Method and system for identifying weak points of power distribution network line network frame

Abstract

The invention relates to the technical field of power grid weak point identification, and discloses a power distribution network line network frame weak point identification method and system, and the method comprises the steps: collecting historical data of each node of a power distribution network line network frame, and carrying out the preprocessing; performing feature extraction according to the preprocessed historical data to obtain key feature data of each node; taking the key feature data corresponding to each sampling time point of each node as a sample, and constructing a sample set; clustering data in the sample set through a clustering algorithm to obtain P classification clusters; acquiring real-time data of each node of the power distribution network line network frame, performing feature extraction to obtain real-time key feature data of each node, and calculating a classification cluster to which each node belongs; and marking the nodes belonging to the target classification cluster as weak points. According to the method, the identification result can better conform to the actual situation, and dynamic monitoring and immediate response can be realized.

Description

Technical Field

[0001] This invention relates to the field of power grid weak point identification technology, and in particular to a method and system for identifying weak points in distribution network line structures. Background Technology

[0002] Weak points in the power distribution network structure refer to critical nodes or areas that may affect the stability and reliability of the power system. Identifying and addressing these weak points helps ensure the continuity and security of power supply.

[0003] Existing technologies typically use association rule methods to mine hidden patterns and relationships in power system operation data, thereby locating potential problem areas. Association rule methods are mainly used to discover static patterns and frequent itemsets, but they are difficult to capture instantaneous changes and dynamic behaviors in power systems. Since association rule methods are usually based on batch processing of historical data, it is difficult to achieve real-time monitoring and immediate response, resulting in processing delays. Rule updates usually require retraining the model, leading to slow response speeds and making them unsuitable for application scenarios that require rapid decision-making. Although the rule sets generated by association rule methods are easy to understand, it is not intuitive to apply these rules to actual operations, especially in complex power systems, where a large number of redundant rules may be generated, increasing the workload of manual screening and interpretation.

[0004] Therefore, there is an urgent need for a method and system for identifying weak points in the power distribution network structure, which can make the identification results more consistent with the actual situation and enable dynamic monitoring and real-time response. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a method for identifying weak points in the power distribution network structure, which can make the identification results more consistent with the actual situation and enable dynamic monitoring and real-time response.

[0007] To address the aforementioned technical problems, this invention provides the following technical solution: a method for identifying weak points in a power distribution network, comprising: collecting historical data of each node in the power distribution network and preprocessing it; extracting features from the preprocessed historical data to obtain key feature data of each node; constructing a sample set by treating the key feature data corresponding to each sampling time point of each node as a sample; clustering the data in the sample set using a clustering algorithm to obtain P classification clusters; acquiring real-time data of each node in the power distribution network, extracting features to obtain real-time key feature data of each node, and calculating the classification cluster to which each node belongs; and marking nodes belonging to the target classification cluster as weak points.

[0008] As a preferred embodiment of the method for identifying weak points in a power distribution network structure as described in this invention, the key feature data includes average voltage, maximum current, power factor, active power and reactive power, voltage fluctuation rate, daily maximum load and load change rate, voltage change rate and frequency of extreme values.

[0009] As a preferred embodiment of the method for identifying weak points in a power distribution network structure according to the present invention, the construction of the sample set includes: for each node, constructing a feature vector based on the key feature data at each sampling time point, and constructing a sample set based on the feature vectors of all nodes.

[0010] As a preferred embodiment of the method for identifying weak points in a power distribution network structure according to the present invention, the step of clustering the data in the sample set using a clustering algorithm includes obtaining an initial classification cluster by clustering the data in the sample set using a clustering algorithm;

[0011] Determine the classification to which the initial classification cluster belongs based on the key feature data of the core points of the initial classification cluster;

[0012] The initial clusters belonging to the same category are merged to obtain P clusters.

[0013] As a preferred embodiment of the method for identifying weak points in a power distribution network structure according to the present invention, the step of determining the category to which the initial classification cluster belongs based on the key feature data of the core points of the initial classification cluster includes,

[0014] The initial classification clusters are defined to belong to the following categories: normal, first deviation, second deviation, and abnormal.

[0015] Set the threshold range for each key feature data and the degree of deviation of each threshold range;

[0016] The threshold range includes a normal threshold range, a first deviation threshold range, a second deviation threshold range, and an abnormal threshold range. The deviation of the normal range is 0, the deviation of the first deviation threshold range is 1, the deviation of the second deviation threshold range is 2, and the deviation of the abnormal threshold range is 3.

[0017] Compare the key feature data of the core point of each initial classification cluster with the threshold range of each key feature to determine whether all key feature data in the core point are within the normal threshold range; if yes, mark the classification of the initial classification cluster corresponding to the core point as normal; otherwise, proceed to the next step.

[0018] The category corresponding to the threshold range with the greatest deviation from the threshold range of the key feature data in the core point is marked as the category to which the initial classification cluster of the core point belongs.

[0019] As a preferred embodiment of the method for identifying weak points in a power distribution network structure according to the present invention, wherein: P classification clusters are obtained, where P is 4; the classification clusters include normal data clusters, first deviation data clusters, second deviation data clusters, and abnormal data clusters.

[0020] As a preferred embodiment of the method for identifying weak points in a power distribution network structure according to the present invention, the calculation of the classification cluster to which each node belongs includes: extracting features from the real-time data of each node in the power distribution network structure to obtain real-time key feature data of each node, and obtaining new samples of each node.

[0021] Calculate the distance from the new sample of each node to the cluster center of each classification cluster, and assign the new sample of each node to the nearest classification cluster.

[0022] This invention provides a system for identifying weak points in the power distribution network structure.

[0023] As a preferred embodiment of the distribution network line structure weak point identification system of the present invention, it includes: a data acquisition module, a feature extraction module, a sample set construction module, a clustering module, a classification module and an identification module;

[0024] The data acquisition module is used to collect historical data from each node of the power distribution network line structure and perform preprocessing.

[0025] The feature extraction module is used to extract features based on the preprocessed historical data to obtain key feature data for each node;

[0026] The sample set construction module is used to construct a sample set by taking the key feature data corresponding to each sampling time point of each node as a sample.

[0027] The clustering module is used to cluster the data in the sample set using a clustering algorithm to obtain P classification clusters;

[0028] The classification module is used to acquire real-time data of each node of the power distribution network, extract features to obtain real-time key feature data of each node, and calculate the classification cluster to which each node belongs.

[0029] The identification module is used to mark nodes belonging to the target classification cluster as weak points.

[0030] The present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of a method for identifying weak points in a power distribution network.

[0031] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of a method for identifying weak points in a power distribution network.

[0032] The beneficial effects of this invention are as follows: This invention considers electrical parameters and incorporates multi-source information such as temperature data, event logs, load characteristics, and meteorological data, providing a more comprehensive approach. When extracting features, in addition to static features, dynamic features such as voltage change rate and extreme value frequency are introduced, which can better reflect the instantaneous state and potential problems of the system. Using clustering methods, natural groupings and patterns in the data can be discovered even without a clear target. By setting specific threshold ranges for each key feature data and assigning corresponding deviations, the consistency and operability of the classification criteria are ensured, improving the accuracy and reliability of classification. It also simplifies the process, making the identification results more consistent with the actual situation, allowing for targeted maintenance and repair measures for different weak points, optimizing resource allocation. For newly acquired real-time data, its key features can be quickly calculated and assigned to the corresponding clusters, achieving dynamic monitoring and immediate response. Attached Figure Description

[0033] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a schematic flowchart of a method for identifying weak points in a power distribution network structure, provided as an embodiment of the present invention.

[0035] Figure 2 This is a schematic diagram of the working module of a distribution network line weak point identification system provided in one embodiment of the present invention. Detailed Implementation

[0036] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0037] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for identifying weak points in a power distribution network structure, including:

[0038] S1: Collect historical data of each node in the power distribution network and perform preprocessing.

[0039] Furthermore, the historical data for each node includes electrical parameters, temperature data, event logs, load characteristics, and meteorological data;

[0040] Among them, electrical parameters may include voltage, current, active power, reactive power, power factor, voltage fluctuation, etc.

[0041] Temperature data can include ambient temperature, conductor temperature, transformer oil temperature, etc.

[0042] Event logs can include trip records, alarm information, maintenance records, etc.

[0043] Load characteristics can include load curves, peak loads, etc.

[0044] Meteorological data can include weather conditions, extreme weather conditions, etc.

[0045] Preprocessing can include operations such as missing value handling, noise filtering, and normalization.

[0046] S2: Extract features from the preprocessed historical data to obtain key feature data for each node.

[0047] Furthermore, key characteristic data and new key characteristic data are determined based on actual conditions or through expert experience methods, time series analysis, and other methods. The key characteristic data in this embodiment include: average voltage, maximum current, power factor, active power and reactive power, voltage fluctuation rate, daily maximum load and load change rate, voltage change rate, and frequency of extreme values.

[0048] By extracting key features (such as average voltage, power factor, and load change rate) from preprocessed historical data, data dimensionality reduction and feature enhancement were achieved. Features highly correlated with weak points in the grid structure (such as voltage fluctuation rate and frequency of extreme values) were selected, and irrelevant noise was filtered out, making subsequent analysis more focused on the core factors of electrical performance and load behavior.

[0049] S3: Take the key feature data corresponding to each sampling time point of each node as a sample and construct a sample set.

[0050] Furthermore, for each node, a feature vector is constructed based on the key feature data at each sampling time point.

[0051] The eigenvector expression is as follows:

[0052] X(n)=[f i(n,t)], i∈{1,2,...,m}, t∈{1,2,...,T}

[0053] Where n represents the nth node, t represents the tth sampling time point, T represents the total number of sampling time points, X(n) represents the feature vector of node n, i.e., a sample, i represents the i-th key feature data, m represents the total number of key feature data, and f i (n,t) represents the value of the i-th key feature data of node n at sampling time point t.

[0054] The sample set is constructed based on the feature vectors of all nodes, and the expression for the sample set is as follows:

[0055] D = {X(n) | n∈N}

[0056] Where D represents the sample set and N represents the total number of nodes.

[0057] In this embodiment, the key feature data (such as average voltage and maximum current) of each node at each sampling time point are organized into a feature vector, and then the feature vectors of all nodes are summarized to form a sample set.

[0058] In an optional embodiment, the sample set can also be constructed using a time-series sliding window. Specifically, for each node, a sliding window of fixed size is used; key feature data from multiple consecutive time points within the window are combined into a sample; for each sampling time point t (except the starting point), a new sample is generated, and the feature vector includes the key feature values ​​of all time points within the window; the feature dimension is expanded to an m×m window size (m is the number of key features);

[0059] Traverse all nodes and valid time points, and summarize all window samples to form a sample set; each sample represents the state evolution of a node within a short window.

[0060] In another alternative embodiment, the sample set can also be constructed by time interval aggregation, specifically, for each node, key feature data is aggregated according to a fixed time interval (such as one day); statistics for each interval are calculated, such as the mean of the daily maximum load and the variance of the voltage change rate.

[0061] Each node generates a sample for each time interval, and the feature vector consists of aggregated statistical values; the original time series data is compressed into statistical features, and samples from all nodes and all time intervals are aggregated to form a sample set; each sample represents the overall state of the node within the interval.

[0062] S4: Cluster the data in the sample set using a clustering algorithm to obtain P classification clusters.

[0063] Furthermore, clustering algorithms are used to cluster the data in the sample set to obtain several initial classification clusters.

[0064] The DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm was used to cluster the data in the sample set, resulting in several initial classification clusters.

[0065] The classification of an initial classification cluster is determined based on the key feature data of the core points of several initial classification clusters.

[0066] The initial classification cluster is defined to belong to the following categories: normal, first deviation, second deviation, and abnormal.

[0067] You can also define the initial classification cluster to which it belongs based on the actual situation.

[0068] Set the threshold range for each key feature data and the degree of deviation for each threshold range. The threshold range includes the normal threshold range, the first deviation threshold range, the second deviation threshold range, and the abnormal threshold range. The deviation of the normal range is 0, the deviation of the first deviation threshold range is 1, the deviation of the second deviation threshold range is 2, and the deviation of the abnormal threshold range is 3.

[0069] The key feature data of the core points of each initial classification cluster are compared with the threshold range of each key feature to determine whether all key feature data of the core point are within the normal threshold range. If so, the classification of the initial classification cluster corresponding to the core point is marked as normal; otherwise, proceed to the next step.

[0070] The category corresponding to the threshold range with the greatest deviation from the threshold range of the key feature data in the core point is marked as the category to which the initial classification cluster of the core point belongs.

[0071] For example, suppose that among the key feature data of the core point of an initial classification cluster, there are key feature data in the normal threshold range, key feature data in the first deviation threshold range, and key feature data in the second deviation threshold range, then the classification of the initial classification cluster corresponding to the core point is the second deviation.

[0072] If, in the core point of an initial classification cluster, there are key feature data within the normal threshold range, key feature data within the second deviation threshold range, and key feature data within the abnormal threshold range, then the classification of the initial classification cluster corresponding to that core point is abnormal.

[0073] The initial clusters belonging to the same category are merged to obtain P clusters.

[0074] In this embodiment, P is 4, and the classification clusters include: normal data cluster, first deviation data cluster, second deviation data cluster, and abnormal data. Among them, the normal data cluster represents nodes that are running completely normally and do not constitute a vulnerability; the first deviation data cluster represents nodes that slightly deviate from the normal range and only need to be monitored, and do not constitute a vulnerability; the second deviation data cluster represents nodes that significantly deviate from the normal range, such as approaching the limit value, and need to be paid attention to immediately, and may constitute a vulnerability; abnormal data represents nodes that have experienced failures or extreme events and definitely constitute a vulnerability.

[0075] By deeply integrating clustering technology with power grid operation and maintenance logic, the system prioritizes tasks through four-level classification to optimize resource allocation; it mitigates data noise by adhering to the principles of core points and maximizing deviations; and it outputs interpretable classification labels to directly guide maintenance actions.

[0076] S5: Obtain real-time data of each node in the power distribution network, extract features to obtain real-time key feature data of each node, and calculate the classification cluster to which each node belongs.

[0077] For the real-time data of each node in the distribution network, feature extraction is performed to obtain the real-time key feature data of each node, and new samples of each node are obtained.

[0078] Calculate the distance from the new sample of each node to the cluster center of each class cluster, and assign the new sample of each node to the nearest class cluster. The distance from the new sample of each node to the cluster center of each class cluster can be calculated using the Euclidean distance formula, as follows:

[0079]

[0080] Wherein, d(X(n)',C j f represents the distance from a new sample at node n to the cluster center of the j-th cluster, Cj represents the j-th cluster, and f i (n,t')' represents the value of the i-th key feature data of node n at the current time point t' (real-time acquisition), μ ji This represents the value of the i-th key feature data of the cluster center of the j-th classification cluster.

[0081] S6: Mark nodes belonging to the target classification cluster as weak points.

[0082] In this embodiment, the target classification cluster includes a second deviation data cluster and an anomaly data cluster.

[0083] Example 2 is the second embodiment of the present invention, which differs from the previous embodiment in that:

[0084] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0085] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0086] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0087] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0088] Example 3, referring to Figure 2 As an embodiment of the present invention, a weak point identification system for power distribution network lines is provided, comprising:

[0089] The data acquisition module is used to collect historical data from each node of the power distribution network and perform preprocessing.

[0090] The feature extraction module, connected to the data acquisition module, is used to extract features from preprocessed historical data to obtain key feature data for each node.

[0091] The sample set construction module, connected to the feature extraction module, is used to construct a sample set by taking the key feature data corresponding to each sampling time point of each node as a sample.

[0092] The clustering module, connected to the sample set construction module, is used to cluster the data in the sample set using a clustering algorithm to obtain several classification clusters;

[0093] The classification module, connected to the clustering module, is used to obtain real-time data of each node in the power distribution network, extract features to obtain real-time key feature data of each node, and calculate the classification cluster to which each node belongs.

[0094] The identification module, connected to the classification module, is used to mark nodes belonging to the target classification cluster as weak points.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for identifying weak points in a power distribution network structure, characterized in that: include, Collect historical data from each node of the power distribution network and perform preprocessing. Feature extraction is performed on the preprocessed historical data to obtain the key feature data of each node; Each node's key feature data at each sampling time point is treated as a sample to construct a sample set. The data in the sample set is clustered using a clustering algorithm to obtain P classification clusters; Real-time data of each node in the power distribution network structure is acquired, features are extracted to obtain real-time key feature data of each node, and the classification cluster to which each node belongs is calculated. Nodes belonging to the target classification cluster are marked as weak points.

2. The method for identifying weak points in a power distribution network structure as described in claim 1, characterized in that: The key characteristic data include average voltage, maximum current, power factor, active power and reactive power, voltage fluctuation rate, daily maximum load and load change rate, voltage change rate and frequency of extreme values.

3. The method for identifying weak points in a power distribution network structure as described in claim 2, characterized in that: The construction of the sample set includes, for each node, constructing a feature vector based on the key feature data at each sampling time point, and constructing a sample set based on the feature vectors of all nodes.

4. The method for identifying weak points 3 in a power distribution network structure as described in claim 3, characterized in that: The step of clustering the data in the sample set using a clustering algorithm includes clustering the data in the sample set using a clustering algorithm to obtain initial classification clusters; Determine the classification to which the initial classification cluster belongs based on the key feature data of the core points of the initial classification cluster; The initial clusters belonging to the same category are merged to obtain P clusters.

5. The method for identifying weak points in a power distribution network structure as described in claim 4, characterized in that: The process of determining the classification of the initial classification cluster based on the key feature data of the core points of the initial classification cluster includes, The initial classification clusters are defined to belong to the following categories: normal, first deviation, second deviation, and abnormal. Set the threshold range for each key feature data and the degree of deviation of each threshold range; The threshold range includes a normal threshold range, a first deviation threshold range, a second deviation threshold range, and an abnormal threshold range. The deviation of the normal range is 0, the deviation of the first deviation threshold range is 1, the deviation of the second deviation threshold range is 2, and the deviation of the abnormal threshold range is 3. Compare the key feature data of the core point of each initial classification cluster with the threshold range of each key feature to determine whether all key feature data in the core point are within the normal threshold range; if yes, mark the classification of the initial classification cluster corresponding to the core point as normal; otherwise, proceed to the next step. The category corresponding to the threshold range with the greatest deviation from the threshold range of the key feature data in the core point is marked as the category to which the initial classification cluster of the core point belongs.

6. The method for identifying weak points in a power distribution network as described in claim 5, characterized in that: The resulting P classification clusters include normal data clusters, first deviation data clusters, second deviation data clusters, and abnormal data clusters.

7. The method for identifying weak points in a power distribution network as described in claim 6, characterized in that: The calculation of the classification cluster to which each node belongs includes extracting features from the real-time data of each node in the distribution network line structure to obtain the real-time key feature data of each node, and obtaining new samples of each node. Calculate the distance from the new sample of each node to the cluster center of each classification cluster, and assign the new sample of each node to the nearest classification cluster.

8. A system for identifying weak points in a power distribution network, using the method for identifying weak points in a power distribution network as described in any one of claims 1 to 7, characterized in that, include: The module includes a data acquisition module, a feature extraction module, a sample set construction module, a clustering module, a classification module, and a recognition module. The data acquisition module is used to collect historical data from each node of the power distribution network line structure and perform preprocessing. The feature extraction module is used to extract features based on the preprocessed historical data to obtain key feature data for each node; The sample set construction module is used to construct a sample set by taking the key feature data corresponding to each sampling time point of each node as a sample. The clustering module is used to cluster the data in the sample set using a clustering algorithm to obtain P classification clusters; The classification module is used to acquire real-time data of each node of the power distribution network, extract features to obtain real-time key feature data of each node, and calculate the classification cluster to which each node belongs. The identification module is used to mark nodes belonging to the target classification cluster as weak points.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for identifying weak points in the power distribution network structure according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for identifying weak points in the power distribution network structure according to any one of claims 1 to 7.

Images