CN117406038B - Tree line discharge fault early identification method and system based on curve difference degree - Google Patents

Abstract

The invention discloses a tree line discharge fault early-stage identification method and a system based on curve difference, which relate to the technical field of fault identification, and the method comprises the following steps: acquiring a first differential current of each phase; when the first differential current exceeds a preset threshold value, a plurality of second differential currents of the phase or the plurality of phases, a plurality of zero sequence currents of the head end and a plurality of zero sequence voltages are obtained; preprocessing a plurality of second differential currents, a plurality of zero sequence currents and a plurality of zero sequence voltages to obtain a data matrix; respectively calculating and obtaining the curve difference degree between the data matrix and the feature matrix of each group of data in the preset sample set; judging the data states of the curve difference degrees, wherein the data states comprise a fault state and a normal state; obtaining a discharge state of one or more phases in the distribution line based on a preset second judging condition; the problem that the tree line discharge fault characteristics are not obvious is solved, the possibility of misjudgment is greatly reduced, and the fault identification accuracy is improved.

Description

Early identification method and system of tree-line discharge fault based on curve difference

Technical Field

The present invention relates to the technical field of fault identification, and more specifically, to an early identification method and system for tree-line discharge faults based on curve difference.

Background technique

my country's distribution network is characterized by wide distribution and large scale. With the development of the power system, more and more distribution lines pass through forest areas and grasslands, which are seriously threatened by the growth of trees. When 10kV distribution overhead lines pass through forest areas, they may be affected by external factors (such as strong winds) and contact with higher branches, thereby causing discharge and constituting tree-line discharge faults. Since the transition resistance of tree-line discharge faults is generally in the order of hundreds of kΩ, its electrical characteristic quantity is weak and difficult to detect and identify, and generally cannot reach the protection start value, so it may exist in the line for a long time. The current flowing through the fault point causes the temperature of the trees to rise, which may ignite the trees and branches and cause fires, further causing tripping accidents, and even threatening the lives and property of the people. Therefore, it is necessary to accurately identify tree-line discharge faults, reduce the risk of wildfires, and ensure the safe and stable operation of distribution lines.

The tree-line discharge fault has large transition resistance, long duration, small fault current, and small characteristic quantity. Considering the complex conditions of the distribution lines, it is easy to be disturbed by factors such as equipment switching operation, unbalanced load and system noise, which will affect the identification of tree-line discharge faults. Therefore, the identification of tree-line discharge faults is difficult and an effective identification method is urgently needed.

Summary of the invention

The object of the present invention is to provide a method and system for early identification of tree-line discharge faults based on curve difference, so as to solve the problems existing in the above-mentioned background technology.

The above technical objectives of the present invention are achieved through the following technical solutions:

In a first aspect, an embodiment of the present application provides a method for early identification of tree line discharge faults based on curve difference, comprising the following steps:

Acquire and monitor the current at the beginning and the end of each phase of the distribution line passing through the forest area, and calculate the first differential current of each phase in the distribution line according to the current at the beginning and the end;

When a first differential current of one or more phases in the distribution line exceeds a preset threshold, a plurality of second differential currents of the phase or phases, as well as a plurality of zero-sequence currents and a plurality of zero-sequence voltages at the head end are acquired in a time sequence within a preset time period;

Preprocessing the plurality of second differential currents, the plurality of zero-sequence currents and the plurality of zero-sequence voltages to obtain a data matrix consisting of the second differential currents, the zero-sequence currents and the zero-sequence voltages;

The curve differences between the data matrix and the characteristic matrix of each group of data in the preset sample set are respectively calculated and obtained, and the preset sample set includes multiple groups of fault data and multiple groups of normal data;

According to a preset first determination condition, determining data states of a plurality of curve differences, the data states including a fault state and a normal state;

According to the proportion of the data state of each curve difference in the multiple curve differences, and based on the KNN algorithm and the preset second judgment condition, the discharge state of one or more phases in the distribution line is obtained, and the discharge state includes tree line discharge fault and no fault.

The beneficial effects of the present invention are as follows: by obtaining the differential current obtained by taking the difference between the currents at the beginning and end of the line fault phase passing through the forest area as the monitoring and identification parameter, the influence of equipment switching operation, unbalanced load and system noise can be eliminated, the anti-interference ability is stronger, and it is more sensitive to the high-resistance grounding faults at the beginning and end of the section passing through the forest area, and ultimately the identification result can be made more accurate; further, by forming a composite criterion through the three electrical parameters of zero-sequence voltage, zero-sequence current and differential current, the fault identification accuracy and anti-interference ability can be improved.

In this scheme, the three electrical parameters of zero-sequence voltage, zero-sequence current and differential current are preprocessed, and the size of each element in the data matrix obtained after processing is linearly transformed to between 0 and 1, so as to remove the data dimension. In this way, the numerical value of the curve can be ignored and the changing trend of the curve can be highlighted. After the preprocessing is completed, each group of data in the data matrix is transformed from the original absolute size representation to the relative size, that is, they are all between 0 and 1, and there is only a difference in rising or falling trends, thereby highlighting the changing trends of each data and improving the feature recognition of the data.

In this scheme, the distance between points in the data matrix and the feature matrix is used to characterize the similarity of the curve trends, thereby defining the curve difference, and using the curve difference as the main benchmark for fault classification. The principle is to refer to the definition of the residuals of the two curves, calculate the distance between the function values of the two curves under the same horizontal coordinate, and reflect the similarity of the curves through the total value of the difference in the function values of all points under all horizontal coordinates; at the same time, the curve difference is calculated. Since the manifestations of faults and normal operations are completely different for different lines, tree conditions, and operating conditions, the growth of tree-line discharge fault currents may also be different for different regions. For fault modes with large differences, the KNN algorithm can achieve excellent results of specialization for different regions.

In this scheme, the change of fault current in the early stage of tree-line discharge is not a sudden change process, but a development process in which the fault current gradually increases monotonically for a certain period of time. It can be seen that in the early stage of tree-line discharge fault, the process of gradually monotonically increasing fault current is a characteristic process unique to tree-line discharge fault. This is used as the basis for judging the fault situation, and the three electrical parameters of zero-sequence voltage, zero-sequence current and differential current are used as composite criteria as judgment parameters, which solves the problem of unclear characteristics of tree-line discharge faults, greatly reduces the possibility of misjudgment, improves the accuracy of fault identification, and effectively avoids igniting trees and branches to cause fires, further causing tripping accidents, and even threatening the safety of people’s lives and property. Ultimately, the purpose of accurately identifying tree-line discharge faults, reducing the risk of wildfires, and ensuring safe and stable operation of distribution lines is achieved.

Based on the above technical solution, the present invention can also be improved as follows.

Further, the preset time period includes a plurality of unit times, and the plurality of second differential currents, the plurality of zero-sequence currents, and the plurality of zero-sequence voltages all correspond to the plurality of unit times; wherein the plurality of second differential currents, the plurality of zero-sequence currents, and the plurality of zero-sequence voltages are preprocessed to obtain a data matrix consisting of the second differential currents, the zero-sequence currents, and the zero-sequence voltages, including:

Performing fast Fourier transform processing on the detection data to obtain multiple waveform spectra corresponding to each group of data in the detection data, the detection data including the second differential current, the zero-sequence current and the zero-sequence voltage;

For multiple waveform spectra of each group of data in the detection data, extract the component amplitude of the 50 Hz frequency band in each waveform spectrum, and determine the component amplitude as point data;

Arrange multiple point data of each group of detection data in time sequence to obtain an intermediate matrix corresponding to the detection data;

The intermediate matrix is standardized to obtain a processed data matrix corresponding to the intermediate matrix.

The beneficial effect of adopting the above further scheme is: it can effectively filter out the noise interference of each data, perform fast Fourier transform on the data corresponding to multiple unit times, and extract the component amplitude of the 50Hz frequency band, thereby converting each group of data in the preset time period into a plurality of point data corresponding to the number of unit times, and standardizing the point data, reducing each point data to the range of 0 to 1, removing the dimension, and then forming the three groups of variables into a data matrix for the operation of subsequent processes.

Furthermore, the above standardization process is represented by a first formula, which is:

In the formula, x(m,n) represents the point data located in the mth row and nth column of the intermediate matrix, _xnmax represents the maximum value of the nth column, _xnmin represents the minimum value of the nth column, and _xnom (m,n) represents the point data after x(m,n) is standardized.

Furthermore, the above curve difference is expressed by a second formula, which is:

Where D _i represents the curve difference, Z(m,n) represents the point data in the mth row and nth column in the data matrix, Z _i (m,n) represents the point data in the mth row and nth column in the feature matrix, a represents the number of unit times in the preset time period, and b represents the number of types of each group of data in the detection data.

Furthermore, the second determination condition is:

If the number of curve difference degrees of the fault state among the plurality of curve difference degrees is greater than the number of curve difference degrees of the normal state, the discharge state is a tree line discharge fault;

If the number of curve differences of the fault state among the plurality of curve differences is smaller than the number of curve differences of the normal state, the discharge state is no fault.

In a second aspect, an embodiment of the present application provides a tree-line discharge fault early identification system based on curve difference, which is applied to any one of the tree-line discharge fault early identification methods based on curve difference in the first aspect, including:

A monitoring module is used to obtain and monitor the current at the beginning and the end of each phase in the distribution line passing through the forest area, and calculate the first differential current of each phase in the distribution line according to the current at the beginning and the end;

A start-up module, used for acquiring multiple second differential currents of one or more phases in the distribution line, as well as multiple zero-sequence currents and multiple zero-sequence voltages at the head end in a time sequence within a preset time period when the first differential current of one or more phases in the distribution line exceeds a preset threshold value;

A data processing module, used for preprocessing the plurality of second differential currents, the plurality of zero-sequence currents and the plurality of zero-sequence voltages to obtain a data matrix consisting of the second differential currents, the zero-sequence currents and the zero-sequence voltages;

A data calculation module, used to respectively calculate and obtain the curve difference between the data matrix and the characteristic matrix of each group of data in the preset sample set, wherein the preset sample set includes multiple groups of fault data and multiple groups of normal data;

A data judgment module, used to judge the data status of the difference of multiple curves according to a preset first judgment condition, the data status including a fault state and a normal state;

The state judgment module is used to obtain the discharge state of one or more phases in the distribution line according to the proportion of the data state of each curve difference in multiple curve differences, based on the KNN algorithm and the preset second judgment condition, and the discharge state includes tree line discharge fault and no fault.

Furthermore, the preset time period includes a plurality of unit times, and the plurality of second differential currents, the plurality of zero-sequence currents and the plurality of zero-sequence voltages all correspond to the plurality of unit times; wherein the data processing module includes:

A first submodule is used to perform fast Fourier transform processing on the detection data to obtain multiple waveform spectra corresponding to each group of data in the detection data, where the detection data includes a second differential current, a zero-sequence current and a zero-sequence voltage;

The second submodule is used to extract the component amplitude of the 50 Hz frequency band in each waveform spectrum for each group of data in the detection data, and determine the component amplitude as point data;

The third submodule is used to arrange multiple point data of each group of data in the detection data in time sequence to obtain an intermediate matrix corresponding to the detection data;

The fourth submodule is used to perform standardization processing on the intermediate matrix to obtain a processed data matrix corresponding to the intermediate matrix.

Furthermore, in the above-mentioned state judgment module, the second judgment condition is:

If the number of curve difference degrees of the fault state among the plurality of curve difference degrees is greater than the number of curve difference degrees of the normal state, the discharge state is a tree line discharge fault;

If the number of curve differences of the fault state among the plurality of curve differences is smaller than the number of curve differences of the normal state, the discharge state is no fault.

In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the methods in the first aspect when executing the computer program.

In a fourth aspect, an embodiment of the present application provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute any one of the methods in the first aspect.

Compared with the prior art, the present invention has at least the following beneficial effects:

In the present application, by obtaining the differential current obtained by taking the difference between the currents at the beginning and end of the line fault phase passing through the forest area as the monitoring and identification parameter, the influence of equipment switching operation, unbalanced load and system noise can be eliminated, the anti-interference ability is stronger, and it is more sensitive to the high-resistance grounding faults at the beginning and end of the section passing through the forest area, which can ultimately make the identification result more accurate; at the same time, by forming a composite criterion with three electrical parameters of zero-sequence voltage, zero-sequence current and differential current, the fault identification accuracy and anti-interference ability can be improved.

In the present application, by means of data preprocessing, the size of each element in the data matrix obtained after processing is linearly transformed to between 0 and 1, thereby removing the data dimension. In this way, the numerical value of the curve can be ignored and the changing trend of the curve can be highlighted. After the preprocessing is completed, each group of data in the data matrix is transformed from the original absolute size representation to a relative size, that is, they are all between 0 and 1, and there is only a difference in rising or falling trends, thereby achieving the highlighting of the changing trends of each data and improving the feature recognition of the data.

In the present application, the distance between points in the data matrix and the feature matrix is used to characterize the similarity of the curve trends; at the same time, the curve difference is calculated. Since the manifestations of faults and normal operations are completely different for different lines, tree conditions, and operating conditions, the growth of tree-line discharge fault currents may also be different in different areas. For fault modes with large differences, the KNN algorithm can achieve excellent results of specialization for different areas.

In the present application, the fault current change in the early stage of tree-line discharge is not a sudden change process, but a development process in which the fault current gradually increases monotonically over a certain period of time. This is used as the basis for judging the fault situation, and the three electrical parameters of zero-sequence voltage, zero-sequence current and differential current are used as composite criteria as judgment parameters. This solves the problem of unclear characteristics of tree-line discharge faults, greatly reduces the possibility of misjudgment, improves the accuracy of fault identification, and effectively avoids igniting trees and branches to cause fires, further causing tripping accidents, and even threatening the safety of people’s lives and property. Ultimately, the purpose of accurately identifying tree-line discharge faults, reducing the risk of wildfires, and ensuring safe and stable operation of distribution lines is achieved.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

FIG1 is a flow chart of an identification method according to an embodiment of the present invention;

FIG2 is a schematic diagram of the connection of the identification system in an embodiment of the present invention;

FIG. 3 is a schematic diagram of the connection of an electronic device in an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not require further definition and explanation in the subsequent drawings.

In the description of the embodiments of the present invention, "plurality" means at least 2.

Example 1

This embodiment provides a method for early identification of tree line discharge faults based on curve difference, as shown in FIG1 , including the following steps:

S1, obtain and monitor the current at the beginning and end of each phase in the distribution line passing through the forest area, and calculate the first differential current of each phase in the distribution line according to the current at the beginning and end.

Specifically, in power supply lines passing through forest areas, current transformers of the same type can be installed in each phase before and after the forest area, and the differential current can be obtained by subtracting the current measured by the current transformer at the incoming end from the current measured at the outgoing end. Among them, for distribution lines, three-phase electricity is often used. Therefore, for each phase of the distribution line, it is necessary to monitor the current at both ends of the forest area to calculate the differential current of each phase.

S2, when the first differential current of one or more phases in the distribution line exceeds a preset threshold, multiple second differential currents of the phase or phases, as well as multiple zero-sequence currents and multiple zero-sequence voltages at the head end are acquired in time sequence within a preset time period.

Specifically, for zero-sequence current and zero-sequence voltage, they can be monitored and obtained by installing zero-sequence current transformers and zero-sequence voltage transformers at the head end of the distribution network entering the forest area; wherein, when the differential current of one or more phases exceeds a preset threshold, data acquisition begins, and the preset threshold can be 1.2 times that of the normal operating state, that is, when the differential current exceeds 1.2 of the normal state, the second differential current, zero-sequence current and zero-sequence voltage of the corresponding phase are continuously acquired within the preset time period in the future.

Among them, the preset time period can be 10s, and the second differential current, zero-sequence current and zero-sequence voltage arranged in time are continuously obtained within this 10 seconds. Within this 10s, the number of data obtained is determined by the acquisition frequency, which can be collected once every 1s, then 10 second differential currents, 10 zero-sequence currents and 10 zero-sequence voltages are obtained within 10s.

S3, preprocessing the plurality of second differential currents, the plurality of zero-sequence currents and the plurality of zero-sequence voltages to obtain a data matrix consisting of the second differential currents, the zero-sequence currents and the zero-sequence voltages.

The purpose of preprocessing is to reduce each data to the range of 0 to 1 and remove the dimension in order to calculate the curve difference.

Optionally, the preset time period includes multiple unit times, and the multiple second differential currents, multiple zero-sequence currents, and multiple zero-sequence voltages all correspond to the multiple unit times. The multiple second differential currents, multiple zero-sequence currents, and multiple zero-sequence voltages are preprocessed to obtain a data matrix consisting of the second differential currents, zero-sequence currents, and zero-sequence voltages, including:

The detection data is processed by fast Fourier transform to obtain a plurality of waveform spectra corresponding to each group of data in the detection data, wherein the detection data includes a second differential current, a zero-sequence current and a zero-sequence voltage.

For the multiple waveform spectra of each group of data in the detection data, the component amplitude of the 50 Hz frequency band in each waveform spectrum is extracted, and the component amplitude is determined as point data.

Arrange multiple point data of each group of detection data in time sequence to obtain an intermediate matrix corresponding to the detection data.

The intermediate matrix is standardized to obtain a processed data matrix corresponding to the intermediate matrix.

Specifically, as mentioned above, the preset time period is 10s, and the acquisition frequency is 1s/time, then the unit time is 1s, so 10 second differential currents, 10 zero-sequence currents and 10 zero-sequence voltages are obtained. Since there are a large number of interference signals in the measured data, especially the harmonic signals and measurement signal interference in the current signal, it is necessary to extract the effective part of the data; the specific processing process is: first, the 10 second differential currents, 10 zero-sequence currents and 10 zero-sequence voltages are fast Fourier transformed to obtain the corresponding waveform frequency spectrum; secondly, extract the component amplitude of the 50Hz frequency band for each 1s waveform spectrum, and use the component amplitude as the data of this 1s, that is, the point data. Thus, the continuous waveform data of 10s is changed into 10 discrete point data, and a total of three groups of 30 point data are obtained; then, the 10 point data in each group of data are arranged in chronological order, and the 10 point data in each group of data are arranged from top to bottom to form a matrix of 10 rows and 3 columns; finally, the obtained matrix is standardized to finally obtain the data matrix within this 10s.

Optionally, the above standardization process is represented by a first formula, which is:

In the formula, x(m,n) represents the point data located in the mth row and nth column of the intermediate matrix, _xnmax represents the maximum value of the nth column, _xnmin represents the minimum value of the nth column, and _xnom (m,n) represents the point data after x(m,n) is standardized.

Among them, each element in the matrix of 10 rows and 3 columns is brought into the first formula for calculation to obtain a processed data matrix of 10 rows and 3 columns.

S4, respectively calculating and obtaining the curve difference between the data matrix and the characteristic matrix of each group of data in the preset sample set, wherein the preset sample set includes multiple groups of fault data and multiple groups of normal data.

Among them, the preset sample set includes multiple groups of fault data and multiple groups of normal data, and both the fault data and the normal data include at least three types of parameters: differential current, zero-sequence current and zero-sequence voltage; for each group of data, a corresponding feature matrix can be obtained, and the method of obtaining the feature matrix is the same as that of the data matrix, which will not be repeated here; for example, if there are 50 groups of fault data and 50 groups of normal data in the sample set, 100 feature matrices can be obtained.

Specifically, the curve difference is calculated by sequentially calculating the 30 point data in the data matrix with the 30 data of the corresponding type in the 100 feature matrices. Therefore, there are 100 curve difference results in the end, and each one has 30 sub-data; among them, the curve difference is defined as the sum of the residuals of the column vectors of the two matrices, reflecting the difference in the numerical change trends represented by the corresponding column vectors of the two matrices.

S5, judging the data states of the plurality of curve differences according to a preset first judgment condition, where the data states include a fault state and a normal state.

Specifically, in the early stage of fault, the fault data shows a monotonically increasing trend, while the normal data shows a random distribution, so the data can be classified according to the size of the curve difference; among them, for one of the curve differences, the 30 sub-data include 3 types of data, so in the early stage of fault, the 3 types of sub-data have a monotonically increasing trend in chronological order, while the normal state is randomly distributed; finally, the data state of each curve difference can be obtained.

Optionally, the above curve difference is expressed by a second formula, which is:

Where D _i represents the curve difference, Z(m,n) represents the point data in the mth row and nth column in the data matrix, Z _i (m,n) represents the point data in the mth row and nth column in the feature matrix, a represents the number of unit times in the preset time period, and b represents the number of types of each group of data in the detection data.

Specifically, in this embodiment, the preset time period is 10s, the unit time is 1s, and a in the second formula is 10; the data matrix includes three types of parameters: differential current, zero-sequence voltage, and zero-sequence current, and b in the second formula is 3. Therefore, the obtained second formula is:

S6, according to the proportion of the data state of each curve difference in the multiple curve differences, and based on the KNN algorithm and the preset second judgment condition, the discharge state of one or more phases in the distribution line is obtained, and the discharge state includes tree line discharge fault and no fault.

The multiple curve differences may be all the calculated curve differences, or may be a part selected from all the calculated curve differences, and the screening condition may be the smallest multiple curve differences.

Optionally, the second determination condition is:

If the number of curve differences of the fault state among the plurality of curve differences is greater than the number of curve differences of the normal state, the discharge state is a tree line discharge fault.

If the number of curve differences of the fault state among the plurality of curve differences is smaller than the number of curve differences of the normal state, the discharge state is no fault.

Among them, the data status of each curve difference has been obtained through the above steps, and the final judgment is based on the proportion of the data status; for example, 100 curve differences are obtained in the above, and the smallest 50 curve differences can be selected from the 100 curve differences. Among these 50 curve differences, if there are more curve differences with a fault state, it indicates that the phase of the distribution network is a tree line discharge fault, otherwise, there are more curve differences in the normal state, which indicates that the phase of the distribution line is fault-free.

Specifically, the present application takes into account that the development process of a tree-line discharge fault may last from tens of seconds to several minutes, and completes fault identification through 10s of data, meeting the requirement for rapid fault identification; at the same time, based on the characteristic that the fault current gradually increases during the tree-line discharge fault, the present application identifies the comprehensive criteria consisting of zero-sequence voltage, zero-sequence current and differential current, and can simultaneously complete the functions of line selection and phase selection, thereby quickly locating the faulty phase. Once a tree-line discharge fault is identified, a warning can be immediately issued to the responsible personnel of the relevant lines, and the fault can be handled in time to avoid causing a fire, which has strong engineering practical significance; in addition, the present method is based on composite criteria for identification, and processes a large amount of data up to 10s, which solves the problem that the characteristics of the tree-line discharge fault are not obvious, greatly reduces the possibility of misjudgment, and improves the accuracy of fault identification.

Example 2

The embodiment of the present application provides a tree-line discharge fault early identification system based on curve difference, which is applied to any one of the tree-line discharge fault early identification methods based on curve difference in embodiment 1, as shown in FIG2, including:

The monitoring module is used to obtain and monitor the current at the beginning and end of each phase in the distribution line passing through the forest area, and calculate the first differential current of each phase in the distribution line based on the current at the beginning and end.

A starting module is used to obtain multiple second differential currents of one or more phases in the distribution line, as well as multiple zero-sequence currents and multiple zero-sequence voltages at the head end in a time sequence within a preset time period when the first differential current of one or more phases in the distribution line exceeds a preset threshold.

The data processing module is used to pre-process the multiple second differential currents, the multiple zero-sequence currents and the multiple zero-sequence voltages to obtain a data matrix consisting of the second differential currents, the zero-sequence currents and the zero-sequence voltages.

Optionally, the preset time period includes a plurality of unit times, and the plurality of second differential currents, the plurality of zero-sequence currents, and the plurality of zero-sequence voltages all correspond to the plurality of unit times; wherein the data processing module includes:

The first submodule is used to perform fast Fourier transform processing on the detection data to obtain multiple waveform spectra corresponding to each group of data in the detection data, and the detection data includes the second differential current, the zero-sequence current and the zero-sequence voltage.

The second submodule is used to extract the component amplitude of the 50 Hz frequency band in each waveform spectrum for each group of data in the detection data, and determine the component amplitude as point data.

The third submodule is used to arrange multiple point data of each group of data in the detection data in time sequence to obtain an intermediate matrix corresponding to the detection data.

The fourth submodule is used to perform standardization processing on the intermediate matrix to obtain a processed data matrix corresponding to the intermediate matrix.

The data calculation module is used to respectively calculate and obtain the curve difference between the data matrix and the characteristic matrix of each group of data in the preset sample set, and the preset sample set includes multiple groups of fault data and multiple groups of normal data.

The data judgment module is used to judge the data status of the difference between multiple curves according to a preset first judgment condition, and the data status includes a fault state and a normal state.

The state judgment module is used to obtain the discharge state of one or more phases in the distribution line according to the proportion of the data state of each curve difference in multiple curve differences, based on the KNN algorithm and the preset second judgment condition, and the discharge state includes tree line discharge fault and no fault.

Optionally, in the above state judgment module, the second judgment condition is:

If the number of curve differences of the fault state among the plurality of curve differences is greater than the number of curve differences of the normal state, the discharge state is a tree line discharge fault.

If the number of curve differences of the fault state among the plurality of curve differences is smaller than the number of curve differences of the normal state, the discharge state is no fault.

Example 3

An embodiment of the present application provides an electronic device, see Figure 3, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any method in Embodiment 1 is implemented.

Example 4

An embodiment of the present application provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute any one of the methods in Example 1.

The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims (10)

1. The early identification method of the tree line discharge faults based on the curve difference degree is characterized by comprising the following steps of:

Acquiring and monitoring currents of the head end and the tail end of each phase in a distribution line passing through a forest zone, and calculating to obtain a first differential current of each phase in the distribution line according to the currents of the head end and the tail end;

when the first differential current of one phase or multiple phases in the distribution line exceeds a preset threshold value, acquiring a plurality of second differential currents of the phase or multiple phases, a plurality of zero sequence currents of the head end and a plurality of zero sequence voltages according to time sequence in a preset time period;

Preprocessing a plurality of second differential currents, a plurality of zero sequence currents and a plurality of zero sequence voltages to obtain a data matrix formed by the second differential currents, the zero sequence currents and the zero sequence voltages;

respectively calculating and obtaining curve difference degrees between the data matrix and feature matrices of each group of data in a preset sample set, wherein the preset sample set comprises a plurality of groups of fault data and a plurality of groups of normal data;

Judging the data states of a plurality of curve difference degrees according to a preset first judging condition, wherein the data states comprise a fault state and a normal state;

And obtaining the discharge state of one or more phases in the distribution line based on the KNN algorithm and a preset second judging condition according to the duty ratio of the data state of each curve difference degree in the curve difference degrees, wherein the discharge state comprises a tree line discharge fault and no fault.

2. The early recognition method of a tree discharge fault based on a curve difference degree according to claim 1, wherein the preset time period includes a plurality of unit times, and a plurality of the second differential currents, the zero sequence currents, and the zero sequence voltages each correspond to a plurality of the unit times; the preprocessing of the second differential currents, the zero sequence currents and the zero sequence voltages to obtain a data matrix composed of the second differential currents, the zero sequence currents and the zero sequence voltages comprises the following steps:

Performing fast Fourier transform processing on the detection data to obtain a plurality of waveform spectrums corresponding to each group of data in the detection data, wherein the detection data comprises a second differential current, a zero sequence current and a zero sequence voltage;

extracting component amplitude values of a 50Hz frequency band in each waveform spectrum for a plurality of waveform spectrums of each group of data in the detection data, and determining the component amplitude values as point data;

arranging a plurality of point data of each group of data in the detection data according to time sequence to obtain an intermediate matrix corresponding to the detection data;

And carrying out standardization processing on the intermediate matrix to obtain a data matrix which is processed and corresponds to the intermediate matrix.

3. The early identification method of a tree line discharge fault based on a curve difference degree according to claim 2, wherein the normalization process is represented by a first formula:

Where x (m, n) represents dot data located in an m-th row and an n-th column in the intermediate matrix, x _nmax represents a maximum value of the n-th column, x _nmin represents a minimum value of the n-th column, and x _nom (m, n) represents dot data after the normalization process of x (m, n).

4. The early recognition method of a tree line discharge fault based on a curve difference degree according to claim 1, wherein the curve difference degree is represented by a second formula, the second formula being:

Wherein D _i represents the curve difference degree, Z (m, n) represents the data of the m-th row and n-th column in the data matrix, Z _i (m, n) represents the data of the m-th row and n-th column in the feature matrix, a represents the number of unit time in a preset time period, and b represents the number of types of each group of data in the detection data.

5. The method for early identifying a tree line discharge fault based on a curve difference degree according to claim 1, wherein the second determination condition is:

if the number of the curve difference degrees of the fault states in the plurality of curve difference degrees is larger than that of the curve difference degrees of the normal states, the discharge state is a tree line discharge fault;

and if the number of the curve difference degrees of the fault state in the plurality of curve difference degrees is smaller than that of the curve difference degrees of the normal state, the discharge state is fault-free.

6. The early recognition system of the tree line discharge faults based on the curve difference degree is applied to the early recognition method of the tree line discharge faults based on the curve difference degree as claimed in any one of claims 1 to 5, and is characterized by comprising the following steps:

The monitoring module is used for acquiring and monitoring the current of the head end and the tail end of each phase in the distribution line passing through the forest zone, and calculating to obtain a first differential current of each phase in the distribution line according to the current of the head end and the tail end;

The starting module is used for acquiring a plurality of second differential currents of one phase or multiple phases, a plurality of zero sequence currents of the head end and a plurality of zero sequence voltages of the phase or multiple phases according to time sequence in a preset time period when the first differential current of the phase or multiple phases in the distribution line exceeds a preset threshold value;

The data processing module is used for preprocessing a plurality of second differential currents, a plurality of zero sequence currents and a plurality of zero sequence voltages to obtain a data matrix formed by the second differential currents, the zero sequence currents and the zero sequence voltages;

The data calculation module is used for respectively calculating and obtaining the curve difference degree between the data matrix and the characteristic matrix of each group of data in a preset sample set, wherein the preset sample set comprises a plurality of groups of fault data and a plurality of groups of normal data;

The data judging module is used for judging a plurality of data states of the curve difference degrees according to a preset first judging condition, wherein the data states comprise a fault state and a normal state;

and the state judging module is used for obtaining the discharge state of one or more phases in the distribution line based on the KNN algorithm and a preset second judging condition according to the duty ratio of the data state of each curve difference degree in the curve difference degrees, wherein the discharge state comprises a tree line discharge fault and no fault.

7. The early recognition system of a tree discharge fault based on a curve difference degree according to claim 6, wherein the preset time period includes a plurality of unit times, and a plurality of the second differential currents, the zero sequence currents, and the zero sequence voltages each correspond to a plurality of the unit times; wherein, the data processing module includes:

The first sub-module is used for performing fast Fourier transform processing on the detection data to obtain a plurality of waveform spectrums corresponding to each group of data in the detection data, wherein the detection data comprises a second differential current, a zero sequence current and a zero sequence voltage;

The second sub-module is used for extracting the component amplitude of the 50Hz frequency band in each waveform spectrum for a plurality of waveform spectrums of each group of data in the detection data, and determining the component amplitude as point data;

A third sub-module, configured to arrange a plurality of point data of each group of data in the detection data according to a time sequence, so as to obtain an intermediate matrix corresponding to the detection data;

and the fourth sub-module is used for carrying out standardization processing on the intermediate matrix to obtain a data matrix which is processed and corresponds to the intermediate matrix.

8. The early recognition system of a tree discharge fault based on a degree of curve difference according to claim 6, wherein in the state judgment module, the second judgment condition is:

if the number of the curve difference degrees of the fault states in the plurality of curve difference degrees is larger than that of the curve difference degrees of the normal states, the discharge state is a tree line discharge fault;

and if the number of the curve difference degrees of the fault state in the plurality of curve difference degrees is smaller than that of the curve difference degrees of the normal state, the discharge state is fault-free.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1-5 when the computer program is executed by the processor.

10. A non-transitory computer readable storage medium storing computer instructions that cause a computer to perform the method of any one of claims 1-5.