th sampling point of (c),
Figure FDA0002669374370000018
is thetajCorresponding to
Figure FDA0002669374370000019
N is 1, …, where m is the number of the fault information sequences, and N is the number of sampling points of m fault information sequences;
Figure FDA0002669374370000021
is composed of
Figure FDA0002669374370000022
Gradient of the nth sample point;
Figure FDA0002669374370000023
is composed of
Figure FDA0002669374370000024
Gradient of the nth sample point, [ theta ]1-θmThe sampling interval of (d) is Δ t;
the step S202 specifically includes:
manhattan distance MD for obtaining every two gradient vectorsi,jDescription of the invention
Figure FDA0002669374370000025
And
Figure FDA0002669374370000026
similarity between:
Figure FDA0002669374370000027
constructing the gradient similarity matrix GS-matrix by using the acquired respective Manhattan distances:
Figure FDA0002669374370000028
4. the grid line fault phase selection method according to claim 2, wherein the step S3 specifically includes:
s301, mapping each Manhattan distance in the gradient similarity matrix into a corresponding color level;
s302, constructing the visual graph by using each color level corresponding to the gradient similarity matrix.
5. The grid line fault phase selection method according to claim 2, wherein the step S301 specifically includes:
mapping each Manhattan distance into a corresponding color level according to a preset relationship, wherein the preset relationship is as follows:
Figure FDA0002669374370000029
wherein the content of the first and second substances,
Figure FDA00026693743700000210
is MDi,jThe color rank of (d); INT is a rounded symbol;
Figure FDA00026693743700000211
and
Figure FDA00026693743700000212
the maximum and minimum values of all elements in the GS-matrix, respectively.
6. The grid line fault phase selection method according to any one of claims 1 to 5, wherein, before the step S4, the grid line fault phase selection method further comprises:
setting the input layer structure of the convolutional neural network model to be the same as the size structure of the visual graph;
and setting an output layer structure of the convolutional neural network model according to the number of the fault phases.
7. The grid line fault phase selection method according to any one of claims 1 to 5, wherein the step S1 specifically includes:
s101, acquiring a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence of the head end and the tail end of each phase on the power grid line during the fault period;
and S102, taking the voltage signal time series, the current signal time series, the active power time series and the reactive power time series of the head end and the tail end of each phase as the plurality of fault information series.
8. A grid line fault phase selection apparatus, comprising:
the system comprises a sequence acquisition module, a fault detection module and a fault detection module, wherein the sequence acquisition module is used for acquiring a plurality of fault information sequences corresponding to a power grid line during a fault period, and the fault information sequences comprise a voltage signal time sequence, a current signal time sequence, an active power time sequence and a reactive power time sequence;
the matrix construction module is used for acquiring the gradient similarity of sampling points in every two fault information sequences and constructing gradient similarity matrixes corresponding to the fault information sequences;
the graph construction module is used for carrying out visualization processing on the gradient similarity matrix so as to construct a visualization graph corresponding to the plurality of fault information sequences;
and the fault phase selection module is used for inputting the visual graph into a convolutional neural network model so as to perform fault phase selection on the power grid line during the fault period.
9. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor realizes the steps of the method of any of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
CN202010928620.7A 2020-09-07 2020-09-07 Power grid line fault phase selection method and device, electronic equipment and storage medium Active CN112163012B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010928620.7A CN112163012B (en) 2020-09-07 2020-09-07 Power grid line fault phase selection method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010928620.7A CN112163012B (en) 2020-09-07 2020-09-07 Power grid line fault phase selection method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN112163012A true CN112163012A (en) 2021-01-01
CN112163012B CN112163012B (en) 2024-03-22

Family

ID=73857721

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010928620.7A Active CN112163012B (en) 2020-09-07 2020-09-07 Power grid line fault phase selection method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112163012B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113420402A (en) * 2021-08-24 2021-09-21 江苏智臻能源科技有限公司 Data feature similarity comparison method based on time sequence features
CN113533903A (en) * 2021-07-16 2021-10-22 南方电网科学研究院有限责任公司 Power distribution network ground fault phase selection method, device, equipment and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106291234A (en) * 2016-07-29 2017-01-04 武汉大学 A kind of transmission line of electricity internal fault external fault based on convolutional neural networks judges and fault phase-selecting method
WO2018133573A1 (en) * 2017-01-18 2018-07-26 华为技术有限公司 Method and device for analyzing service survivability

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106291234A (en) * 2016-07-29 2017-01-04 武汉大学 A kind of transmission line of electricity internal fault external fault based on convolutional neural networks judges and fault phase-selecting method
WO2018133573A1 (en) * 2017-01-18 2018-07-26 华为技术有限公司 Method and device for analyzing service survivability

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113533903A (en) * 2021-07-16 2021-10-22 南方电网科学研究院有限责任公司 Power distribution network ground fault phase selection method, device, equipment and storage medium
CN113533903B (en) * 2021-07-16 2023-02-28 南方电网科学研究院有限责任公司 Power distribution network ground fault phase selection method, device, equipment and storage medium
CN113420402A (en) * 2021-08-24 2021-09-21 江苏智臻能源科技有限公司 Data feature similarity comparison method based on time sequence features
CN113420402B (en) * 2021-08-24 2021-12-14 江苏智臻能源科技有限公司 Data feature similarity comparison method based on time sequence features

Also Published As

Publication number Publication date
CN112163012B (en) 2024-03-22

Similar Documents

Publication Publication Date Title
CN112149554B (en) Training and fault detection methods of fault classification model and related devices
CN112163012A (en) Power grid line fault phase selection method and device, electronic equipment and storage medium
CN112967357B (en) Spectrum map construction method based on convolutional neural network
CN111667010A (en) Sample evaluation method, device and equipment based on artificial intelligence and storage medium
CN111325159A (en) Fault diagnosis method and device, computer equipment and storage medium
CN115392037A (en) Equipment fault prediction method, device, equipment and storage medium
CN110580488A (en) Multi-working-condition industrial monitoring method, device, equipment and medium based on dictionary learning
CN114841199A (en) Power distribution network fault diagnosis method, device, equipment and readable storage medium
CN115313930A (en) Method for predicting excitation current of large-scale steam turbine generator in real time
CN112633362B (en) Rotary machine self-adaptive fault diagnosis method, device, equipment and medium
CN113537614A (en) Construction method, system, equipment and medium of power grid engineering cost prediction model
CN115689061B (en) Wind power ultra-short term power prediction method and related equipment
CN112801315A (en) State diagnosis method and device for power secondary equipment and terminal
CN114648227B (en) Method, device, equipment and storage medium for detecting abnormal time sequence of aviation hydraulic pump station
CN116796649A (en) SPEI coarse resolution data space downscaling method and device based on machine learning
CN116467665A (en) Fault diagnosis method, device, computer equipment and storage medium for nuclear reactor
CN116296233A (en) Test data validity analysis system and method for transonic continuous wind tunnel
CN115619449A (en) Deep learning model-based fertilizer price prediction method, device and storage medium
CN115795360A (en) Cable fault detection method based on artificial neural network
CN112488528A (en) Data set processing method, device, equipment and storage medium
CN116415510B (en) Breaker temperature rise prediction method and system based on phase space reconstruction and neural network
CN117109919A (en) Health diagnosis method and model of rolling bearing
CN116505522B (en) Simulation method, simulation platform and equipment for operation of power system
CN117314883B (en) Power distribution network fault line selection method and system based on EWT and VGGNet
CN117272186A (en) Wind driven generator bearing life prediction method and device based on hybrid model

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