CN117706278B - Fault line selection method and system for power distribution network and readable storage medium - Google Patents

Abstract

The invention relates to the technical field of power systems, in particular to a fault line selection method, a fault line selection system and a readable storage medium of a power distribution network, wherein the method comprises the following steps: after the power distribution network fails, zero sequence current in a preset time window of each feeder line and zero sequence voltage of a bus are obtained; according to the zero sequence current and the zero sequence voltage, two-dimensionally processing a feeder line in the power distribution network in a preset short-time window zero sequence instantaneous power curve cluster through a KPCA algorithm to determine a corresponding principal component score; and performing BIRCH clustering according to the principal component score to determine whether the feeder is a fault feeder. Whether the feeder line is a line with faults or not is judged through principal component score clustering, so that the target fault feeder line in the power distribution network can be accurately and rapidly determined when the current amount of the power distribution network is small, and the detection precision is improved. The problem of how to quickly identify a faulty feeder line when a single-phase earth fault occurs in a low-current grounded power distribution network is solved.

Description

Fault line selection method, system and readable storage medium for distribution network

Technical Field

The present invention relates to the technical field of power systems, and in particular to a method and system for fault line selection in a distribution network, and a readable storage medium.

Background technique

With the rapid development of my country's distribution network, the number and scale of distribution networks are gradually increasing, which has added a lot of workload to the operation and maintenance of distribution networks. How to effectively ensure and improve the safe and reliable operation of distribution networks is related to the safety and reliability of users' electricity use. Therefore, power supply companies must face a very important challenge, that is, how to continuously ensure and improve the safety and reliability of distribution networks.

The neutral point of my country's 10kV distribution network is generally ungrounded or resonantly grounded. When a single-phase grounding fault occurs, the current flowing through the fault point is very small, and it does not affect the symmetry of the system. It can run with the fault for a period of time, so it is called a small current grounding system. The main advantage of the small current grounding system is that the current flowing into the earth is small, and some interfering instantaneous faults can not cause protection action, effectively improving the reliability of power supply. Among various short-circuit faults, about 70% of power outages are caused by single-phase grounding faults in distribution network lines, and a large part of single-phase grounding faults are grounding arc faults. If they are not handled for a long time, they are still easy to cause fire accidents. Especially in recent years, with the rapid development of the economy and the increasing scale of cities, the proportion of cable lines invested in distribution network lines is getting higher and higher. When a single-phase grounding fault is formed due to some factors, due to the significant increase in the system's ground capacitance, a large grounding current will be generated, and the arc is difficult to extinguish by itself. Long-term operation is easy to burn equipment and lines, and in severe cases, a developmental fault will form or cause a wildfire. According to current research, a single-phase grounding fault in the distribution network will directly lead to an arc fault, and an arc fault is difficult to extinguish during the transient process. Therefore, the fault must be removed in time when a single-phase grounding fault occurs to prevent it from developing into an arc fault.

In the relevant technical solutions of the distribution network, the neutral point of the distribution network is generally ungrounded or resonantly grounded. When a single-phase grounding fault occurs, the current flowing through the fault point is very small, and it does not affect the symmetry of the system. It can operate with the fault for a period of time, so it is called a small current grounding distribution network. The main advantage of the small current grounding distribution network is that the current flowing into the earth is small, and some interfering instantaneous faults can not cause protection action, which effectively improves the reliability of power supply.

However, when the inventors were conceiving and implementing this solution, they found that there were at least the following defects: since the current in the small-current grounding distribution network is small, when a single-phase grounding fault occurs in the distribution network, the traditional fault line selection device is prone to difficulty in identifying the faulty feeder and has the defect of insufficient detection accuracy.

The KPCA-BIRCH clustering algorithm uses kernel functions to complete nonlinear transformations on principal component analysis (PCA), further mapping the data set to a linearly separable higher-dimensional feature space, expressing the main information of the original data with data of fewer dimensions, and then performing unsupervised clustering of the obtained data through the BIRCH algorithm to determine the best clustering data and the number of clustering layers. Compared with the PCA algorithm, the KPCA-BIRCH algorithm is more adaptable to nonlinear equation processing similar to distribution network fault conditions, and through data dimensionality reduction and dual algorithm fusion unsupervised clustering, it effectively realizes adaptive, high-precision, and high-efficiency clustering without rigid clustering features, thereby significantly improving the accuracy of fault line selection and reducing the problem of low robustness of line selection in previous algorithms.

Summary of the invention

The main purpose of the present invention is to provide a fault line selection method for a distribution network, aiming to solve the problem of how to quickly identify a faulty feeder when a single-phase grounding fault occurs in a small current grounding distribution network.

To achieve the above object, the present invention provides a method for fault line selection in a distribution network, the method comprising:

Obtain the zero-sequence current of each feeder and the zero-sequence voltage of the busbar within the preset time window after a distribution network fault occurs;

According to the zero-sequence current and the zero-sequence voltage, a zero-sequence instantaneous power curve cluster of the feeder in the distribution network in a preset short-time window is processed in two dimensions by a KPCA algorithm to determine a corresponding principal component score;

BIRCH clustering is performed according to the principal component scores to determine whether the feeder is a faulty feeder.

Optionally, before the step of two-dimensionalizing a preset short-time window zero-sequence instantaneous power curve cluster of the feeder in the distribution network by using a KPCA algorithm according to the zero-sequence current and the zero-sequence voltage to determine a corresponding principal component score, the step further includes:

Acquire first intercepted data, and use the first intercepted data as the upper limit of the short-time window, wherein the first intercepted data is data intercepted at a first preset time interval before the fault;

And, obtaining second intercepted data, and using the second intercepted data as the lower limit of the short-time window, wherein the second intercepted data is data intercepted at a second preset time interval after the fault;

Determine a short time window interception interval according to the short time window upper limit and the short time window lower limit;

Determine the zero-sequence instantaneous power curve cluster of the distribution network in the short-time window interception interval as the short-time window zero-sequence instantaneous power curve cluster;

Among them, the first preset time length is shorter than the second preset time length.

Optionally, the step of two-dimensionally processing a zero-sequence instantaneous power curve cluster of a feeder in the distribution network in a preset short-time window by a KPCA algorithm according to the zero-sequence current and the zero-sequence voltage to determine a corresponding principal component score includes:

Determining a target instantaneous power curve in a short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

Based on the KPCA algorithm, determining the two-dimensional coordinates of the target instantaneous power curve, wherein the two-dimensional coordinates represent the fault zero-sequence power of the feeder;

The two-dimensional coordinates are used as the principal component scores.

The two-dimensional coordinates are obtained by the KPCA algorithm, and the formula is as follows:

;

in is the generating matrix of the feature space, ,..., Represents the feature samples in the feature space, N is the number of samples, K is the gram matrix, and the elements in the matrix , is the eigenvalue.

Optionally, the step of performing BIRCH clustering according to the principal component score to determine whether the feeder is a faulty feeder includes:

The two-dimensional curve clusters are grouped, and the failure modes existing in the data are discovered through unsupervised clustering, so as to mine the internal coupling relationship of the data;

The unsupervised clustering method is to automatically and unsupervisedly discover the fault modes existing in the data without anyone setting the number of clustering layers;

The optimal clustering data value k is determined by the BIRCH algorithm. The optimal clustering data value k is jointly acted on by the silhouette coefficient _Si and the CH index, and the optimal number of layers is iterated. The larger the silhouette coefficient Si and the CH index, the better the number of clusters.

At the same time, determining whether the feeder associated with the principal element score is a normal feeder or a faulty feeder;

The optimal clustering data value k has two stratification numbers: 1 and 2:

When the optimal cluster data value k is 1, it indicates that the fault does not belong to the feeder, and the feeder associated with the principal component score is determined to be a normal feeder;

When the optimal clustering data value k is 2, it means that the fault belongs to the feeder, and the BIRCH algorithm will sensitively prompt the abnormal point (fault point) and determine that the feeder associated with the principal component score is the faulty feeder;

The BIRCH clustering algorithm has the following formula:

Among them, a(i) represents the average value of the distance from point i to all other points in the cluster it belongs to; b(i) represents the minimum value of the average value of the distance from point i to all points in a cluster that does not contain it, B is the variance between different clusters, W is the variance of all data points within the cluster, n is the total number of data points, and the CH value involves the number of clusters and the trace of the deviation matrix between classes.

Optionally, after obtaining the zero-sequence current of each feeder within a preset time window and the zero-sequence voltage of the busbar after a fault occurs in the distribution network, the method further includes:

Determining whether the zero-sequence voltage is greater than a preset phase voltage threshold;

If so, based on the zero-sequence current and the zero-sequence voltage, the feeder in the distribution network is subjected to two-dimensional processing in a preset short-time window zero-sequence instantaneous power curve cluster by KPCA algorithm to determine the corresponding principal component score.

In addition, to achieve the above-mentioned purpose, the present invention further provides a fault line selection system for a distribution network, the fault line selection system for a distribution network comprising:

A data acquisition module is used to obtain the zero-sequence current of each feeder and the zero-sequence voltage of the busbar within a preset time window after a fault occurs in the distribution network;

A numerical calculation module, used to determine the principal element score corresponding to the feeder in the distribution network in a preset short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

A logic judgment module is used to determine whether the feeder is a faulty feeder after clustering according to the principal component score.

Optionally, the data acquisition module further includes:

A zero-sequence voltage acquisition unit is used to acquire the zero-sequence voltage of the busbar through a voltage transformer installed on the busbar;

The zero-sequence current acquisition unit is used to acquire the zero-sequence current of each feeder through the current transformer installed on each feeder.

Optionally, the numerical calculation module further includes:

A signal calculation unit, used for constructing a start signal when the instantaneous value of the collected zero-sequence voltage is greater than a preset voltage threshold;

An instantaneous power curve calculation unit, used for determining a target instantaneous power curve in a short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

The KPCA calculation unit is used to determine the two-dimensional coordinates of the target instantaneous power curve based on the KPCA-BIRCH cluster analysis method, and the two-dimensional coordinates represent the fault zero-sequence power of the feeder.

Optionally, the logic judgment module further includes:

A zero-sequence voltage judgment unit, used to determine whether the zero-sequence voltage is greater than a preset phase voltage threshold;

Wherein, if yes, the step of determining the principal element score corresponding to the feeder in the distribution network in the preset short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage is performed;

A fault line selection judgment unit is used to determine whether the feeder is a faulty feeder after performing BIRCH clustering on the principal component scores.

In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium on which a fault line selection program for a distribution network is stored, and when the fault line selection program for the distribution network is executed by a processor, the steps of the fault line selection method for the distribution network as described above are implemented.

The present invention provides a fault line selection method and system for a distribution network and a readable storage medium. The method extracts the zero-sequence current of a feeder and the zero-sequence voltage of a busbar at a preset power frequency period after a fault occurs, and then calculates the principal component scores corresponding to the zero-sequence current and the zero-sequence voltage in a short-time window zero-sequence instantaneous power curve cluster. The principal component scores are used to determine whether the feeder line is a line with a fault, so that even when the current of the distribution network is small, a target fault feeder in the distribution network can be accurately and quickly determined, thereby improving the detection accuracy.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of the architecture of a fault line selection system for a distribution network according to an embodiment of the present invention;

Figure 2 is a schematic diagram of a distribution network simulation model;

FIG3 is a schematic flow chart of a first embodiment of a method for fault line selection in a distribution network according to the present invention;

FIG4 is a schematic diagram of a single-phase ground fault distribution network;

5 is a schematic flow chart of a second embodiment of a method for fault line selection in a distribution network according to the present invention;

FIG6 is a schematic flow chart of a third embodiment of a method for fault line selection in a distribution network according to the present invention;

FIG7 is a schematic diagram of a cluster of zero-sequence instantaneous power curves of each feeder in a distribution network in a short time window;

FIG8 is a schematic flow chart of a fourth embodiment of a method for fault line selection in a distribution network according to the present invention;

FIG9 is a schematic diagram showing the distribution of principal component scores of healthy lines and faulty lines obtained based on the KPCA-BIRCH cluster analysis results;

10 is a schematic flow chart of a fifth embodiment of a method for fault line selection in a distribution network according to the present invention;

The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings.

Detailed ways

The fault line selection method for the distribution network described in the present invention can be used to protect distribution networks of different voltage levels. Depending on the scenario, the method can be flexibly configured on 10-35kV overhead lines, cable lines, and overhead-cable hybrid lines. It can accurately identify single-phase grounding faults in the distribution network, and the protection action can be timely, and the fault can be isolated and cleared in time, thereby improving the stability of the power system.

In order to better understand the above technical solution, exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

As an implementation scheme, FIG1 is a schematic diagram of the architecture of the hardware operating environment of the fault line selection system of the distribution network involved in the embodiment of the present invention.

As shown in Figure 1, the system includes a data acquisition module 101, a numerical calculation module 102 and a logic judgment module 103. The data acquisition module 101 is used to obtain the zero-sequence current of each feeder and the zero-sequence voltage of the bus within a preset time window after a fault occurs in the distribution network; the numerical calculation module 102 is used to determine the corresponding principal component score of the feeder in the distribution network in the preset short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage; the logic judgment module 103 is used to determine whether the feeder is a faulty feeder after clustering the principal component score. Wherein:

The data acquisition module 101 may include a zero-sequence voltage acquisition unit 1011 and a zero-sequence current acquisition unit 1012. The zero-sequence voltage acquisition unit 1011 is used to acquire the zero-sequence voltage of the bus through the voltage transformer installed on the bus; the zero-sequence current acquisition unit 1012 is used to acquire the zero-sequence current of each feeder through the current transformer installed on each feeder.

The numerical calculation module 102 may include a signal calculation unit 1021, an instantaneous power curve calculation unit 1022, and a KPCA calculation unit 1023. The signal calculation unit 1021 is used to construct a start signal when the instantaneous value of the collected zero-sequence voltage is greater than a preset voltage threshold, the instantaneous power curve calculation unit 1022 is used to determine the target instantaneous power curve in the short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage, and the KPCA calculation unit 1023 is used to determine the two-dimensional coordinates of the target instantaneous power curve based on the KPCA-BIRCH cluster analysis method, the two-dimensional coordinates representing the fault zero-sequence power of the feeder, and the two-dimensional coordinates are used as the principal component score.

The logic judgment module 103 may include a zero-sequence voltage judgment unit 1031 and a fault line selection judgment unit 1032. The zero-sequence voltage judgment unit 1031 is used to determine whether the zero-sequence voltage is greater than a preset phase voltage threshold, wherein if so, the step of determining the principal component score corresponding to the feeder in the distribution network in the preset short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage is performed; the fault line selection judgment unit 1032 is used to determine whether the feeder is a fault feeder after performing BIRCH clustering on the principal component score.

In addition, the fault line selection system for the distribution network shown in FIG1 further includes a memory 104 and a processor 105. The memory 104 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 103 is used to store a fault line selection program for the distribution network as a computer-readable storage medium, and the processor 105 may be used to call the fault line selection program for the distribution network stored in the memory 104 and perform the following operations:

Obtain the zero-sequence current of each feeder and the zero-sequence voltage of the busbar within the preset time window after a distribution network fault occurs;

Determine, according to the zero-sequence current and the zero-sequence voltage, a principal element score corresponding to a feeder in the distribution network in a preset short-time window zero-sequence instantaneous power curve cluster;

Determine whether the feeder is a faulty feeder according to the principal component score.

In one embodiment, the processor 105 may be used to call the fault line selection program of the power distribution network stored in the memory 104, and perform the following operations:

Acquire first intercepted data, and use the first intercepted data as the upper limit of the short-time window, wherein the first intercepted data is data intercepted at a first preset time interval before the fault;

And, obtaining second intercepted data, and using the second intercepted data as the lower limit of the short-time window, wherein the second intercepted data is data intercepted at a second preset time interval after the fault;

Determine a short time window interception interval according to the short time window upper limit and the short time window lower limit;

Determine the zero-sequence instantaneous power curve cluster of the distribution network in the short-time window interception interval as the short-time window zero-sequence instantaneous power curve cluster;

Among them, the first preset time length is shorter than the second preset time length.

In one embodiment, the processor 105 may be used to call the fault line selection program of the power distribution network stored in the memory 104, and perform the following operations:

Determining a target instantaneous power curve in a short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

Based on the KPCA algorithm, determining the two-dimensional coordinates of the target instantaneous power curve, wherein the two-dimensional coordinates represent the fault zero-sequence power of the feeder;

The two-dimensional coordinates are used as the principal component scores.

In one embodiment, the processor 105 may be used to call the fault line selection program of the power distribution network stored in the memory 104, and perform the following operations:

BIRCH clustering is performed according to the principal component scores to determine whether the feeder is a faulty feeder.

In one embodiment, the processor 105 may be used to call the fault line selection program of the power distribution network stored in the memory 104, and perform the following operations:

Determining whether the zero-sequence voltage is greater than a preset phase voltage threshold;

If so, the step of determining the principal component score corresponding to the feeder in the distribution network in a preset short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage is performed.

Based on the hardware architecture of the fault line selection system of the distribution network based on the above-mentioned power system technology, an embodiment of the fault line selection method of the distribution network of the present invention is proposed.

Refer to Figure 2, which is a schematic diagram of the distribution network simulation model. The distribution network simulation model is a distribution network arc simulation model built according to the actual operation of the distribution network. PSCAD/EMTDC is used to establish the distribution network simulation model shown in Figure 2. The 110kV/10kV substation has a total of six outgoing lines, 4 overhead lines, namely L1=20km, L2=24km, L4=16km, L6=12km, and 2 pure cable lines, namely L3=16km and L5=15km. Among them, the positive sequence impedance of the overhead line is: R1=0.45Ω/km, L1=1.172mH/km, C1=6.1nF/km, and the zero sequence impedance is: R0=0.7Ω/km, L0=3.91mH/km, C0=3.8nF/km; the positive sequence impedance of the cable feeder is: R1=0.075Ω/km, L1=0.254mH/km, C1=318nF/km, and the zero sequence impedance is: R0=0.102Ω/km, L0=0.892mH/km, C0=212nF/km. The neutral point of the distribution system is not grounded, and a single-phase grounding fault is set in the simulation model. The fault points are set at 10km, 11km, 12km, 13km, 14km, and 15km away from the first busbar on the feeder L1. The initial fault angle is 90° and the transition resistance is 0Ω.

3, in a first embodiment, the fault line selection method of the distribution network includes the following steps:

Step S10, obtaining the zero-sequence current of each feeder and the zero-sequence voltage of the bus within a preset time window after a fault occurs in the distribution network;

In this embodiment, referring to FIG. 4 , which is a schematic diagram of a single-phase ground fault distribution network, when a fault occurs in the distribution network, the zero-sequence current of the feeder and the zero-sequence voltage of the bus in the distribution network at a preset period after the fault occurs are firstly collected.

Exemplarily, the distribution network fault is a single-phase grounding fault, the fault initial angle is 90°, and the transition resistance is 0Ω. The zero-sequence current of the corresponding feeder and the zero-sequence voltage of the busbar generated by the distribution network line under the single-phase grounding fault are obtained.

Optionally, the zero-sequence voltage of the busbar may be acquired through a voltage transformer installed on the busbar, and the zero-sequence current of each feeder may be acquired through a current transformer installed on each feeder.

Optionally, the preset power frequency cycle may be a quarter of the power frequency cycle after a fault.

In this embodiment, the zero-sequence current refers to the average value of the three-phase current of the feeder, and the zero-sequence voltage refers to the average value of the three-phase voltage of the bus. These data can be used as inputs for subsequent steps.

Step S20, determining the principal component score corresponding to the feeder in the distribution network in a preset short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

In this embodiment, after obtaining the zero-sequence current of the feeder and the zero-sequence voltage of the bus, the short-time window zero-sequence instantaneous power curve cluster technology is used to calculate the principal component score of each feeder within a preset time window. Specifically, the short-time window zero-sequence instantaneous power curve cluster technology can take the zero-sequence current and zero-sequence voltage of the feeder as input, and calculate the zero-sequence instantaneous power curve of the feeder within the time window. These power curves can be used to determine the score of each feeder within the preset time window.

In this embodiment, the principal component score is a data analysis technology, which is usually used for dimensionality reduction and feature extraction of multidimensional data. In this embodiment, the principal component score is used to determine the faulty feeder. The principal component score is characterized by a set of new variable values obtained after linear transformation of the original data. These variables are sorted according to the variance of the data, where the first variable is called the first principal component, the second variable is called the second principal component, and so on. Each principal component represents a specific combination of the original data, which makes the principal component explain the data changes to the greatest extent among all possible combinations.

Step S30: determining whether the feeder is a faulty feeder according to the principal component score.

In this embodiment, after the principal component scores are determined, whether the feeder is a faulty feeder is determined after clustering based on the principal component scores.

Optionally, in some embodiments, the principal component score of each feeder within a preset time window may be compared with a predefined threshold. If the score exceeds the threshold, the feeder may be determined to be a faulty feeder. Otherwise, the feeder is not a faulty feeder. Different thresholds may be set to adapt to different environments and scenarios.

Optionally, in some other implementations, BIRCH clustering is performed according to the principal component score to determine whether the feeder is a faulty feeder, and a feeder associated with the principal component score is determined to be a faulty feeder.

In the technical solution provided in this embodiment, when a fault occurs in the distribution network, the zero-sequence current of the feeder and the zero-sequence voltage of the busbar at a preset power frequency cycle after the fault occurs are extracted, and then the principal component scores corresponding to the zero-sequence current and zero-sequence voltage in the short-time window zero-sequence instantaneous power curve cluster are calculated. Clustering is performed according to the principal component scores to determine whether the feeder line is a faulty line, so that even when the current in the distribution network is small, the target fault feeder in the distribution network can be accurately and quickly determined, thereby improving the detection accuracy.

5 , in the second embodiment, based on any embodiment, step S20 includes:

Step S21, determining a target instantaneous power curve in a short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

Step S22, determining the two-dimensional coordinates of the target instantaneous power curve based on the KPCA algorithm, wherein the two-dimensional coordinates represent the fault zero-sequence power of the feeder;

Step S23: using the two-dimensional coordinates as the principal component score.

Optionally, in this embodiment, after obtaining the zero-sequence current of the feeder and the zero-sequence voltage of the busbar at a preset power frequency cycle after the distribution network fails, a target instantaneous power curve in the short-time window zero-sequence instantaneous power curve cluster is determined according to the zero-sequence current and the zero-sequence voltage.

Specifically, you can follow the steps below:

First, the zero-sequence current and zero-sequence voltage are subjected to discrete Fourier transform (DFT) to obtain their frequency spectra, where only the fundamental component (i.e., the power frequency component) needs to be considered.

Next, the zero-sequence instantaneous power is calculated and divided into several segments according to the time window, each segment contains data of several power frequency cycles. For each time segment, their zero-sequence instantaneous power curves are merged into an average curve to obtain the short-time window zero-sequence instantaneous power curve.

Next, the target instantaneous power curve is processed using the KPCA algorithm to determine the two-dimensional coordinates representing the fault zero-sequence power of the feeder. Specifically, the processing can be performed according to the following steps:

(1) For each target instantaneous power curve, calculate its first n principal component scores to obtain an n-dimensional vector.

(2) Put the n-dimensional vectors of all target instantaneous power curves into a matrix and perform KPCA dimensionality reduction to obtain a two-dimensional coordinate system.

(3) The coordinates of each target instantaneous power curve in the two-dimensional coordinate system are the required two-dimensional coordinates, and the abscissa is the first principal component score of the target instantaneous power curve.

(4) The horizontal coordinate of the two-dimensional coordinate is used as the principal component score of the feeder. Specifically, for each feeder, the horizontal coordinate of the corresponding target instantaneous power curve is used as the principal component score of the feeder.

Exemplarily, there are two principal component scores, namely KP1 and KP2, and a 2-dimensional vector instantaneous power matrix ΔP ₀ is constructed according to KP1 and KP2:

In this matrix, each sample selects 5 sampling points before the initial moment of the fault and 20 sampling points after the initial moment of the fault. A 36×25 matrix is formed by 36 historical sample data, and KPCA-BIRCH clustering is performed. The cumulative contribution rate of the fault information contained in KP1 and KP2 is greater than 97%. KP1 and KP2 are used to represent the fault zero-sequence power, and the two-dimensional coordinate X (KP1, KP2) representing the fault zero-sequence power of the feeder is obtained.

Finally, the two-dimensional coordinate X, namely (KP1, KP2), is used as the principal component score to determine the faulty feeder based on the value using BIRCH clustering.

It should be noted that the zero-sequence currents corresponding to different feeders are different, so the corresponding target instantaneous power curves in the short-time window zero-sequence instantaneous power curve cluster are also different.

In the technical solution provided in this embodiment, the target instantaneous power curve is determined by using the zero-sequence current and the zero-sequence voltage, so that the target fault feeder in the distribution network can be accurately and quickly determined even when the current of the distribution network is small, thereby improving the detection accuracy.

6 , in a third embodiment, based on any embodiment, before step S10, the following further includes:

Step S40, obtaining first intercepted data, and using the first intercepted data as the upper limit of the short-time window; and obtaining second intercepted data, and using the second intercepted data as the lower limit of the short-time window;

Step S50, determining a short time window interception interval according to the short time window upper limit and the short time window lower limit;

Step S60: determining the zero-sequence instantaneous power curve cluster of the distribution network in the short-time window interception interval as the short-time window zero-sequence instantaneous power curve cluster.

Optionally, the present embodiment provides a method for constructing a short-time window zero-sequence instantaneous power curve cluster. In the present embodiment, the size of the interception interval of the short-time window is first determined, and the first intercepted data intercepted at the first preset time length before the fault is used as the upper limit of the short-time window, and the second intercepted data intercepted at the second preset time length after the fault is used as the lower limit of the short-time window. The zero-sequence instantaneous power curve cluster of the distribution network in the short-time window interception interval is determined as the short-time window zero-sequence instantaneous power curve cluster. The first preset time length is less than the second preset time length.

Optionally, the first preset time length may be data with an interval of 0.2 ms before the fault.

Optionally, the second preset time length may be data at an interval of 1 ms after a fault.

For example, referring to Figure 7, Figure 7 is a schematic diagram of a cluster of zero-sequence instantaneous power curves of each feeder in a distribution network in a short time window. It can be seen that there is an obvious difference between the faulty line and the sound line (normal line).

In the technical solution provided in this embodiment, a short-time window zero-sequence instantaneous power curve cluster is formed by intercepting the zero-sequence instantaneous power curve cluster of the distribution network within the short-time window interval, which provides a prerequisite for the subsequent determination of the corresponding principal component score of the feeder in the distribution network in the preset short-time window zero-sequence instantaneous power curve cluster based on the zero-sequence current and zero-sequence voltage. Then, after BIRCH clustering of the data, the target fault feeder in the distribution network can be accurately and quickly determined even when the current of the distribution network is small, thereby improving the detection accuracy.

8 , in a fourth embodiment, based on any embodiment, step S30 includes:

S31, performing BIRCH clustering according to the principal component scores to determine whether the feeder is a faulty feeder.

Optionally, in this embodiment, whether the feeder is a faulty feeder is determined based on the BIRCH clustering of the principal component scores.

For example, referring to Figure 9, Figure 9 is a schematic diagram of the distribution of principal component scores of healthy lines and faulty lines obtained based on the KPCA-BIRCH cluster analysis results. The two-dimensional curve clusters are grouped, and the fault modes existing in the data are automatically and unsupervisedly discovered without anyone setting the number of clustering layers, and the internal coupling relationship of the data is mined;

At the same time, the BIRCH algorithm uses a tree model to determine the best clustering data value k. By combining the silhouette coefficient _Si and the CH index, the k value is used to iterate the best number of layers. The larger the values of the two indicators, the better the number of clusters. At the same time, it is determined whether the feeder associated with the principal component score is a normal feeder or a faulty feeder;

The value of k can be 1 or 2, two stratification numbers:

When the k value is 1, it indicates that the fault does not belong to the feeder, and the feeder associated with the principal element score is determined to be a normal feeder;

When the k value is 2, it indicates that the fault belongs to the feeder, and the BIRCH algorithm will sensitively prompt the abnormal point (fault point) and determine that the feeder associated with the principal component score is the faulty feeder;

The BIRCH clustering algorithm has the following formula:

Among them, a(i) means the average value of the distance from point i to all other points in its cluster; b(i) is the minimum value of the average value from point i to all points in a cluster that does not contain it, B is the variance between different clusters, W is the variance of all data points within the cluster, n is the total number of data points, and the CH value involves the number of clusters and the trace of the deviation matrix between classes. In the end, the two-dimensional coordinate points will be effectively divided into two feature clusters, the principal component score of the class with the least occupied will be selected, and then the corresponding fault feeder will be selected.

In the technical solution provided in this embodiment, normal feeders and faulty feeders are distinguished by BIRCH clustering of principal component scores, so that the target faulty feeder in the distribution network can be accurately and quickly determined even when the current in the distribution network is small, thereby improving detection accuracy.

Referring to FIG. 10 , in a fifth embodiment, based on any embodiment, after step S10, the following further includes:

Step S70, determining whether the zero-sequence voltage is greater than a preset phase voltage threshold;

Step S80: If yes, execute the step of determining the principal component score corresponding to the feeder in the distribution network in the preset short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage.

Optionally, in this embodiment, a bus zero-sequence voltage sensor is provided in the distribution network, and the instantaneous value of the bus zero-sequence voltage is obtained through the sensor. At the same time, a voltage threshold detection module is provided, and when the instantaneous value of the bus zero-sequence voltage is greater than a preset phase voltage threshold, the module sends a power-on signal to the fault recording line selection device.

The fault recording and line selection device starts after receiving the power-on signal and begins to record the zero-sequence voltage of each feeder within a preset period after the fault occurs. During this period, the fault recording and line selection device will transmit the recorded data to the host computer for subsequent data analysis and processing.

It should be noted that in order to ensure the accuracy and stability of the data, a high-precision voltage sensor should be selected, and the data should be properly filtered and corrected. In addition, the selection of the preset cycle should be adjusted according to the specific situation to fully ensure the integrity and accuracy of the data.

For example, let the zero-sequence voltage of the bus be , the voltage threshold is ,in, Generally, 0.15 is taken. Indicates the rated voltage of the busbar.

like more than the , the fault line selection device starts immediately and records the zero-sequence current of one cycle after the fault occurs.

In this embodiment, when the instantaneous value of the zero-sequence voltage of the bus is greater than the preset phase voltage threshold, it is preliminarily determined that a fault has occurred in the distribution network, and the fault recording and line selection device is started to execute step S20. Combined with the aforementioned fault line selection method for the distribution network, when a fault occurs in the distribution network, a judgment can be made in a timely manner and the line can be accurately selected, thereby improving the detection accuracy and ensuring the safety and stability of the system.

In addition, it can be understood by a person skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the fault line selection system of the distribution network to implement the process steps of the embodiment of the above method.

Therefore, the present invention also provides a computer-readable storage medium, which stores a fault line selection program for a distribution network. When the fault line selection program for a distribution network is executed by a processor, the various steps of the fault line selection method for a distribution network described in the above embodiment are implemented.

The computer-readable storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which are computer-readable storage media that can store program codes.

It should be noted that since the storage medium provided in the embodiment of the present application is the storage medium used to implement the method of the embodiment of the present application, based on the method introduced in the embodiment of the present application, the person skilled in the art can understand the specific structure and deformation of the storage medium, so it is not repeated here. All storage media used in the method of the embodiment of the present application belong to the scope of protection of this application.

It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

The present invention is described with reference to the flowcharts and/or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the processes and/or boxes in the flowchart and/or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

It should be noted that in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claim. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims (7)

1. The fault line selection method for the power distribution network is characterized by comprising the following steps of:

After the power distribution network fails, zero sequence current in a preset time window of each feeder line and zero sequence voltage of a bus are obtained;

According to the zero sequence current and the zero sequence voltage, two-dimensionally processing a feeder line in the power distribution network in a preset short-time window zero sequence instantaneous power curve cluster through a KPCA algorithm to determine a corresponding principal component score;

performing BIRCH clustering according to the principal component score to determine whether the feeder is a fault feeder;

The short-time window zero-sequence instantaneous power curve cluster is a two-dimensional curve cluster, and the step of determining principal component scores of the zero-sequence current and the zero-sequence voltage in the short-time window zero-sequence instantaneous power curve cluster comprises the following steps:

determining a target instantaneous power curve in a short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

Determining two-dimensional coordinates of the target instantaneous power curve based on a KPCA algorithm, wherein the two-dimensional coordinates represent fault zero-sequence power of the feeder line;

taking the two-dimensional coordinates as the principal component scores;

the two-dimensional coordinates are obtained through a KPCA algorithm, and the formula is as follows:

；

；

Wherein the method comprises the steps of Generating a matrix for a feature space,/>Representing feature samples in a feature space, N being the number of samples, K being a gram matrix, elements/>，/>Is a characteristic value;

the step of determining the fault feeder line in the power distribution network after using BIRCH clustering according to the principal component score comprises the following steps:

Grouping the two-dimensional curve clusters, finding out fault modes existing in data in an unsupervised clustering mode, and mining the internal coupling relation of the data;

Determining an optimal clustering data value k through a BIRCH algorithm, and iterating out an optimal layering number through combining the contour coefficient S _i with the CH index to jointly act on the optimal clustering data value k;

simultaneously determining that the feeder associated with the principal component score is a normal feeder or a fault feeder;

the optimal clustering data value k has two layering numbers of 1 and 2:

When the k value of the optimal cluster data value is 1, indicating that the fault does not belong to the feeder line, and determining the feeder line associated with the principal component score as a normal feeder line;

when the k value of the optimal cluster data value is 2, representing that the fault belongs to the feeder line, and determining the feeder line associated with the principal component score as the fault feeder line;

The BIRCH clustering algorithm has the following formula:

；

；

Wherein a (i) represents the average of the distances of point i to all other points in the cluster in which it is located; b (i) represents the minimum of the average of points i to all points within a cluster that does not contain it, B is the variance between different clusters, W is the variance of data points within all clusters, n is the total number of data points, and the CH value relates to the number of clusters and the trace of the dispersion matrix between the classes;

after the step of obtaining the zero sequence current in the preset time window of each feeder line and the zero sequence voltage of the bus after the power distribution network fails, the method further comprises the following steps:

determining whether the zero sequence voltage is greater than a preset phase voltage threshold;

And if so, carrying out two-dimensional processing on the feeder line in the power distribution network in a preset short-time window zero sequence instantaneous power curve cluster through a KPCA algorithm according to the zero sequence current and the zero sequence voltage so as to determine a corresponding principal component score.

2. The fault line selection method of a power distribution network according to claim 1, wherein before the step of determining the corresponding principal component score by performing two-dimensional processing on a feeder line in the power distribution network through a KPCA algorithm on a preset short-time window zero sequence instantaneous power curve cluster according to the zero sequence current and the zero sequence voltage, the fault line selection method further comprises:

acquiring first intercepted data, and taking the first intercepted data as the upper limit of a short time window, wherein the first intercepted data is intercepted data which is intercepted at intervals of a first preset duration before a fault;

obtaining second intercepted data, and taking the second intercepted data as the lower limit of a short time window, wherein the second intercepted data is intercepted data which is intercepted at intervals of a second preset time length after a fault;

Determining a short time window interception interval according to the short time window upper limit and the short time window lower limit;

Determining a zero sequence instantaneous power curve cluster of the power distribution network in the short time window interception section as the zero sequence instantaneous power curve cluster of the short time window;

The first preset duration is smaller than the second preset duration.

3. A system for implementing a fault line selection method for a power distribution network according to claim 1, the system comprising:

the data acquisition module is used for acquiring zero sequence current and zero sequence voltage of a bus in a preset time window of each feeder line after the power distribution network fails;

the numerical calculation module is used for determining corresponding principal component scores of feeder lines in the power distribution network in a preset short-time window zero sequence instantaneous power curve cluster according to the zero sequence current and the zero sequence voltage;

and the logic judgment module is used for determining whether the feeder line is a fault feeder line according to the principal component score.

4. The system of claim 3, wherein the data acquisition module further comprises:

the zero sequence voltage acquisition unit is used for acquiring the zero sequence voltage of the bus through a voltage transformer arranged on the bus;

the zero sequence current acquisition unit is used for acquiring the zero sequence current of each feeder line through a current transformer arranged on each feeder line.

5. The system of claim 3, wherein the numerical computation module further comprises:

the signal calculation unit is used for constructing a starting signal when the acquired instantaneous value of the zero sequence voltage is greater than a preset voltage threshold value;

The instantaneous power curve calculation unit is used for determining a target instantaneous power curve in a short-time window zero-sequence instantaneous power curve cluster according to the zero-sequence current and the zero-sequence voltage;

and the KPCA calculation unit is used for determining the two-dimensional coordinates of the target instantaneous power curve based on a KPCA algorithm, wherein the two-dimensional coordinates represent the fault zero sequence power of the feeder line.

6. The system of claim 3, wherein the logic determination module further comprises:

the zero sequence voltage judging unit is used for determining whether the zero sequence voltage is larger than a preset phase voltage threshold value or not;

If yes, executing the step of determining corresponding principal component scores of feeder lines in the power distribution network in a preset short-time window zero sequence instantaneous power curve cluster according to the zero sequence current and the zero sequence voltage;

and the fault line selection judging unit is used for determining whether the feeder is a fault feeder after BIRCH clustering is carried out on the principal component scores.

7. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a fault line selection program of an electrical distribution network, which when executed by a processor, implements the steps of the fault line selection method of an electrical distribution network according to any one of claims 1 to 2.