Wind deflection monitoring and early warning system and device for ice-covered distribution lines

Abstract

The present invention relates to the technical field of angle measurement, and particularly relates to a wind deflection monitoring and early warning system and device for ice-covered distribution lines, which acquires a sequence of wind deflection angle values of the distribution line; determines whether the wind deflection of the distribution line is abnormal based on the correlation between the sequence of wind deflection angle values and the sequence of wind force values; if the wind deflection of the distribution line is abnormal, obtains the bending degree of the distribution line according to the line shape of the distribution line; detects the severity of the shaking of the distribution line to obtain the degree of ice and frost influence on the distribution line; determines the reason for the abnormal wind deflection angle value of the distribution line according to the magnitude relationship between the bending degree and the degree of ice and frost influence, thereby improving the analysis accuracy of the reason for the abnormal wind deflection angle value of the distribution line.

Description

Technical Field

[0001] The present invention relates to the technical field of angle measurement, and particularly to a wind deviation monitoring and early warning system and device for icing distribution lines. Background Art

[0002] The wind deviation of a distribution line refers to the phenomenon of deviation or inclination of the distribution line under the action of wind. The wind deviation of the distribution line has the following impacts: insufficient safety distance: the wind deviation may cause the distance between the distribution line and surrounding objects (such as trees, buildings) to be insufficient, leading to short circuits or discharges; increased mechanical stress: the mechanical stress borne by the distribution line and the pole tower increases, which may damage the equipment; line tripping: severe wind deviation may trigger line tripping, resulting in power outages. Therefore, accurate wind deviation angle data is crucial for timely detecting potential risks and ensuring the safe operation of the distribution line. Usually, a wind deviation angle sensor is used to detect the wind deviation situation of the distribution line.

[0003] The interaction effect and icing effect between distribution lines are important factors affecting the detection accuracy of the wind deviation angle sensor of the distribution line. Factors such as multiple distribution lines, jumpers, and the bending of the distribution line will cause deviations in the measured wind deviation angle value, affecting the change of the wind deviation angle, making it difficult for the wind deviation angle sensor to provide accurate data. When there is snow or ice on the distribution line, the distribution line is unevenly stressed, and the wind deviation angle is more complexly affected by mechanical changes, resulting in deviations in the measured wind deviation angle value. The existence of these factors may cause large deviations in the wind deviation angle values measured by the wind deviation angle sensor. Therefore, the wind deviation angle is a crucial parameter in the wind deviation monitoring and early warning of icing distribution lines. However, the prior art cannot accurately analyze whether the reason for the deviation of the measured wind deviation angle is caused by the bending of the distribution line or the icing on the distribution line, so subsequent wind deviation monitoring and early warning cannot be carried out for the accurate reason. Summary of the Invention

[0004] In order to solve the technical problem that the prior art cannot accurately identify the reason for the deviation of the measured wind deviation angle, the purpose of the present invention is to provide a wind deviation monitoring and early warning system and device for icing distribution lines, and the specific technical solutions adopted are as follows:

[0005] In the first aspect of the present invention, a wind deviation monitoring and early warning system for icing distribution lines is provided, including:

[0006] A data acquisition module for acquiring a sequence of wind deviation angle values of the distribution line;

[0007] A data processing module for determining whether the wind deviation of the distribution line is abnormal based on the correlation between the sequence of wind deviation angle values and the sequence of wind force values;

[0008] If the wind deviation of the distribution line is abnormal, then obtain the bending degree of the distribution line according to the line shape of the distribution line;

[0009] Detect the severe shaking condition of the distribution line to obtain the degree of ice and frost influence on the distribution line;

[0010] Determine the abnormal cause of the wind deflection angle value of the distribution line according to the magnitude relationship between the bending degree and the degree of ice and frost influence.

[0011] In an exemplary embodiment, the determination process of whether the wind deflection of the distribution line is abnormal includes:

[0012] Segment the wind deflection angle value sequence and the wind force value sequence in the same segmentation manner to obtain a plurality of wind deflection angle value sequence segments and a plurality of wind force value sequence segments respectively;

[0013] Based on the similarity between the wind deflection angle value sequence segment and the wind force value sequence segment at the same position, obtain a similarity sequence;

[0014] If the similarity sequence does not show an increasing trend, it is determined that the wind deflection of the distribution line is abnormal.

[0015] In an exemplary embodiment, the obtaining process of the bending degree includes:

[0016] Obtain a plurality of consecutive video frame images of the distribution line;

[0017] Extract the connected regions of the distribution line in the video frame images to obtain the skeleton of the distribution line;

[0018] Take the two-dimensional coordinates of all pixel points of the skeleton as the input of the principal component analysis method to obtain the two largest projection values and the corresponding projection vectors;

[0019] According to the projection value difference between the two largest projection values and the angle difference between the corresponding projection vectors, obtain the bending degree of the distribution line.

[0020] In an exemplary embodiment, the obtaining of the bending degree according to the projection value difference between the two largest projection values and the angle difference between the corresponding projection vectors includes:

[0021] According to the magnitude relationship between the projection value difference and the angle difference, obtain the sub-bending degree of the distribution line in each video frame image;

[0022] According to the sub-bending degree corresponding to each video frame image, obtain the bending degree of the distribution line.

[0023] In an exemplary embodiment, the obtaining process of the projection value difference includes:

[0024] Obtain the ratio of the second projection value to the first projection value as the projection value difference; the first projection value and the second projection value are the two largest projection values, and the first projection value is greater than the second projection value;

[0025] The process of obtaining the angle difference includes:

[0026] Obtain the angles corresponding to the projection vectors of the two largest projection values, and calculate the absolute value of the angle difference;

[0027] Take the ratio of the absolute value of the angle difference to the first angle as the angle difference; the first angle is the largest angle among the angles of the two largest projection values.

[0028] In an exemplary embodiment, the process of obtaining the degree of ice and frost influence includes:

[0029] Obtain the intersection-over-union ratio of the power distribution line connectivity regions in each adjacent two video frame images to form an intersection-over-union ratio sequence;

[0030] Obtain the degree of drastic change of the intersection-over-union ratio according to the intersection-over-union ratio sequence;

[0031] Obtain the degree of ice and frost influence on the power distribution line according to the degree of drastic change of the intersection-over-union ratio.

[0032] In an exemplary embodiment, the process of obtaining the degree of drastic change of the intersection-over-union ratio includes:

[0033] Obtain the two-dimensional coordinate points of each intersection-over-union ratio in the intersection-over-union ratio sequence; the abscissa of the two-dimensional coordinate point of the intersection-over-union ratio is the serial number of the intersection-over-union ratio in the intersection-over-union ratio sequence, and the ordinate of the two-dimensional coordinate point of the intersection-over-union ratio is the value of the intersection-over-union ratio;

[0034] Cluster the two-dimensional coordinate points of each intersection-over-union ratio, and obtain the abscissa sequence of each cluster. The abscissa sequence is formed by sorting the number of intersection-over-union ratios within the same abscissa range in each cluster in the order of the abscissa range;

[0035] Obtain the difference sequence of the abscissa sequence of the cluster, and obtain the variance of the difference sequence;

[0036] Obtain the maximum variance among the variances of each cluster as the degree of drastic change of the intersection-over-union ratio.

[0037] In an exemplary embodiment, the process of determining the abnormal cause of the wind deviation angle value includes:

[0038] If the degree of bending is greater than or equal to the degree of ice frost influence, it is determined that the abnormal wind deviation angle value of the power distribution line is caused by the bending of the power distribution line; if the degree of bending is less than the degree of ice frost influence, it is determined that the abnormal wind deviation angle value of the power distribution line is caused by icing on the power distribution line.

[0039] In an exemplary embodiment, the wind deviation monitoring and warning system further includes:

[0040] A warning module, configured to, if the degree of bending is greater than or equal to the degree of ice frost influence, correct the current wind deviation angle according to the degree of bending; compare the corrected current wind deviation angle with a preset wind deviation angle warning threshold, and if the corrected current wind deviation angle is greater than or equal to the preset wind deviation angle warning threshold, output an alarm signal;

[0041] If the degree of bending is less than the degree of ice frost influence, compare the degree of ice frost influence with a preset ice frost influence degree threshold, and if the degree of ice frost influence is greater than or equal to the preset ice frost influence degree threshold, output an alarm signal.

[0042] In a second aspect of the present invention, there is provided a wind deviation monitoring and warning device for an ice-covered power distribution line, including: a memory and a processor; the memory is connected to the processor; the memory is used to store program instructions; the processor is used to, when the program instructions are executed, implement the steps of a wind deviation monitoring and warning method for an ice-covered power distribution line as follows:

[0043] Obtain a sequence of wind deviation angle values of the power distribution line;

[0044] Based on the correlation between the wind deviation angle value sequence and the wind force value sequence, determine whether the wind deviation of the power distribution line is abnormal;

[0045] If the wind deviation of the power distribution line is abnormal, obtain the degree of bending of the power distribution line according to the line shape of the power distribution line;

[0046] Detect the severity of the shaking of the power distribution line to obtain the degree of ice frost influence on the power distribution line;

[0047] According to the magnitude relationship between the degree of bending and the degree of ice frost influence, determine the cause of the abnormal wind deviation angle value of the power distribution line.

[0048] The present invention has the following beneficial effects: The present invention first determines whether the wind deviation of the power distribution line is abnormal based on the correlation between the wind deviation angle value sequence and the wind force value sequence of the power distribution line. If the wind deviation of the power distribution line is abnormal, the degree of bending of the power distribution line is obtained according to the line shape of the power distribution line; and the severity of the shaking of the power distribution line is detected to obtain the degree of ice frost influence on the power distribution line. Finally, in combination with the magnitude relationship between the degree of bending and the degree of ice frost influence, the accurate cause of the abnormal wind deviation angle value of the power distribution line is determined. Description of the Drawings

[0049] Figure 1 is a schematic diagram of the hardware structure composition of a wind deviation monitoring and early warning system for ice-covered distribution lines provided by an embodiment of the present invention;

[0050] Figure 2 is a schematic diagram of the software module composition of a wind deviation monitoring and early warning system for ice-covered distribution lines provided by an embodiment of the present invention;

[0051] Figure 3 is a flowchart of a wind deviation monitoring and early warning method for ice-covered distribution lines provided by an embodiment of the present invention;

[0052] Figure 4 is a decision flowchart for determining whether the wind deviation of a distribution line is abnormal provided by an embodiment of the present invention;

[0053] Figure 5 is a flowchart for obtaining the bending degree of a distribution line provided by an embodiment of the present invention;

[0054] Figure 6 is a schematic diagram of the relationship between the skeleton of a distribution line and the main direction provided by an embodiment of the present invention;

[0055] Figure 7 is a flowchart for obtaining the degree of ice and frost influence on a distribution line provided by an embodiment of the present invention;

[0056] Figure 8 is a flowchart for obtaining the degree of drastic change in the intersection-over-union ratio provided by an embodiment of the present invention;

[0057] Figure 9 is a schematic diagram of a histogram provided by an embodiment of the present invention;

[0058] Figure 6 In, the solid line represents the skeleton of the distribution line, the dashed line represents the main direction obtained by the principal component analysis method of the distribution line skeleton, and the solid circular dots represent the inflection points on the distribution line;

[0059] Figure 9 In, x is the horizontal axis, representing each abscissa range; y is the vertical axis, representing the number of intersection-over-union ratios in each abscissa range in the density cluster, with the unit of individual. Detailed Implementation Manner

[0060] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the specific implementation manners, structures, features and their effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The data information collected in this application has been obtained with full consent and authorization, and the collection, use and processing of relevant information need to comply with the relevant laws, regulations and standards of relevant countries and regions.

[0062] This embodiment provides a wind deviation monitoring and warning system for icing distribution lines, and the monitoring object is a section of distribution line. The length of the distribution line is determined according to the actual situation.

[0063] This embodiment provides a data detection hardware part corresponding to a wind deviation monitoring and warning system for icing distribution lines, including: a wind deviation angle sensor, a wind speed sensor, and an image sensor. Among them, the wind deviation angle sensor is used to detect the wind deviation angle value of the distribution line, the wind speed sensor is used to detect the wind speed value at the location of the distribution line, and the image sensor is used to obtain the video frame image of the distribution line, so as to obtain the image information of the distribution line.

[0064] On the basis of ensuring that each sensor can effectively detect data, the specific installation positions of the wind deviation angle sensor, the wind speed sensor, and the image sensor are determined according to the actual situation.

[0065] In an exemplary embodiment, a specific installation method of the wind deviation angle sensor, the wind speed sensor, and the image sensor is given as follows.

[0066] For the wind speed sensor: It can be installed at both ends of the distribution line, especially in open areas along the line or areas with strong wind, and the wind speed sensor is fixed on the electric tower of the distribution line to avoid interference from other obstacles such as buildings and trees.

[0067] For the wind deviation angle sensor: It can be fixed near the conductor of the distribution line, so as to directly measure the wind condition of the distribution line, and the data is closer to the actual situation, or installed at the connection point between the electric tower, the distribution line and the tower, and the jumper. Therefore, the installation position of the wind deviation angle sensor needs to ensure that the influence of the wind on the entire distribution line can be completely captured, and the wind deviation angle can be recorded in real time when the wind speed is relatively high.

[0068] For the image sensor: It can be installed at the turning points, suspension points of the power distribution line or places vulnerable to external environment, and specifically fixed on the corresponding utility poles, brackets or towers, so as to monitor the bending areas of the power distribution line or the places where bending may occur. Moreover, a small and portable image sensor can be used, and the image sensor can be a high-definition monitoring camera, set facing the power distribution line, capable of obtaining high-definition images of the power distribution line. Further, the image sensor can also be equipped with a wind direction sensor (such as an anemometer) to ensure that the system can automatically adjust the monitoring angle of the image sensor according to the wind force.

[0069] Therefore, each sensor provided in this embodiment can be fixed on the corresponding tower or support structure through a bracket to ensure the stability of the sensor. The bracket can also have an adjustment function to fine-tune the direction of the sensor as needed. Further, the sensor can also be equipped with an automatic adjustment device (such as an electric pan-tilt head), which automatically adjusts the detection angle of the sensor according to the wind direction. It should be understood that on the basis of being able to reliably obtain the corresponding signals of the power distribution line, this embodiment is not limited to the specific installation positions and installation methods of each sensor.

[0070] The hardware part of a wind deviation monitoring and warning system for iced power distribution lines provided in this embodiment can include a data processing device, such as a server or a computer mainframe, etc., taking the server as an example. As Figure 1 shown, the server is signal-connected to each sensor, which can be connected by wire or wirelessly.

[0071] Each sensor collects data according to a preset collection frequency, and the preset collection frequency is set according to the actual situation. As an example, this embodiment takes once per second. The server receives the data information detected by each sensor and processes the data.

[0072] Software modules are set in the server, including a data acquisition module and a data processing module, and each software module realizes the functions of the corresponding module through its respective corresponding method steps. Therefore, a wind deviation monitoring and warning system for iced power distribution lines provided in this embodiment is configured in the server, including a data acquisition module and a data processing module, as Figure 2 shown. Both the data acquisition module and the data processing module can be software modules, and the functions of the corresponding modules are realized through the corresponding method steps.

[0073] As Figure 3As shown, they are the method steps corresponding to each module. The data acquisition module is used to acquire the sequence of wind deflection angle values of the distribution line. The data processing module is used to determine whether the wind deflection of the distribution line is abnormal based on the correlation between the sequence of wind deflection angle values and the sequence of wind force values; if the wind deflection of the distribution line is abnormal, then according to the line shape of the distribution line, the degree of bending of the distribution line is obtained; the severe shaking condition of the distribution line is detected, thereby obtaining the degree of ice and frost influence on the distribution line; according to the magnitude relationship between the degree of bending and the degree of ice and frost influence, the reason for the abnormal wind deflection angle value of the distribution line is determined.

[0074] Therefore, the implementation processes of the data acquisition module and the data processing module are essentially a method for wind deflection monitoring and early warning for ice-covered distribution lines. The following specifically describes each method step in conjunction with the accompanying drawings.

[0075] Step 1: Acquire the sequence of wind deflection angle values of the distribution line.

[0076] The wind deflection angle sensor acquires the wind deflection angle values of the distribution line at a preset acquisition frequency, the wind speed sensor acquires the wind speed value at the location of the distribution line, that is, the wind force value, at a preset acquisition frequency; the image sensor acquires the video frame images of the distribution line at a preset acquisition frequency. The preset acquisition frequencies of these three sensors are equal, and the three acquire synchronously.

[0077] In this embodiment, a monitoring time period is preset, and the duration of this monitoring time period is set according to actual needs. During this monitoring time period, multiple wind deflection angle values, wind force values, and video frame images can be acquired.

[0078] Sort the multiple wind deflection angle values within this monitoring time period in chronological order to obtain the sequence of wind deflection angle values. Similarly, sort the multiple wind force values within this monitoring time period in chronological order to obtain the sequence of wind force values.

[0079] Step 2: Determine whether the wind deflection of the distribution line is abnormal based on the correlation between the sequence of wind deflection angle values and the sequence of wind force values.

[0080] Since the subsequent steps are processed on the basis of the abnormal wind deflection of the distribution line, therefore, it is first necessary to determine whether the wind deflection of the distribution line is abnormal. Through the relationship between the change in wind force and the change in wind deflection angle, under normal circumstances, if the wind force increases, the line is blown, and then the correlation between the wind deflection angle and the wind force will become larger. If the wind deflection angle data does not change with the change in wind force after it increases, it is most likely caused by the bending or ice and frost coverage of the distribution line. Therefore, determine whether the wind deflection of the distribution line is abnormal based on the correlation between the sequence of wind deflection angle values and the sequence of wind force values.

[0081] In an exemplary embodiment, as Figure 4 shown, the determination process of whether the wind deflection of the distribution line is abnormal includes:

[0082] Step 2-1: Segment the wind deflection angle value sequence and the wind force value sequence in the same segmentation manner to obtain multiple wind deflection angle value sequence segments and multiple wind force value sequence segments respectively.

[0083] Segment the wind deflection angle value sequence and the wind force value sequence in the same segmentation manner. In an exemplary embodiment, the segmentation manner can be segmentation according to the number of data, that is, in chronological order, every preset number of data constitutes a segment, so as to be segmented into multiple segments. Specifically: Segment the wind deflection angle value sequence in the manner that every preset number of data constitutes a segment to obtain multiple wind deflection angle value sequence segments; Segment the wind force value sequence in the manner that every preset number of data constitutes a segment to obtain multiple wind force value sequence segments. The number of data included in the wind deflection angle value sequence segments and the wind force value sequence segments is the same. Taking 10 as an example for the preset number, then, in order, every 10 wind deflection angle values of the wind deflection angle value sequence form a segment to obtain multiple wind deflection angle value sequence segments; every 10 wind force values of the wind force value sequence form a segment to obtain multiple wind force value sequence segments.

[0084] Step 2-2: Obtain a similarity sequence based on the similarity between the wind deflection angle value sequence segment and the wind force value sequence segment at the same position.

[0085] Through the segmentation in Step 2-1, there is a corresponding relationship in position between each wind deflection angle value sequence segment and each wind force value sequence segment, that is: the wind deflection angle value sequence segment at the first position and the wind force value sequence segment at the first position are a pair of sequence segments with a corresponding relationship, the wind deflection angle value sequence segment at the second position and the wind force value sequence segment at the second position are a pair of sequence segments with a corresponding relationship, and so on.

[0086] Obtain the similarity between the wind deflection angle value sequence segment and the wind force value sequence segment at the same position. Specifically: Obtain the similarity between the wind deflection angle value sequence segment at the first position and the wind force value sequence segment at the first position to obtain the first similarity; Obtain the similarity between the wind deflection angle value sequence segment at the second position and the wind force value sequence segment at the second position to obtain the second similarity, and so on, until obtaining the similarity between the wind deflection angle value sequence segment at the last position and the wind force value sequence segment at the last position to obtain the last similarity.

[0087] Among them, the specific manner of similarity is set according to actual needs, such as: cosine similarity, Pearson correlation coefficient, etc. This embodiment takes cosine similarity as an example.

[0088] Construct a similarity sequence, and the similarity sequence includes the first similarity, the second similarity,..., the last similarity.

[0089] Step 2-3: If the similarity sequence does not show an increasing trend, it is determined that the wind deflection of the power distribution line is abnormal.

[0090] As can be seen from the above analysis, when the wind force increases, the line is blown, and the correlation between the wind deflection angle and the wind force will become larger. Therefore, if the similarity sequence shows an increasing trend, that is, the similarity gradually increases, it is determined that the wind deflection of the power distribution line is normal; if the similarity sequence does not show an increasing trend, it is determined that the wind deflection of the power distribution line is abnormal, and the wind deflection value is inaccurate.

[0091] In an exemplary embodiment, the detection process of whether the similarity sequence is increasing is given as follows: taking the order value of each element in the similarity sequence as the abscissa and the similarity value as the ordinate, obtaining multiple coordinate points, taking all the coordinate points as the input of the principal component analysis PCA algorithm, obtaining multiple two-dimensional projection vectors, and the projection value corresponding to each projection vector, recording the projection vector corresponding to the maximum projection value as the direction vector, obtaining the angle corresponding to the direction vector, for example: obtaining the arctangent value of the ratio of the ordinate to the abscissa of the direction vector, denoted as s. If s is greater than 0, the similarity sequence shows an increasing trend, that is, the wind deflection angle value is normal and there is no data distortion due to external influence. In this case, it can be directly involved in the final early warning. If s is less than or equal to 0, it is determined that the wind deflection of the power distribution line is abnormal. As another implementation, it is also possible to perform a linear fitting based on the coordinate points of each element in the similarity sequence, and then obtain the slope of the fitted line. If the slope is greater than 0, the similarity sequence shows an increasing trend. If the slope is less than or equal to 0, it is determined that the wind deflection of the power distribution line is abnormal.

[0092] Step 3: If the wind deflection of the power distribution line is abnormal, obtain the bending degree of the power distribution line according to the line shape of the power distribution line.

[0093] If the wind deflection of the power distribution line is abnormal, it is necessary to analyze whether it is due to the increase in the weight of the power distribution line caused by the accumulation of more ice and frost or due to the bending of the power distribution line. First, based on the determination that the wind deflection of the power distribution line is abnormal, this step obtains the bending degree of the power distribution line according to the line shape of the power distribution line.

[0094] In an exemplary embodiment, as Figure 5 shown, a specific implementation process of obtaining the bending degree of the power distribution line according to the line shape of the power distribution line is given:

[0095] Step 3-1: Obtain multiple consecutive video frame images of the power distribution line.

[0096] Obtain multiple consecutive video frame images of the power distribution line through an image sensor. Subsequently, it is used to detect the bending degree of the power distribution line in the video frame images according to the video frame images.

[0097] Step 3-2: Extract the connected components of the power distribution line in the video frame image to obtain the skeleton of the power distribution line.

[0098] The following explains for any arbitrary video frame image. Through image segmentation, obtain the power distribution line area in the video frame image. For example, since there is a large difference between the gray value of the power distribution line and that of other parts of the background, the background and foreground can be segmented by the gray value difference, thereby obtaining the power distribution line area. Then, obtain the connected components of the power distribution line by the method of connected component extraction for the power distribution line area. Finally, perform skeleton extraction on the connected components of the power distribution line to obtain the skeleton of the power distribution line.

[0099] Step 3-3: Take the two-dimensional coordinates of all pixel points of the skeleton as the input of the principal component analysis method to obtain the two largest projection values and the corresponding projection vectors.

[0100] Construct an image two-dimensional coordinate system with the length and width of the video frame image, map all pixel points of the skeleton into the image two-dimensional coordinate system, and obtain the two-dimensional coordinates of all pixel points of the skeleton in the image two-dimensional coordinate system. Take the two-dimensional coordinates of all pixel points of the skeleton as the input of the principal component analysis PCA method to obtain multiple projection values and the corresponding projection vectors. Sort all the projection values from largest to smallest. The largest projection value represents the most important feature of the skeleton. Therefore, obtain the two largest projection values, that is, the largest projection value and the second largest projection value second only to the largest projection value.

[0101] For the convenience of explanation, set the two largest projection values as the first projection value and the second projection value respectively, and the first projection value is greater than the second projection value, that is: define the largest projection value as the first projection value, and define the second largest projection value second only to the largest projection value as the second projection value. And obtain the projection vector corresponding to the first projection value and the projection vector corresponding to the second projection value. Define the projection vector corresponding to the first projection value as the first projection vector, and define the projection vector corresponding to the second projection value as the second projection vector.

[0102] The larger the first projection value is relative to the second projection value, that is, the greater the gap between the second projection value and the first projection value, the better the largest projection value represents the shape of the skeleton, and the closer the shape of the skeleton is to a linear shape, that is, the smaller the bending degree of the power distribution line.

[0103] Step 3-4: Obtain the bending degree of the power distribution line according to the projection value difference between the two largest projection values and the angle difference between the corresponding projection vectors.

[0104] Obtain the projection value difference between the two largest projection values. In an exemplary embodiment, obtain the absolute value of the difference between the first projection value and the second projection value, and then calculate the ratio of the absolute value of the difference to the first projection value. This ratio represents the representativeness of the largest projection value. The greater the representativeness, the closer the shape of the skeleton is to a linear shape, that is, the smaller the bending degree of the power distribution line. Then, calculate the difference between the value 1 and this ratio, and use the obtained difference as the projection value difference, which represents the bending degree of the power distribution line. Correspondingly, as a simpler calculation method, directly calculate the ratio of the second projection value to the first projection value as the projection value difference.

[0105] The reason for also needing to combine the angle difference between the projection vectors corresponding to the two largest projection values to obtain the bending degree of the power distribution line is that if the bending degree of the power distribution line is represented by the morphological characteristics of only the two largest projection values, some inflection points on the power distribution line will be ignored. Sometimes, the power distribution line may cause the wind deflection angle sensor to be inaccurate due to the inflection points. As Figure 6 shown, the solid line represents the skeleton of the power distribution line, the dashed line represents the main direction obtained by the principal component analysis method for the skeleton of the power distribution line, and the three solid circular points represent the inflection points on the power distribution line. It can be seen that the main direction cannot represent the inflection points well.

[0106] For Figure 6 the problem shown in, this embodiment represents it through the angle difference between the projection vectors corresponding to the two largest projection values. If the angle difference between these two projection vectors is smaller, it means that the number of bending points (i.e., Figure 6 the solid points in) is less, or the bending degree of the bending points is smaller or the bending transition is more gentle, then the bending degree of the power distribution line is smaller. Correspondingly, if the angle difference between these two projection vectors is larger, it means that the bending degree of the power distribution line is larger.

[0107] Therefore, obtain the angle values of the first projection vector and the second projection vector respectively. In an exemplary embodiment, for the first projection vector, take the arctangent value of the ratio of the ordinate to the abscissa in the first projection vector as the angle value of the first projection vector. For the second projection vector, obtain the angle value of the second projection vector through the same method.

[0108] Then, obtain the angle difference between the angle value of the first projection vector and the angle value of the second projection vector. In an exemplary embodiment, calculate the absolute value of the angle difference between the angle value of the first projection vector and the angle value of the second projection vector; then obtain the maximum angle value among the angle value of the first projection vector and the angle value of the second projection vector, and define this maximum angle value as the first angle. Then, calculate the ratio of the absolute value of the angle difference to the first angle as the angle difference. The larger the angle difference, the fewer the bending points on the power distribution line, and the higher the representativeness of the largest projection value for the bending degree of the power distribution line.

[0109] Next, according to the magnitude relationship between the projection value difference and the angle difference, the sub-bending degree of the power distribution line in the video frame image is obtained. In this embodiment, the larger value of the projection value difference and the angle difference is obtained, and the larger value is used as the sub-bending degree of the power distribution line in the video frame image. For example, if the projection value difference is greater than the angle difference, the projection value difference is used as the sub-bending degree of the power distribution line in the video frame image.

[0110] By using the above process, the sub-bending degree of the power distribution line in each video frame image is obtained. Finally, according to the sub-bending degree corresponding to each video frame image, the bending degree of the power distribution line is obtained. In an exemplary embodiment, the maximum value among the sub-bending degrees corresponding to each video frame image is used as the bending degree of the power distribution line.

[0111] The obtained bending degree of the power distribution line can also represent the probability that the wind deflection angle value is inaccurate due to the bending of the power distribution line. The greater the bending degree of the power distribution line, the more likely it is that the inaccurate wind deflection angle value is caused by the bending of the power distribution line.

[0112] Step 4: Detect the severity of the shaking of the power distribution line to obtain the ice and frost influence degree of the power distribution line.

[0113] Due to wind deflection anomalies, it may be caused by the bending of the power distribution line or by the ice and frost covering the power distribution line. Therefore, it is also necessary to obtain the ice and frost influence degree of the power distribution line, which characterizes the influence degree of the ice and frost covering the power distribution line on the power distribution line.

[0114] If the wind force increases, but in the video frame image, the change in the bending degree of the power distribution line is small, it indicates that it is more likely that the large deviation of the wind deflection angle value is caused by the ice and frost covering the power distribution line. The increase in wind force is the premise of the abnormal calculation. Therefore, the ice and frost influence degree of the power distribution line can be obtained only according to the change in the deformation of the power distribution line, and the ice and frost influence degree characterizes the probability of the deviation of the wind deflection angle value caused by the ice and frost.

[0115] Under the action of the wind force, the ice and frost covering the power distribution line will cause the power distribution line to shake violently. Therefore, detect the severity of the shaking of the power distribution line to obtain the ice and frost influence degree of the power distribution line.

[0116] In an exemplary embodiment, as Figure 7 shown, the following gives a specific obtaining process of the ice and frost influence degree:

[0117] Step 4-1: Obtain the intersection-over-union ratio of the connected regions of the power distribution line in each adjacent two video frame images to form an intersection-over-union ratio sequence.

[0118] For multiple consecutive video frame images, based on the power distribution line connected regions in each video frame image obtained above, obtain the intersection and union of the power distribution line connected regions in every two adjacent video frame images, and calculate the ratio of the intersection to the union to obtain the intersection over union ratio.

[0119] The larger the intersection over union ratio, the larger the intersecting part of the power distribution line connected regions in two adjacent video frame images, that is, the larger the overlapping region, indicating a lower degree of shaking of the power distribution line. This means that the ice-covered power distribution line is less affected by wind. In this case, if there is a large change in the bending degree of the ice-covered power distribution line in consecutive video frame images, it is very likely that the wind deflection angle value is inaccurate due to the bending change of the power distribution line, rather than due to ice and frost.

[0120] Since one intersection over union ratio is obtained for every two adjacent video frame images, combining multiple consecutive video frame images, multiple intersection over union ratios are obtained. Combining these intersection over union ratios with the order of the video frame images forms an intersection over union ratio sequence.

[0121] Step 4-2: Obtain the degree of drastic change of the intersection over union ratio according to the intersection over union ratio sequence.

[0122] If the intersection over union ratio is constantly fluctuating, it indicates that the ice-covered power distribution line is affected by wind and is in an unstable state. It is more likely that the ice-covered power distribution line is adhered to the wire support, such as a utility pole, due to the influence of ice and frost.

[0123] Therefore, obtain the degree of drastic change of the intersection over union ratio according to the intersection over union ratio sequence. In an exemplary embodiment, as Figure 8 shown, the following gives a specific obtaining process of the degree of drastic change of the intersection over union ratio:

[0124] Step 4-2-1: Obtain the two-dimensional coordinate points of each intersection over union ratio in the intersection over union ratio sequence.

[0125] Taking the serial number of the intersection over union ratio in the intersection over union ratio sequence as the abscissa and the value of the intersection over union ratio as the ordinate, construct an intersection over union ratio two-dimensional coordinate system. According to the abscissa and coordinate value of the intersection over union ratio two-dimensional coordinate system, determine the two-dimensional coordinate points of each intersection over union ratio. The abscissa of the two-dimensional coordinate point is the serial number of the intersection over union ratio in the intersection over union ratio sequence, and the ordinate of the two-dimensional coordinate point is the value of the intersection over union ratio.

[0126] Step 4-2-2: Cluster the two-dimensional coordinate points of each intersection over union ratio to obtain the abscissa sequence of each cluster.

[0127] Cluster the two-dimensional coordinate points of all intersection over union ratios. In an exemplary embodiment, a density clustering method (such as the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering method) is used for clustering to obtain multiple density clusters. If the distribution of the two-dimensional coordinate points of the intersection over union ratios shows an unstable state, many intersection over union ratios with a large difference in abscissa but similar values will be grouped together. Furthermore, according to the missing situation of the abscissa of the intersection over union ratios in the density cluster, the instability of the change of the intersection over union ratio data can be obtained.

[0128] Construct multiple abscissa ranges, and the size of the abscissa range is set according to actual needs. For example, 10 abscissas are included in one abscissa range.

[0129] For any density cluster, obtain the number of intersection over union ratios in the same abscissa range in this density cluster, so as to obtain the number of intersection over union ratios in each abscissa range. It should be understood that for some abscissa ranges where there are no intersection over union ratios, the corresponding number of intersection over union ratios is 0. Sort the number of intersection over union ratios in each abscissa range in this density cluster according to the order of each abscissa range to obtain an abscissa sequence.

[0130] In an exemplary embodiment, histograms can be statistically calculated for each abscissa range and the number of intersection over union ratios in each abscissa range in this density cluster to obtain the histogram corresponding to this density cluster. Among them, the abscissa of this histogram is each abscissa range, and the ordinate is the number of intersection over union ratios in each abscissa range in this density cluster. As Figure 9 shown, x represents the horizontal axis of the histogram, y represents the vertical axis of the histogram, and the unit of the vertical axis is individual. If in the histogram, the greater the deviation of the height of adjacent bars, that is, the greater the deviation of the number of intersection over union ratios in adjacent abscissa ranges, the greater the instability of the change of the intersection over union ratios in this density cluster.

[0131] Step 4-2-3: Obtain the difference sequence of the abscissa sequence of the cluster and calculate the variance of the difference sequence.

[0132] Obtain the absolute value of the difference in the height of adjacent bars in this density cluster, that is, the absolute value of the difference in the number of intersection over union ratios in adjacent abscissa ranges in the abscissa sequence, and construct a difference sequence according to the absolute value of the difference in each number of intersection over union ratios. Then calculate the variance of this difference sequence. The greater the variance, the higher the degree of severity of the change of the intersection over union ratio.

[0133] Step 4-2-4: Obtain the maximum variance among the variances of each cluster as the degree of severity of the change of the intersection over union ratio.

[0134] In the above manner, the variances corresponding to each density cluster are obtained, and the maximum variance is determined from the variances corresponding to each density cluster. This maximum variance serves as the final degree of drastic change in the intersection over union. The greater the degree of drastic change in the intersection over union, the greater the instability of the change in the intersection over union, that is, the more unstable the state of the ice-covered distribution line affected by the wind.

[0135] Step 4-3: Obtain the degree of ice and frost influence on the distribution line according to the degree of drastic change in the intersection over union.

[0136] In an exemplary embodiment, the obtained degree of drastic change in the intersection over union is normalized, and the normalized degree of drastic change in the intersection over union is used as the degree of ice and frost influence on the distribution line. Among them, the normalization method can be: , represents the processing object, and exp represents the exponential function with the natural constant e as the base.

[0137] Step 5: Determine the abnormal cause of the wind deviation angle value of the distribution line according to the magnitude relationship between the degree of bending and the degree of ice and frost influence.

[0138] Through the above steps, the degree of bending and the degree of ice and frost influence of the distribution line are obtained respectively. The degree of bending of the distribution line represents the probability that the wind deviation is caused by the bending of the distribution line, and the degree of ice and frost influence represents the probability that the wind deviation is caused by the cumulative ice covering the distribution line.

[0139] Then, according to the magnitude relationship between the degree of bending and the degree of ice and frost influence, the abnormal cause of the wind deviation angle value of the distribution line is determined. Specifically: if the degree of bending is greater than or equal to the degree of ice and frost influence, it is determined that the abnormal wind deviation angle value of the distribution line is caused by the bending of the distribution line; if the degree of bending is less than the degree of ice and frost influence, it is determined that the abnormal wind deviation angle value of the distribution line is caused by the icing of the distribution line.

[0140] In an exemplary embodiment, the wind deviation monitoring and early warning system further includes an early warning module, and the early warning module is used to execute the following early warning strategy:

[0141] If the degree of bending is greater than or equal to the degree of ice and frost influence, that is, if the abnormal wind deviation angle value of the distribution line is caused by the bending of the distribution line, the current wind deviation angle is corrected according to the obtained degree of bending. Specifically: the corrected wind deviation angle = the current wind deviation angle * (1 - w0), where w0 represents the degree of bending of the distribution line. Then, compare the corrected current wind deviation angle with the preset wind deviation angle early warning threshold. If the corrected current wind deviation angle is greater than or equal to the preset wind deviation angle early warning threshold, an alarm signal is output to facilitate the staff to take relevant measures in time. Among them, the specific value of the preset wind deviation angle early warning threshold is set by the desired early warning sensitivity. For example: if a higher early warning sensitivity is desired, the preset wind deviation angle early warning threshold can be set smaller so that it is easier to reach the early warning condition.

[0142] Among them, the relevant measures taken by the staff may include: The staff bring new wires: If the bending causes the distribution line to no longer meet the original tension requirements, it may be necessary to replace part of the distribution line to ensure that the distribution line does not bend excessively due to wind or other factors; Use straightening tools: Straighten the distribution line through professional tools (such as tension adjusters or straightening instruments) to restore the normal tension and straightness of the distribution line; Inspect and reinforce the distribution line: Inspect the distribution line to ensure that there is no further damage caused by other factors, and take reinforcement measures to avoid similar problems in the future.

[0143] If the degree of bending is less than the degree of ice and frost influence, that is, if the abnormal wind deflection angle value of the distribution line is caused by icing on the distribution line, then compare the degree of ice and frost influence of the distribution line with the preset ice and frost influence degree threshold. If the degree of ice and frost influence is greater than or equal to the preset ice and frost influence degree threshold, an alarm signal is output to facilitate the staff to take relevant measures in time. Among them, the specific value of the preset ice and frost influence degree threshold is set by the desired warning sensitivity. For example: If a higher warning sensitivity is desired, the preset ice and frost influence degree threshold can be set smaller so that it is easier to reach the warning condition.

[0144] Among them, the relevant measures taken by the staff may include: Heating the distribution line: Electric heating tapes or heating equipment can be used to prevent the re-accumulation of ice and frost; Strengthening line protection: For distribution lines that are often affected by ice and frost, anti-freezing equipment should be considered for installation and special cold-resistant conductors should be used to reduce the influence of ice and frost on the distribution line.

[0145] This embodiment also provides a wind deflection monitoring and warning device for iced distribution lines, including: a memory and a processor; the memory is connected to the processor; the memory is used to store program instructions; the processor is used to, when the program instructions are executed, implement Figure 3 the steps of a wind deflection monitoring and warning method for iced distribution lines as shown. Since the steps of the wind deflection monitoring and warning method for iced distribution lines have been specifically described above, they will not be elaborated here.

[0146] It should be noted that: The above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0147] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. An icing-orientated wind deviation monitoring and early warning system for distribution lines, characterized in that, Including: A data acquisition module, configured to acquire a sequence of wind deflection angle values of a distribution line; A data processing module, configured to determine whether the wind deflection of the distribution line is abnormal based on the correlation between the sequence of wind deflection angle values and the sequence of wind force values; If the wind deflection of the distribution line is abnormal, then obtain the degree of bending of the distribution line according to the line shape of the distribution line; Detect the severity of the shaking of the distribution line, thereby obtaining the degree of ice and frost influence on the distribution line; Determine the cause of the abnormal wind deflection angle value of the distribution line according to the magnitude relationship between the degree of bending and the degree of ice and frost influence; The process of obtaining the degree of ice and frost influence includes: Acquire a plurality of consecutive video frame images of the distribution line; Obtain the intersection-over-union ratio of the connected regions of the distribution line in every two adjacent video frame images, and form a sequence of intersection-over-union ratios; Obtain the severity of the change in the intersection-over-union ratio according to the sequence of intersection-over-union ratios; Obtain the degree of ice and frost influence on the distribution line according to the severity of the change in the intersection-over-union ratio; The process of obtaining the severity of the change in the intersection-over-union ratio includes: Obtain the two-dimensional coordinate points of each intersection-over-union ratio in the sequence of intersection-over-union ratios; the abscissa of the two-dimensional coordinate point of the intersection-over-union ratio is the serial number of the intersection-over-union ratio in the sequence of intersection-over-union ratios, and the ordinate of the two-dimensional coordinate point of the intersection-over-union ratio is the value of the intersection-over-union ratio; Cluster the two-dimensional coordinate points of each intersection-over-union ratio, and obtain the sequence of abscissas of each cluster, where the sequence of abscissas is the number of intersection-over-union ratios within the same abscissa range in the cluster sorted in the order of the abscissa range; Obtain the difference sequence of the sequence of abscissas of the cluster, and obtain the variance of the difference sequence; Obtain the maximum variance among the variances of each cluster as the severity of the change in the intersection-over-union ratio.

2. The wind deviation monitoring and early warning system for ice-covered distribution lines according to claim 1, characterized in that, The process of determining whether the wind deflection of the distribution line is abnormal includes: Segment the sequence of wind deflection angle values and the sequence of wind force values in the same segmentation manner, respectively obtaining a plurality of segments of the sequence of wind deflection angle values and a plurality of segments of the sequence of wind force values; Obtain a sequence of similarities based on the similarity between the segment of the sequence of wind deflection angle values and the segment of the sequence of wind force values at the same position; If the sequence of similarities does not show an increasing trend, then determine that the wind deflection of the distribution line is abnormal.

3. The wind deviation monitoring and early warning system for ice-covered distribution lines according to claim 1, characterized in that, The process of obtaining the degree of bending includes: Extract the connected regions of the distribution line in the video frame image to obtain the skeleton of the distribution line; Use the two-dimensional coordinates of all pixel points of the skeleton as the input of the principal component analysis method to obtain the two largest projection values and the corresponding projection vectors; Obtain the degree of bending of the distribution line according to the projection value difference between the two largest projection values and the angle difference between the corresponding projection vectors.

4. The wind deviation monitoring and early warning system for icing distribution lines according to claim 3, characterized in that, The process of obtaining the degree of bending of the distribution line according to the projection value difference between the two largest projection values and the angle difference between the corresponding projection vectors includes: Obtain the sub-degree of bending of the distribution line in each video frame image according to the magnitude relationship between the projection value difference and the angle difference; Obtain the degree of bending of the distribution line according to the sub-degree of bending corresponding to each video frame image.

5. The wind deviation monitoring and early warning system for ice-covered distribution lines according to claim 3 or 4, characterized in that, The process of obtaining the projection value difference includes: Obtain the ratio of the second projection value to the first projection value as the projection value difference; the first projection value and the second projection value are the two largest projection values, and the first projection value is greater than the second projection value. The process of obtaining the angle difference includes: Obtain the angles corresponding to the projection vectors of the two largest projection values and calculate the absolute value of the angle difference. Take the ratio of the absolute value of the angle difference to the first angle as the angle difference; the first angle is the largest angle among the angles of the two largest projection values.

6. The wind deviation monitoring and early warning system for ice-covered distribution lines according to claim 1, characterized in that The process of determining the reason for the abnormal value of the wind deviation angle includes: If the degree of bending is greater than or equal to the degree of ice frost influence, it is determined that the abnormal value of the wind deviation angle of the distribution line is caused by the bending of the distribution line; if the degree of bending is less than the degree of ice frost influence, it is determined that the abnormal value of the wind deviation angle of the distribution line is caused by the icing of the distribution line.

7. The wind deviation monitoring and early warning system for ice-covered distribution lines according to claim 6, characterized in that, The wind deviation monitoring and warning system further includes: An early warning module, which is used to, if the degree of bending is greater than or equal to the degree of ice frost influence, correct the current wind deviation angle according to the degree of bending; compare the corrected current wind deviation angle with a preset wind deviation angle early warning threshold, and if the corrected current wind deviation angle is greater than or equal to the preset wind deviation angle early warning threshold, output an alarm signal. If the degree of bending is less than the degree of ice frost influence, compare the degree of ice frost influence with a preset ice frost influence degree threshold, and if the degree of ice frost influence is greater than or equal to the preset ice frost influence degree threshold, output an alarm signal.

8. A wind deviation monitoring and early warning device for icing distribution lines, characterized by comprising: A memory and a processor; The memory is connected to the processor; The memory is used to store program instructions; The processor is used to, when the program instructions are executed, implement the steps of a wind deviation monitoring and warning method for an icing distribution line as follows: Obtain the wind deviation angle value sequence of the distribution line; Based on the correlation between the wind deviation angle value sequence and the wind force value sequence, determine whether the wind deviation of the distribution line is abnormal; If the wind deviation of the distribution line is abnormal, obtain the degree of bending of the distribution line according to the line shape of the distribution line; Detect the severity of the shaking of the distribution line to obtain the degree of ice frost influence on the distribution line; According to the magnitude relationship between the degree of bending and the degree of ice frost influence, determine the reason for the abnormal value of the wind deviation angle of the distribution line; The process of obtaining the degree of ice frost influence includes: Obtain multiple consecutive video frame images of the distribution line; Obtain the intersection-over-union ratio of the connected regions of the distribution line in each adjacent two video frame images to form an intersection-over-union ratio sequence; According to the intersection-over-union ratio sequence, obtain the severity of the change in the intersection-over-union ratio; According to the severity of the change in the intersection-over-union ratio, obtain the degree of ice frost influence on the distribution line; The process of obtaining the severity of the change in the intersection-over-union ratio includes: Obtain the two-dimensional coordinate points of each intersection-over-union ratio in the intersection-over-union ratio sequence; the abscissa of the two-dimensional coordinate point of the intersection-over-union ratio is the serial number of the intersection-over-union ratio in the intersection-over-union ratio sequence, and the ordinate of the two-dimensional coordinate point of the intersection-over-union ratio is the value of the intersection-over-union ratio. Cluster the two-dimensional coordinate points of each intersection over union (IoU) to obtain the abscissa sequence of each cluster. The abscissa sequence is formed by sorting the number of IoUs within the same abscissa range in the cluster according to the order of the abscissa range. Obtain the difference sequence of the abscissa sequence of the cluster and calculate the variance of the difference sequence. Obtain the maximum variance among the variances of all clusters as the degree of drastic change in the IoU.

Images