A method and system for online intelligent monitoring of transmission lines

Abstract

The present application relates to the technical field of data processing, and in particular to a method and system for online intelligent monitoring of power transmission lines. The method comprises the following steps: clustering multidimensional data points to obtain multiple clusters, calculating the degree of abnormality of the clusters; generating an early warning signal in response to the degree of abnormality being greater than a preset threshold; the multidimensional data points include: the temperature, humidity, and wind speed of the power transmission line; wherein, for any cluster, the weighted Euclidean distance between each data point contained in the cluster and the cluster center is calculated and normalized to obtain a normalized result, and the mean of the sum of the normalized results of all clusters is used as the degree of abnormality. The present application has the effect of improving the accuracy of transmission line fault identification.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a method and system for online intelligent monitoring of power transmission lines. Background Art

[0002] In modern power systems, transmission lines are a crucial component of power transmission, and their safe and stable operation directly impacts the reliability of power supply. However, traditional transmission line inspection methods have numerous limitations, such as low manual inspection frequency, slow response time, and insufficient ability to identify hidden dangers. These issues make it difficult to detect line faults in a timely manner, increasing operational risks in the power system.

[0003] In recent years, with the rapid development of smart grids, online monitoring technology has been widely used. By utilizing sensors to monitor transmission lines in real time and comprehensively analyzing multi-dimensional data on environmental factors such as temperature, humidity, and wind speed, a basis for fault prediction is provided. For example, patent application CN118209817A discloses a method and system for identifying transmission line faults based on environmental characteristics. This system fuses acquired multi-dimensional data and performs dimensionality reduction on the initial multi-dimensional fusion features to obtain target multi-dimensional fusion features. Based on these target multi-dimensional fusion features, transmission line faults are identified.

[0004] For multidimensional data on power transmission lines, due to the different distribution characteristics of data in different dimensions, dimensionality reduction can only retain some features, while ignoring other features, resulting in inaccurate transmission line fault identification results. The iterative self-organizing clustering algorithm, which clusters multidimensional data, can accurately obtain the characteristics of each dimension through clustering, thereby more accurately identifying transmission line faults.

[0005] During the clustering process, the traditional iterative self-organizing clustering algorithm updates the clusters by calculating the Euclidean distance between any multidimensional data point and the cluster center. However, since different dimensions of multidimensional data of transmission lines have different distribution characteristics, directly using the Euclidean distance to iteratively update the clusters will misclassify some multidimensional data points into other clusters, resulting in inaccurate clustering results, further affecting the fault identification results of transmission lines. Summary of the Invention

[0006] In order to solve the technical problem of large errors in fault identification of transmission lines, the present application provides a method and system for online intelligent monitoring of transmission lines.

[0007] In a first aspect, the present application provides a method for online intelligent monitoring of a transmission line, which adopts the following technical solutions:

[0008] A method for online intelligent monitoring of power transmission lines comprises the steps of: clustering multidimensional data points to obtain multiple clusters, calculating the abnormality degree of the clusters; generating an early warning signal in response to the abnormality degree being greater than a preset threshold; the multidimensional data points include: temperature, humidity, and wind speed of the power transmission line; wherein, for any cluster, the weighted Euclidean distance between each data point contained in the cluster and the cluster center is calculated and normalized to obtain a normalized result, and the mean of the sum of the normalized results of all clusters is used as the abnormality degree; the calculation formula of the weighted Euclidean distance is:

[0009] , Indicates the The first Data points and The weighted Euclidean distance between the cluster centers of the clusters, Indicates the Data points and The cluster center of the cluster is in The difference in dimensions, Indicates the The weighting coefficients of the dimensions, Indicates the total number of dimensions.

[0010] The beneficial effects are: using the respective weights of dimensions such as temperature, humidity, and wind speed to weight the dimensions to obtain the weighted Euclidean distance, and calculating the degree of anomaly based on the weighted Euclidean distance to reflect the abnormal situation of the transmission line monitoring data, so as to achieve the purpose of real-time monitoring of the transmission line status. By monitoring key parameters such as temperature, humidity, and wind speed, abnormal situations can be discovered in time, thereby effectively preventing power system failures and improving the safety and reliability of power grid operation.

[0011] Optional, weighting factor The calculation formula is: ,in, represents the total number of clusters, Indicates the The first cluster The total number of data points in each dimension, Indicates the The total number of data points involved in clustering of the dimension, Indicates the The first The periodic intensity of the dimension, Represents the standard normalization function.

[0012] The beneficial effect is that the weight of the dimension is calculated according to the periodicity strength of the same dimension in all clusters. The stronger the periodicity strength, the greater the weight corresponding to the dimension. is the confidence level of the periodicity strength of the kth cluster. The larger its value, the more data points there are in the cluster, and the higher the confidence level of the periodicity strength of the cluster. The number of data points and the periodicity strength are taken into consideration.

[0013] Optional, weighting factor The calculation formula is: ,in, represents the total number of clusters, Indicates the The first The periodic intensity of the dimension, Represents the standard normalization function.

[0014] The beneficial effects are: it is applicable to application scenarios where the distribution of data points in the cluster is uniform, or the number of data points has little effect on the periodicity intensity, and it simplifies the calculation process of the weighted coefficient, making it easier to implement and calculate.

[0015] Optionally, the periodic intensity is calculated as: ,in, Indicates the The periodic intensity of the dimensions, Indicates the The total number of data points in the dimensions, Indicates the Dimension The data value of each data point; Indicates the The mean of all data points in a dimension; Indicates the The standard deviation of the data values ​​of all data points in the dimension, Indicates An exponential function with base .

[0016] The beneficial effects are: Quantify the periodicity strength, The larger the value of , the stronger the periodicity of the dimension, and vice versa. Indicates that the dimension The greater the difference between the data point and the dimension data, the greater the value of the dimension. The greater the difference between the data point and the data of that dimension. It indicates the average level of the difference between all data points in this dimension and the data of this dimension. The smaller the value, the closer all data points in this dimension are. The higher the consistency of all data points in this cluster, the more obvious the periodic characteristics of the data in this dimension are.

[0017] Optionally, the periodic intensity is calculated as: ,in, Indicates the The periodic intensity of the dimensions, Indicates the The total number of data points in the dimensions, Indicates the Dimension The data value of each data point; Indicates the The mean of all data points in a dimension; Indicates the The standard deviation of the data values ​​of all data points in the dimension, Indicates the The standard deviation of all data points at the time of difference in the dimension, Indicates An exponential function with base .

[0018] The beneficial effects are: It indicates the average level of the difference between all data points in this dimension and the data of this dimension. The smaller its value is, the closer all data points in this dimension are. The consistency of all data points in this cluster is high, that is, the more obvious the periodic characteristics of the data in this dimension are. It indicates the degree of fluctuation of the differential moments of all data points of the dimension data. The smaller the value, the smaller the degree of fluctuation of the differential moments of all data points of the dimension data. The closer the differential moments are, the closer the time intervals of the dimension data are, that is, the stronger the periodicity of the dimension.

[0019] Optionally, during the clustering process, for any cluster in any iterative round, the data points are sorted in ascending order, and the difference values ​​of adjacent data points at corresponding moments are calculated to obtain a first-order difference sequence.

[0020] Optionally, multi-dimensional data points are clustered to obtain multiple clusters, including the following steps: in each clustering round, for any data point, the data point is divided into the cluster where the cluster center with the smallest initial weighted Euclidean distance to the data point is located, all data points are divided, and then the cluster center of each cluster is updated according to the data values ​​of all data points in different dimensions in each cluster, until the preset clustering end condition is reached and the clustering is stopped, thereby obtaining multiple clusters.

[0021] In a second aspect, the present application provides an online intelligent monitoring system for transmission lines, which adopts the following technical solutions:

[0022] A transmission line online intelligent monitoring system comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned transmission line online intelligent monitoring method is implemented.

[0023] The beneficial effect is that the above-mentioned online intelligent monitoring method for transmission lines is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.

[0024] This application has the following technical effects:

[0025] 1. The weighted Euclidean distance is obtained by weighting the dimensions such as temperature, humidity, and wind speed using their respective weights. The degree of anomaly is calculated based on the weighted Euclidean distance to reflect the abnormality of the transmission line monitoring data, thereby achieving the purpose of real-time monitoring of the transmission line status. By monitoring key parameters such as temperature, humidity, and wind speed, anomalies can be discovered in a timely manner, thereby effectively preventing power system failures and improving the safety and reliability of power grid operation.

[0026] 2. Calculate the weight of the dimension based on the periodicity strength of the same dimension across all clusters. The stronger the periodicity strength, the greater the weight corresponding to the dimension. The calculation of the weighting coefficient takes into account both the number of data points and the periodicity strength. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] By reading the detailed description below with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present application will become readily understood. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding numbers represent the same or corresponding parts.

[0028] Figure 1 This is a method flow chart of an online intelligent monitoring method for transmission lines in an embodiment of the present application.

[0029] Figure 2 This is a method flow chart of step S1 in a method for online intelligent monitoring of transmission lines in an embodiment of the present application. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0031] It should be understood that when the terms "first," "second," etc. are used in the claims, specification, and drawings of this application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the specification and claims of this application indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0032] The present application discloses an online intelligent monitoring method for a power transmission line. Figure 1 , including steps S1 and S2, specifically as follows:

[0033] S1: Cluster the multi-dimensional data points to obtain multiple clusters, and calculate the abnormality degree of the clusters.

[0034] In one embodiment, a temperature sensor, humidity sensor, and wind speed meter are placed on a transmission tower to collect temperature, humidity, and wind speed data on the transmission line. The collection cycle is preset to 2 seconds. The data collected in each dimension at the same time is treated as a multidimensional data point.

[0035] Clustering is performed on multi-dimensional data points to obtain multiple clusters. This application uses an iterative self-organizing clustering algorithm for clustering.

[0036] The iterative self-organizing clustering algorithm continuously optimizes clusters by assigning data points to clusters with the closest cluster centers over multiple clustering rounds. Each clustering round recalculates the cluster centers of each cluster until the cluster centers no longer change or the preset number of iterations is reached. This iterative self-organizing clustering algorithm is currently available and will not be described in detail here.

[0037] Traditional iterative self-organizing clustering algorithms calculate the Euclidean distance between any data point and all cluster centers during repeated clustering rounds, and then assign the data point to the cluster with the cluster center that has the smallest Euclidean distance. However, because different dimensions have different impacts on clustering results, it is necessary to weight each dimension using its weights to obtain the weighted Euclidean distance from the data point to all cluster centers.

[0038] Specifically, in the process of continuous iteration of clustering rounds, in order to facilitate distinction and description, the weighted Euclidean distance between any data point in the cluster and the cluster center of any cluster is defined as the initial weighted Euclidean distance, denoted as After clustering is completed, the weighted Euclidean distance from any data point in the cluster to the cluster center of the cluster is defined as the weighted Euclidean distance, recorded as .

[0039] The calculation formula for the initial weighted Euclidean distance is

[0040] , Indicates the Data points and The initial weighted Euclidean distance of the cluster centers of the clusters, Indicates the Data points and The cluster center of the cluster is in The difference in dimensions, Indicates the The weighting coefficients of the dimensions, Indicates the total number of dimensions. The calculation method of the weighted coefficient is the same as that of the subsequent weighted coefficients, which will not be described here. The mean of all data points in the cluster is used as the cluster center of the cluster. The data point is not necessarily in a cluster.

[0041] Reference Figure 2 , step S1 includes steps S10 to S12, which are specifically as follows:

[0042] S10: In each clustering round, for any data point, the data point is divided into the cluster where the cluster center with the smallest initial weighted Euclidean distance to the data point is located. All data points are divided, and then the cluster center of each cluster is updated according to the data values ​​of all data points in different dimensions in each cluster.

[0043] S11: After the last iteration is completed, clustering is completed and the clustering result is obtained.

[0044] The clustering end condition may be that the cluster center no longer changes or the number of clustering iterations reaches a preset number, and clustering is stopped. The preset number of clustering iterations can be determined by the implementer according to the specific implementation situation and is not specifically limited in this application. For example, the number of clustering iterations is 100.

[0045] S12: For any cluster, calculate the weighted Euclidean distance between each data point contained in the cluster and the cluster center and normalize it to obtain the normalized result. Traverse all clusters and take the mean of the sum of the normalized results of all clusters as the abnormality degree.

[0046] In one embodiment, the calculation of the abnormality level can be expressed as follows:

[0047] Where, Indicates the abnormality degree of the transmission line monitoring data, The total number of clusters representing the clustering results, represents the cluster number, Indicates the The standard deviation of the weighted Euclidean distance between all data points in a cluster and the cluster center of the cluster, Represents the standard normalization function.

[0048] For any cluster, if The smaller the value, the more concentrated the data in the cluster is, and the lower the abnormality of the cluster is. The average abnormality of all clusters is taken as the abnormality of the transmission line monitoring data.

[0049] , Indicates the The first Data points and The weighted Euclidean distance between the cluster centers of the clusters, Indicates the Data points and The cluster center of the cluster is in The difference in dimensions, Indicates the The weighting coefficients of the dimensions, Indicates the total number of dimensions.

[0050] In one embodiment, the weighting coefficients The calculation formula is:

[0051] ,in, represents the total number of clusters, Indicates the The first cluster The total number of data points in each dimension, Indicates the The total number of data points involved in clustering of the dimension, Indicates the The first The periodic intensity of the dimension, Represents the standard normalization function.

[0052] The weight of the dimension is calculated based on the periodicity strength of the same dimension in all clusters. The stronger the periodicity strength, the greater the weight corresponding to the dimension. is the confidence level of the periodicity strength of the kth cluster. The larger its value, the more data points there are in the cluster, and the higher the confidence level of the periodicity strength of the cluster. The number of data points and the periodicity strength are taken into consideration.

[0053] In one embodiment, the weighting coefficients The calculation formula can also be:

[0054] ,in, represents the total number of clusters, Indicates the The first The periodic intensity of the dimension, This embodiment is suitable for application scenarios where the data points in the clusters are evenly distributed or the number of data points has little effect on the periodicity intensity, and simplifies the calculation process of the weighting coefficient, making it easier to implement and calculate.

[0055] During the clustering process, multidimensional data points with relatively close Euclidean distances will be clustered in the same cluster. Since Euclidean distance is an intuitive distance measurement between multidimensional data points, for transmission line monitoring data, due to the differences in data of different dimensions, some multidimensional data are very close in Euclidean distance, but have large differences in some dimensions.

[0056] Therefore, for multidimensional data, different dimensions have different impacts on the overall data. Traditional clustering algorithms cannot effectively account for periodicity in data with significant periodicity, resulting in reduced clustering effectiveness. For example, in transmission line monitoring data, temperature and humidity exhibit certain periodicity over time, significantly impacting clustering results. Furthermore, for any dimension, the stronger the periodicity, increasing the weight of that dimension can more effectively cluster data points with similar periods into the same cluster, optimizing the clustering results. Furthermore, by increasing the weight of dimensions with strong periodicity, the impact of non-critical or noisy dimensions can be relatively reduced, thereby improving overall clustering accuracy.

[0057] Therefore, the present application obtains the periodicity strength of each dimension and obtains the weight of the dimension according to the periodicity strength of the dimension.

[0058] In one embodiment, the periodicity strength is calculated as follows:

[0059] In the clustering process, for any cluster in any iterative round, the data points are sorted in ascending order, and the corresponding time of the next multidimensional data point is subtracted from the collection time of the previous multidimensional data point to obtain the differential time of the data point, that is, the differential value of the corresponding time of adjacent data points is calculated to obtain the first-order difference sequence.

[0060] Calculate the periodic intensity using the formula: ,in, Indicates the The periodic intensity of the dimensions, Indicates the The total number of data points in the dimensions, Indicates the Dimension The data value of each data point; Indicates the The mean of all data points in a dimension; Indicates the The standard deviation of the data values ​​of all data points in the dimension, Indicates the The standard deviation of all data points at the time of difference in the dimension, Indicates An exponential function with base .

[0061] The larger the value of , the stronger the periodicity of the dimension, and vice versa. Indicates that the dimension The greater the difference between the data point and the dimension data, the greater the value of the dimension. The greater the difference between the data point and the data of that dimension.

[0062] Depend on By transforming the formula, we can get: It indicates the average level of the difference between all data points in this dimension and the data of this dimension. The smaller the value, the closer all data points in this dimension are. The higher the consistency of all data points in this cluster, the more obvious the periodic characteristics of the data in this dimension are.

[0063] It indicates the degree of fluctuation of the differential moments of all data points of the dimension data. The smaller the value, the smaller the degree of fluctuation of the differential moments of all data points of the dimension data. The closer the differential moments are, the closer the time intervals of the dimension data are, that is, the stronger the periodicity of the dimension.

[0064] At this point, the periodicity strength of any cluster in any cluster can be obtained.

[0065] In one embodiment, the calculation formula for periodicity intensity may also be:

[0066] ,in, Indicates the The periodic intensity of the dimensions, Indicates the The total number of data points in the dimensions, Indicates the Dimension The data value of each data point; Indicates the The mean of all data points in a dimension; Indicates the The standard deviation of the data values ​​of all data points in the dimension, Indicates This embodiment provides a simple and direct method for calculating periodic intensity, which is suitable for application scenarios requiring less computational effort.

[0067] S2: In response to the abnormality level being greater than a preset threshold, a warning signal is generated.

[0068] If the abnormality of the transmission line monitoring data is greater than the preset threshold , then generate an early warning signal to prompt, otherwise, continue monitoring without alarming. Preset threshold The implementation personnel can limit the value according to the specific implementation situation. For example, the preset threshold The value is set to 0.7. The early warning signal may be a text message such as "abnormal warning" sent to the monitoring platform.

[0069] An embodiment of the present application further discloses an online intelligent monitoring system for transmission lines, comprising a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, the online intelligent monitoring method for transmission lines according to the present application is implemented.

[0070] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0071] In this application, the aforementioned memory can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of, accessible to, or connected to the device.

[0072] Although this specification has shown and described a number of embodiments of the present application, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, variations, and alternatives without departing from the spirit and scope of the present application. It should be understood that in practicing the present application, various alternatives to the embodiments of the present application described herein may be employed.

[0073] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for online intelligent monitoring of power transmission lines, characterized in that: Including steps: Cluster multi-dimensional data points to obtain multiple clusters, and calculate the abnormality degree of the clusters; In response to the abnormality level being greater than a preset threshold, an early warning signal is generated; the multi-dimensional data points include: temperature, humidity and wind speed of the transmission line; For any cluster, the weighted Euclidean distance between each data point in the cluster and the cluster center is calculated and normalized to obtain the normalized result. The mean of the sum of the normalized results of all clusters is used as the abnormality degree. The calculation formula of the weighted Euclidean distance is: , Indicates the The first Data points and The weighted Euclidean distance between the cluster centers of the clusters, Indicates the Data points and The cluster center of the cluster is in The difference in dimensions, Indicates the The weighting coefficients of the dimensions, Indicates the total number of dimensions; Weighting coefficient The calculation formula is: ,in, represents the total number of clusters, Indicates the The first cluster The total number of data points in each dimension, Indicates the The total number of data points involved in clustering of the dimension, Indicates the The first The periodic intensity of the dimension, represents the standard normalization function; or Weighting coefficient The calculation formula is: ,in, represents the total number of clusters, Indicates the The first The periodic intensity of the dimension, represents the standard normalization function; The calculation formula of periodic intensity is: ,in, Indicates the The periodic intensity of the dimensions, Indicates the The total number of data points in the dimensions, Indicates the Dimension The data value of each data point; Indicates the The mean of all data points in a dimension; Indicates the The standard deviation of the data values ​​of all data points in the dimension, Indicates An exponential function with base .

2. The method for online intelligent monitoring of power transmission lines according to claim 1, characterized in that: The calculation formula of periodic intensity is: ,in, Indicates the The periodic intensity of the dimensions, Indicates the The total number of data points in the dimensions, Indicates the Dimension The data value of each data point; Indicates the The mean of all data points in a dimension; Indicates the The standard deviation of the data values ​​of all data points in the dimension, Indicates the The standard deviation of all data points at the time of difference in the dimension, Indicates An exponential function with base .

3. The method for online intelligent monitoring of power transmission lines according to claim 2, characterized in that: In the clustering process, for any cluster in any iterative round, the data points are sorted in ascending order, and the difference values ​​of adjacent data points at corresponding moments are calculated to obtain the first-order difference sequence.

4. The method for online intelligent monitoring of power transmission lines according to claim 1, characterized in that: Clustering multi-dimensional data points to obtain multiple clusters includes the following steps: In each clustering round, for any data point, the data point is divided into the cluster where the cluster center with the smallest initial weighted Euclidean distance to the data point is located. All data points are divided, and then the cluster center of each cluster is updated according to the data values ​​of all data points in different dimensions in each cluster until the preset clustering end condition is reached and clustering is stopped, resulting in multiple clusters.

5. An online intelligent monitoring system for power transmission lines, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for online intelligent monitoring of a power transmission line according to any one of claims 1 to 4 is implemented.

Images