Conductor galloping detection method, device, equipment and storage medium

By performing real-time image processing on the spacer bars on the wire, building an amplitude model and fitting sequence, and automatically monitoring the wire dance, solving the problem of low efficiency in the existing technology, and achieving high-precision and low-cost monitoring effects.

CN118587659BActive Publication Date: 2024-11-19SANLI VIDEO FREQUENCY SCI & TECH SHENZHEN
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411066459.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-11-19
Estimated Expiration
2044-08-05

AI Technical Summary

Technical Problem

The existing wire dance detection methods rely on manual inspection or sensors, which are costly and cannot be monitored in real time, resulting in inefficient monitoring.

Method used

By using the spacer on the wire as a marker, image processing is collected in real time, real-time amplitude model and fitting sequence are constructed, the vibration trend and dance information of the wire are determined, and automated monitoring is achieved.

Benefits of technology

It improves the accuracy and efficiency of wire dance detection, reduces labor costs, avoids the problems of high difficulty in algorithm development and large image processing errors, and realizes accurate and real-time monitoring of wire dance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118587659B_ABST
    Figure CN118587659B_ABST
Patent Text Reader

Abstract

The present invention provides a wire dancing detection method, device, equipment and storage medium. The method uses a spacer on a wire as a marker for detecting the dancing of the wire, calculates the center position of the marker in each frame image collected in real time based on the reference position of the marker, and the distance between the center position and the reference position is used to reflect the dancing of the wire. The distance of the center position of the marker relative to the reference position in each frame image and the frame number are formed into real data, and a real sequence is constructed. The real sequence is then used to construct the latest real-time amplitude model in real time, and then the real-time amplitude model is used to construct a fitting sequence based on the target frame number of the real data of the corresponding time period, and the fitting sequence is used as a basis to obtain the dancing information of the wire. This greatly improves the accuracy and efficiency of wire dancing detection.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of image intelligent analysis technology, and in particular to a wire galloping detection method, device, equipment and storage medium. Background Art

[0002] The conductor galloping of transmission lines has long been an important factor affecting the safety of power transmission lines. It can cause transmission lines to trip, some bolts on tower crossarms to fall off, and damage to tower crossarms, jumpers, insulators and hardware, making the power grid operation situation extremely severe. Therefore, it is necessary to detect the conductor galloping.

[0003] However, current methods for detecting conductor vibration usually rely on manual inspections or installing sensors on the conductors to monitor the vibration of the conductors. These methods have problems such as high cost, reliance on human resources and inability to monitor in real time, which limits their effectiveness and scope in practical applications and leads to low monitoring efficiency. Summary of the invention

[0004] In view of this, in order to at least solve the technical problem of low wire monitoring efficiency in the related art, the present invention provides a wire dancing detection method, device, equipment and storage medium.

[0005] A first aspect of the present invention provides a wire galloping detection method, the method comprising:

[0006] Processing the current frame image collected in real time to obtain the center position of a predetermined marker in the current frame image; wherein the marker is a spacer rod on the conductor;

[0007] The frame number of the current frame image and the distance value between the center position of the marker in the current frame image and the reference position form a corresponding real array and record it in the real sequence;

[0008] When the first total number of real arrays included in the real sequence reaches the first set threshold for the first time, or when the first total number has reached the first set threshold and the real arrays included in the real sequence are updated, a linear fitting process is performed on the first set threshold number of real arrays that are most recently recorded in ascending order of frame numbers, with the frame number of each real array as an independent variable and the distance value as a dependent variable, to construct a real-time amplitude model of the marker;

[0009] Input the target frame number of the real array currently recorded most recently in the real sequence into the real-time amplitude model to obtain the fitting distance value corresponding to the target frame number, and form a corresponding fitting array with the target frame number and its corresponding fitting distance value, and record it in the fitting sequence;

[0010] When a second total number of fitting arrays included in the fitting sequence reaches a second set threshold for the first time, or when the second total number has reached the second set threshold and the fitting arrays included in the fitting sequence are updated, determining a current vibration trend of the marker according to the second set threshold number of fitting arrays newly recorded;

[0011] When the current vibration trend is different from the previous vibration trend, determining the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend; wherein the type of the trend inflection point is a peak or a trough;

[0012] When the determined trend turning points include crests and troughs, the dancing information of the wire is determined according to the fitting distance values ​​and / or frame numbers corresponding to the latest crests and troughs respectively; the dancing information includes vibration frequency and / or amplitude.

[0013] A second aspect of the present invention provides a wire dancing detection device, the device comprising:

[0014] The image processing module is configured to: process the current frame image collected in real time to obtain the center position of the predetermined marker in the current frame image; wherein the marker is a spacer on the wire;

[0015] The real sequence construction module is configured to: form a corresponding real array from the frame number of the current frame image and the distance value between the center position of the marker in the current frame image and the reference position and record it in the real sequence;

[0016] The amplitude model building module is configured to: when the first total number of real arrays included in the real sequence reaches the first set threshold for the first time, or when the first total number has reached the first set threshold and the real arrays included in the real sequence are updated, perform linear fitting processing on the first set threshold number of real arrays that have been recorded most recently in ascending order of frame numbers, use the frame number of each real array as an independent variable and the distance value as a dependent variable, and build a real-time amplitude model of the marker;

[0017] The fitting sequence building module is configured to: input the target frame number of the real array currently recorded most recently in the real sequence into the real-time amplitude model to obtain the fitting distance value corresponding to the target frame number, and form a corresponding fitting array with the target frame number and its corresponding fitting distance value, and record it in the fitting sequence;

[0018] a trend determination module, configured to: determine the current vibration trend of the marker according to the second set threshold number of fitting arrays newly recorded when the second total number of fitting arrays included in the fitting sequence reaches the second set threshold value for the first time, or when the second total number has reached the second set threshold value and the fitting arrays included in the fitting sequence are updated;

[0019] The inflection point determination module is configured to: when the current vibration trend is different from the previous vibration trend, determine the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend; wherein the type of the trend inflection point is a peak or a trough;

[0020] The wire dancing determination module is configured to: when the determined trend turning point includes a peak and a trough, determine the dancing information of the wire according to the fitting distance value and / or frame number corresponding to the latest peak and trough respectively; the dancing information includes vibration frequency and / or amplitude.

[0021] A third aspect of the present invention provides an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor is used to execute the machine executable instructions to implement the wire dancing detection method provided in the first aspect above.

[0022] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the wire galloping detection method provided in the first aspect is implemented.

[0023] Compared with the prior art, the present invention has at least the following beneficial effects:

[0024] The wire dancing detection method, device, equipment and storage medium provided by the embodiments of the present invention, on the one hand, by using the spacer rods on the wire as markers for detecting the wire dancing, the calculation difficulty and prediction point loss problems caused by directly monitoring the wire can be avoided, thereby reducing the errors and computational complexity caused thereby, which is conducive to reducing the difficulty of algorithm development and the difficulty of identifying the markers, and effectively improving the recognition efficiency; on the other hand, by calculating the center position of the marker in each frame of the real-time collected image based on the reference position of the marker, and the distance between the center position and the reference position, the movement of the marker, that is, the dancing of the wire, can be reflected, which can more accurately reflect the actual dancing of the wire, which is conducive to improving the detection of the wire. The detection accuracy of wire dancing; on the other hand, by forming real data from the distance and frame number of the center position of the marker in each frame image relative to the reference position, and constructing a real sequence, and then using the real sequence to construct the latest real-time amplitude model in real time, and then using the real-time amplitude model to calculate the fitting distance value corresponding to each target frame number based on the target frame number of the real data of the corresponding time period, and then using the target frame number and the corresponding fitting distance value as the fitting array, constructing a fitting sequence, and using the fitting sequence as the basis to obtain the dancing information of the wire, it can well avoid the phenomenon that the detection result characterizes that there are many negligible tiny dancing of the wire due to some interference or noise other than the dancing.

[0025] Therefore, combining the above aspects, not only can the automation of wire dancing monitoring be realized, without relying on manual patrols, which greatly reduces labor costs and improves the efficiency of wire dancing monitoring; but it also avoids the problems of high difficulty in developing algorithms, large image processing errors and low dancing detection accuracy caused by direct monitoring of wires, and by using linear fitting to calculate the vibration trend of the wire in real time, rather than simply calculating based on the center position of the marker, it can avoid the large error in the calculation of the wire dancing frequency caused by the easy data jitter of the single camera screen. Therefore, the various schemes provided in the embodiments of the present invention can greatly improve the accuracy and efficiency of wire dancing detection.

[0026] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of a wire galloping detection method according to an exemplary embodiment of the present invention;

[0028] Figure 2 is a structural block diagram of a wire dancing detection device according to an exemplary embodiment of the present invention;

[0029] Figure 3It is a hardware structure diagram of an electronic device in which a wire dancing detection device is located according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0030] Here, the exemplary embodiments will be described in detail, and the examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. Moreover, the implementations described in the following exemplary embodiments do not limit the present invention, and the structural, methodological, or functional changes made by ordinary technicians in the field according to these implementations are all included in the protection scope of the present invention.

[0031] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0032] It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0033] In order to at least solve the technical problems of high cost and low efficiency of wire monitoring in related technologies, an embodiment of the present invention provides a wire dancing detection method. The basic inventive concept is: on the one hand, by using the spacer rods on the wire as markers for detecting the dancing of the wire, the calculation difficulty and prediction point loss problems caused by directly monitoring the wire can be avoided, thereby reducing the errors and computational complexity caused by this, which is conducive to reducing the difficulty of algorithm development and the difficulty of identifying markers, and effectively improving the recognition efficiency; on the other hand, by calculating the center position of the marker in each frame of the real-time acquired image based on the reference position of the marker, and the distance between the center position and the reference position, the movement of the marker, that is, the dancing of the wire, can be reflected, which can more accurately reflect the wire. The real dancing situation is conducive to improving the detection accuracy of wire dancing; on the other hand, by forming real data from the distance and frame number of the center position of the marker in each frame image relative to the reference position, and constructing a real sequence, and then using the real sequence to construct the latest real-time amplitude model in real time, and then using the real-time amplitude model to calculate the fitting distance value corresponding to each target frame number based on the target frame number of the real data of the corresponding time period, and then using the target frame number and the corresponding fitting distance value as the fitting array to construct a fitting sequence, and using the fitting sequence as the basis to obtain the dancing information of the wire, it can well avoid the phenomenon that the detection result characterizes that there are many negligible tiny dancing of the wire due to some interference or noise other than the dancing.

[0034] Therefore, combining the above aspects, not only can the automation of wire dancing monitoring be realized, without relying on manual patrols, which greatly reduces labor costs and improves the efficiency of wire dancing monitoring; but it also avoids the problems of high difficulty in developing algorithms, large image processing errors and low dancing detection accuracy caused by direct monitoring of wires, and by using linear fitting to calculate the vibration trend of the wire in real time, rather than simply calculating based on the center position of the marker, it can avoid the large error in the calculation of the wire dancing frequency caused by the easy data jitter of the single camera screen. Therefore, the various schemes provided in the embodiments of the present invention can greatly improve the accuracy and efficiency of wire dancing detection.

[0035] The wire galloping detection method according to the embodiment of the present invention is described in more detail below, but the present invention should not be limited thereto.

[0036] The wire dancing detection method provided in the embodiment of the present invention can be applied to a video surveillance device, which can be deployed in an electric power scenario to monitor the wire area and process the images collected under monitoring to obtain the wire dancing situation. Of course, the wire dancing detection method provided in the embodiment of the present invention can also be applied to a monitoring center that is separated from the video surveillance device but has a communication connection. The monitoring center can receive image data sent by the video surveillance device, and process the image data by executing the wire dancing detection method provided in the embodiment of the present invention to obtain the wire dancing situation.

[0037] See also Figure 1 , Figure 1 : is a flow chart of a wire dancing detection method provided by the present invention according to an exemplary embodiment; the wire dancing detection method comprises:

[0038] In step S100, the current frame image collected in real time is processed to obtain the center position of a predetermined marker in the current frame image; wherein the marker is a spacer rod on the wire;

[0039] In step S200, the frame number of the current frame image and the distance value between the center position of the marker in the current frame image and the reference position are formed into a corresponding real array and recorded in the real sequence;

[0040] In step S300, when the first total number of real arrays included in the real sequence reaches the first set threshold for the first time, or when the first total number has reached the first set threshold and the real arrays included in the real sequence are updated, a linear fitting process is performed on the first set threshold number of real arrays that are newly recorded in ascending order of frame numbers, with the frame number of each real array as the independent variable and the distance value as the dependent variable, to construct a real-time amplitude model of the marker;

[0041] In step S400, the target frame number of the real array currently recorded most recently in the real sequence is input into the real-time amplitude model to obtain the fitting distance value corresponding to the target frame number, and the target frame number and its corresponding fitting distance value form a corresponding fitting array and record it in the fitting sequence;

[0042] In step S500, when the second total number of fitting arrays included in the fitting sequence reaches the second set threshold for the first time, or when the second total number has reached the second set threshold and the fitting arrays included in the fitting sequence are updated, the current vibration trend of the marker is determined according to the second set threshold number of fitting arrays newly recorded;

[0043] In step S600, when the current vibration trend is different from the previous vibration trend, the type and frame number of the trend inflection point are determined according to the current vibration trend and the previous vibration trend; wherein the type of the trend inflection point is a peak or a trough;

[0044] In step S700, when the determined trend turning point includes a peak and a trough, the dancing information of the wire is determined according to the fitting distance value and / or frame number corresponding to the latest peak and trough respectively; the dancing information includes vibration frequency and / or amplitude.

[0045] In the power inspection scenario, a video surveillance device can be installed in the conductor area of ​​the transmission line, and then the video surveillance device is used to collect images of the conductor area in real time, and the detection results of the conductor galloping are obtained by using steps S100 to S700. The process of conductor galloping detection is described below:

[0046] After the image of the wire area is captured in real time by the video surveillance equipment, the current frame image captured in real time can be processed through step S100 to identify the center position of the marker in the current frame image, wherein the marker and its center position can be identified by a pre-trained marker recognition model. The specific recognition principle can be found in the relevant technology and will not be elaborated here.

[0047] The marker may be any spacer on the conductor. However, in order to better reflect the actual galloping of the conductor and further improve the detection accuracy of the conductor galloping, in some embodiments, the conductor galloping detection method provided by the embodiment of the present invention further provides a solution for obtaining a spacer on the conductor that is representative of the galloping. That is, the conductor galloping detection method provided by the embodiment of the present invention may also include a process of determining the marker, and the determination process includes:

[0048] In step S010, each frame of the image in the wire region image sequence of a set time length is processed to obtain the center position of each spacer bar on the wire in each frame of the image;

[0049] In step S020, for each spacer bar, the maximum displacement of the spacer bar in the set time period is determined according to all center positions of the spacer bar in the set time period;

[0050] In step S030, a spacer bar having the largest maximum displacement value among all the spacers is used as a marker.

[0051] It should be noted that, in general, steps S010 to S030 must be executed before executing step S100, and since all the spacers will not change or change significantly in their subsequent positions once they are installed on the conductor, unless the spacers fall due to bad weather or aging. Therefore, generally speaking, as long as the steps S010 to S030 for determining the marker are executed once, the determined marker can always be used by step S100, so there is no need to repeat steps S010 to S030. However, if it is found that the above-mentioned marker cannot be identified during the execution of step S100, in order to ensure that the conductor dancing result can be detected smoothly and the detection accuracy can be guaranteed as much as possible, in some embodiments, when the marker cannot be identified during the processing of the current frame image collected in real time, the conductor dancing detection method provided by the embodiment of the present invention may also include:

[0052] In step S040, the process of determining the marker is returned to obtain a new marker;

[0053] In step S050, the predetermined marker is updated to the new marker.

[0054] That is, when it is found that the above-mentioned marker cannot be identified during the execution of step S100, steps S010-S030 can be executed to redefine the marker, and the marker determined at this time replaces the original marker to perform subsequent wire dancing monitoring.

[0055] The following describes the working process of obtaining the marker through steps S010 to S030:

[0056] First, the video equipment can be controlled to continuously work and set the duration, so as to collect and obtain the wire area image sequence, and then by step S010, the recognition process of all the spacer bars is carried out to each frame image in the wire area image sequence, for example, the model framework of the deep learning model YOLOV4 can be used as the image recognition model, thus using the YOLOV4 model as the one-stage target detection algorithm, to skip the first stage of generating the selected area in the traditional two-stage target detection algorithm, to directly generate the category probability and position coordinates of the spacer bars, thus realizing that a single detection can obtain the final spacer bar detection result, faster, and can better meet the needs of video real-time analysis. After identifying all the spacer bars in each frame image, the real-time position rectangular coordinates of all the spacer bars in each frame image can be obtained, and the real-time position rectangular coordinates can be understood as the coordinates of the circumscribed rectangle of the spacer bars, and the left vertex coordinates and the right vertex coordinates of the circumscribed rectangle can be included, and then based on the mathematical calculation principle, the center point coordinates of the circumscribed rectangle can be calculated, and thus the center position of each spacer bar on the wire in each frame image can be obtained.

[0057] Next, execute step S020, for each spacer bar, calculate its maximum displacement in the set time length according to all the center positions of the spacer bar in the set time length, in some examples, the distance between the two center positions corresponding to each two adjacent frames of images can be calculated, and the maximum value of the distance is used as the maximum displacement. However, in order to more accurately reflect the size of the movable range of the spacer bar, so as to better obtain the spacer bar with the highest dancing flexibility, in some embodiments, the wire dancing detection method provided by the embodiment of the present invention also provides a better maximum displacement determination scheme, that is, in the step S020, the step of determining the maximum displacement of the spacer bar in the set time length according to all the center positions of the spacer bar in the set time length includes: calculating the distance between each two center positions of the spacer bar in the set time length, and using the maximum distance therein as the maximum displacement of the spacer bar in the set time length.

[0058] That is, without considering the time proximity, the maximum movable range of the entire set duration is considered, and the maximum displacement that best reflects the dancing flexibility of the spacer rod is selected.

[0059] After obtaining the maximum displacement of each spacer bar through any of the above embodiments, step S030 can be executed to use a spacer bar with the largest maximum displacement value among all the spacers as a marker, thereby obtaining a spacer bar with the highest dancing flexibility, which can better reflect the dancing of the conductor.

[0060] It can be seen that the marker predetermined in step S100 is the marker mentioned in any of the above embodiments. After the center position of the marker in the current frame image is determined in step S100, step S200 can be executed to further calculate the distance value between the center position and the predetermined reference position, and then the frame number and distance value of the current frame image are formed into a set of real arrays and then recorded in the real sequence.

[0061] In the above, the reference position refers to the position of the marker when it is stationary, which can be obtained by processing based on the image of the marker when it is stationary. However, for video surveillance equipment, it is impossible to determine whether the marker is stationary, and thus it is impossible to know which image is the image of the marker when it is stationary. At this time, human intervention is required, which may cause the acquisition of the reference position to be too subjective and unable to meet real-time requirements. Therefore, in order to solve this technical problem, in some embodiments, the wire dancing detection method provided in the embodiment of the present invention also provides an automatic determination scheme for the reference position, that is, the wire dancing detection method provided in the embodiment of the present invention also includes a process for determining the reference position of the marker, and the determination process includes:

[0062] In step S061, a corresponding circumscribed circle is determined according to all center positions of the marker in the set time period, and the center of the circumscribed circle is used as the reference position of the marker; or

[0063] In step S062, the mean values ​​of the horizontal coordinates and the vertical coordinates of all the center positions of the marker in the set time period are calculated to obtain the reference position of the marker.

[0064] It can be understood that the wire dancing detection method provided in the embodiment of the present invention provides two solutions for determining the reference position, one of which is the solution recorded in step S061, and the other is the solution recorded in step S062.

[0065] For step S061, a circumscribed polygon can be generated based on all the center positions of the marker in the set time, and then the circumscribed circle of the circumscribed polygon is obtained, and then the center coordinates of the circumscribed circle are obtained, and the center coordinates of the circumscribed circle are the reference position of the marker. The calculation principle of the circumscribed polygon, the circumscribed circle and the center coordinates thereof can be referred to in the relevant technology, which will not be described in detail here.

[0066] For step S062, the calculation process of the reference position can be simplified and the calculation efficiency of the reference position can be improved. Since all the center positions of the marker within the set time length are known, the horizontal coordinate mean and the vertical coordinate mean can be directly calculated using all the center positions. The calculated horizontal coordinate mean and vertical coordinate mean can be used as the reference position of the marker.

[0067] It can be seen from this that the reference position in step S200 is calculated in advance and only needs to be applied later. There is no need to calculate the reference position every time step S200 is executed.

[0068] In addition, the duration set above can be set according to actual experience or needs. For example, it can be set to 1 minute, but it is not limited thereto.

[0069] After continuously processing each current frame image collected in real time through step S100 and step S200, many groups of real arrays will be obtained. Therefore, the total number of arrays in the real sequence will also continue to increase. Therefore, when the first total number of real arrays contained in the real sequence reaches the first set threshold for the first time, step S300 can be executed to perform linear fitting processing on the first set threshold real arrays in the order of frame number from small to large, with the frame number as the independent variable and the distance value as the dependent variable, to construct a real-time amplitude model of the marker at this time, the amplitude model can be a curve model, or it can be understood as a function expression, for example, y=f(x), where y represents the distance value, x represents the frame number, and y=f(x) represents the functional relationship between the distance value and the frame number. The first set threshold represents the number of real arrays that can meet the construction requirements of the amplitude model, for example, it can be 10, or it can be 11 or more, and the embodiment of the present invention does not make specific limitations on this.

[0070] After obtaining the above-mentioned real-time amplitude model, step S400 can be executed to calculate the fitting distance value corresponding to the current frame image at this time, that is, the target frame number of the current frame image, that is, the target frame number of a real array most recently recorded in the real array contained in the real sequence, is input into the real-time amplitude model corresponding to the current frame image, and then the fitting distance value corresponding to the target frame number is obtained, and then the target frame number and the fitting distance value are formed into a set of fitting arrays and recorded in the fitting sequence.

[0071] Since steps S100~S200 are continuously executed in a loop, during this continuous loop execution process, each time a frame of image is processed, the arrays in the real sequence and the fitting sequence will be updated once, so the arrays contained in the real sequence and the fitting sequence gradually increase. Therefore, in order to obtain the real-time amplitude model corresponding to each frame of image, for the real sequence, it is necessary to recalculate the real-time amplitude model corresponding to the current frame of image through step S300. Therefore, in addition to being triggered when the first total number of real arrays contained in the real sequence reaches the first set threshold for the first time, the execution of step S300 can also be triggered when the first total number of real arrays contained in the real sequence has reached the first set threshold and the real arrays contained in the real sequence are updated. When triggered by the latter, the current real-time amplitude model of the marker is constructed based on the first set threshold real arrays recorded most recently.

[0072] After each update of the real-time amplitude model, step S400 is executed once to input the target frame number of the real array most recently recorded in the real array into the real-time amplitude model currently updated to obtain the corresponding fitting distance value, and form a set of fitting arrays and record them in the fitting sequence.

[0073] It can be seen that in the real sequence, except for the first set threshold real arrays of the latest record, other arrays will not be used to construct the current real-time amplitude model, thereby ensuring the real-time nature of subsequent vibration trends.

[0074] As for the fitting sequence, it can be known from the above related records that since the fitting distance value is calculated based on the current latest frame number each time the real-time amplitude model is updated, the arrays included in the fitting sequence will gradually increase. When the second total number of the fitting arrays included in the fitting sequence reaches the second set threshold for the first time, step S400 can be executed to process the second set threshold number of fitting arrays, for example, to perform linear fitting processing to obtain the corresponding trend curve, and then to know the current vibration trend of the marker.

[0075] After that, due to the above reasons, the number of arrays included in the fitting sequence continues to increase. Therefore, in order to obtain the latest current vibration trend, after the total number of fitting arrays included in the fitting sequence has reached the second set threshold, each update of the fitting arrays included therein will trigger the re-execution of step S500 to re-determine the latest current vibration trend of the marker. It can be understood that the execution of step S500 is triggered not only when the second total number of fitting arrays included in the fitting sequence reaches the second set threshold for the first time, but also when the second total number of fitting arrays included in the fitting sequence has reached the second set threshold and the fitting arrays included in the fitting sequence are updated. When triggered by the latter, the current vibration trend of the marker is determined based on the latest record of the second set threshold number of fitting arrays.

[0076] It can be seen that in the fitting sequence, other fitting arrays except the second set threshold fitting arrays of the latest record will not be used to determine the current vibration trend, thereby ensuring the real-time nature of the current vibration trend.

[0077] In some embodiments, in order to improve the accuracy of determining the current vibration trend, the wire galloping detection method provided by the embodiment of the present invention further provides a scheme for determining the current vibration trend, that is, in step S500, the step of determining the current vibration trend of the marker according to the second set threshold fitting array of the latest record includes:

[0078] In step S510, when it is determined that the corresponding trend curve includes an upward trend and a downward trend according to the second set threshold number of fitting arrays recorded in the latest record, the total number of arrays corresponding to the upward trend and the downward trend are determined respectively, and the trend with the larger total number of corresponding arrays is taken as the current vibration trend of the marker;

[0079] In step S520, when the corresponding trend curve is determined to be a monotonous rising curve according to the second set threshold number of fitting arrays recorded most recently, it is determined that the current vibration trend of the marker is an rising trend;

[0080] In step S530, when it is determined that the corresponding trend curve is a monotonically decreasing curve according to the second set threshold denoising values ​​recorded most recently, it is determined that the current vibration trend of the marker is a decreasing trend.

[0081] For step S510, when the trend curve contains both an upward trend and a downward trend, the total number of arrays is used as the basis to take the trend with the larger total number of arrays as the current vibration trend of the marker. For example, assuming that the total number of arrays corresponding to the upward trend is 4 and the total number of arrays corresponding to the downward trend is 6, and the total number of arrays corresponding to the downward trend is larger than the total number of arrays corresponding to the upward trend, the current vibration trend of the marker is a downward trend. It can be seen that if the second set threshold is an even number, it may happen that the total number of arrays corresponding to the upward trend is 5 and the total number of arrays corresponding to the downward trend is also 5, which causes step S510 to fail to execute smoothly. Therefore, in order to solve this technical problem, in some embodiments, the second set threshold can be set to an odd number and can meet the requirements for constructing the trend curve. In some examples, the second set threshold can be set to 9 or 11, but is not limited thereto.

[0082] In addition, in order to facilitate the recording of the current vibration trend, a number can be used to express the current vibration trend. For example, if the current vibration trend is an upward trend, it can be recorded as 1; conversely, if the current vibration trend is a downward trend, it can be recorded as -1. It should be understood that the vibration trend is a trend that characterizes the amplitude, for example, whether the amplitude is decreasing or increasing with the frame number.

[0083] After determining the current vibration trend of the marker, the current vibration trend of the marker can be compared with the previous vibration trend to determine whether the two are the same. If the two are different, it means that the marker has a peak or a trough. At this time, it is necessary to find the new peak or trough. Therefore, step S600 is executed to determine whether the trend inflection point is a peak or a trough based on the current vibration trend and the previous vibration trend. A set of trend inflection point type determination schemes are given below:

[0084] In some embodiments, in step S600, the step of determining the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend may include:

[0085] In step S610, when the current vibration trend is an upward trend and the previous vibration trend is a downward trend, the type of the corresponding trend inflection point is determined to be a peak, and the frame number and fitting distance value corresponding to the peak are obtained;

[0086] In step S620, when the current vibration trend is a downward trend and the previous vibration trend is an upward trend, it is determined that the type of the corresponding trend inflection point is a trough, and the frame number and fitting distance value corresponding to the trough are obtained.

[0087] After recording the type and frame number of the inflection point, it can be determined whether the recorded inflection point already contains a peak and a trough. If so, it means that the wire has completed a dancing cycle, so the dancing information of the wire can be calculated. If not, for example, there may be only one peak and the trough has not appeared yet, then the dancing information of the wire cannot be calculated at this time, and it is necessary to wait until the trough appears before calculating.

[0088] Therefore, when the determined trend turning points include peaks and troughs, step S700 can be executed to determine the dancing information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs. The following describes the process of determining the dancing information of the wire:

[0089] When the dancing information includes vibration frequency, in step S700, the step of determining the dancing information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs respectively includes:

[0090] In step S710, the difference between the frame number corresponding to the latest peak and the frame number corresponding to the latest trough is calculated;

[0091] In step S720, the ratio of the number of frames collected per second when collecting images to the difference is calculated to obtain the vibration frequency of the wire.

[0092] When the dancing information includes amplitude, in step S700, the step of determining the dancing information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs respectively includes:

[0093] In step S730, the difference between the fitting distance value corresponding to the latest crest and the fitting distance value corresponding to the latest trough is calculated to obtain the amplitude of the wire.

[0094] When the dancing information includes vibration frequency and amplitude, in step S700, the step of determining the dancing information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs respectively includes:

[0095] In step S710, the difference between the frame number corresponding to the latest peak and the frame number corresponding to the latest trough is calculated;

[0096] In step S720, the ratio of the number of frames collected per second when collecting images to the difference is calculated to obtain the vibration frequency of the wire;

[0097] In step S730, the difference between the fitting distance value corresponding to the latest crest and the fitting distance value corresponding to the latest trough is calculated to obtain the amplitude of the wire.

[0098] It should be noted that there is no strict execution order for step S730 and step S710, and the two can be executed in parallel or in series.

[0099] After obtaining the dancing information of the wire, in order to timely warn the wire with abnormal dancing and timely intervene in the wire with abnormal dancing to avoid irreversible consequences, in some embodiments, the wire dancing detection method provided by the embodiment of the present invention also provides an alarm scheme for abnormal dancing, that is, after determining that the dancing information of the wire is obtained, the wire dancing detection method may also include:

[0100] In step S800, when the vibration frequency of the wire is greater than a set frequency threshold, or the amplitude of the wire is greater than a set amplitude threshold, an alarm message is sent to a server or to a bound client.

[0101] The frequency threshold and the amplitude threshold may be set based on experience or experiments, and are not limited in this regard in the embodiment of the present invention.

[0102] In addition, since the number of arrays in the real sequence and the fitting sequence will increase with the number of collected frames, the gradually increasing arrays will occupy a certain amount of storage space and will also affect the data acquisition efficiency. Therefore, in some embodiments, in order to solve this technical problem while ensuring the comprehensiveness, stability and accuracy of the current vibration trend construction, for the real sequence and the fitting sequence, when it is determined that there is a peak or a trough based on the current vibration trend, all arrays before the currently determined peak or trough in the real sequence and the fitting sequence can be deleted, thereby achieving the purpose of reducing the storage space occupied by useless data and improving data acquisition efficiency. When a peak or a trough is obtained, it indicates that the direction of the current vibration trend has reached a comprehensive and stable state. Even if the array before the peak or the trough is deleted, it will not cause adverse effects on the subsequent determination of the vibration trend.

[0103] It is worth noting that the embodiments or technical features of the wire dancing detection method provided by the present invention can be combined with each other to form a new technical solution as long as there is no combination contradiction.

[0104] Corresponding to the above-mentioned embodiment of the wire dancing detection method, the embodiment of the present invention further provides a wire dancing detection device, such as Figure 2 As shown, Figure 2 : is a structural block diagram of a wire dancing detection device according to an exemplary embodiment of the present invention. The wire dancing detection device 200 may include:

[0105] The image processing module 210 is configured to: process the current frame image collected in real time to obtain the center position of the predetermined marker in the current frame image; wherein the marker is a spacer on the wire;

[0106] The real sequence construction module 220 is configured to: form a corresponding real array from the frame number of the current frame image and the distance value between the center position of the marker in the current frame image and the reference position and record it in the real sequence;

[0107] The amplitude model building module 230 is configured to: when the first total number of real arrays included in the real sequence reaches the first set threshold for the first time, or when the first total number has reached the first set threshold and the real arrays included in the real sequence are updated, perform linear fitting processing on the first set threshold number of real arrays that have been recorded most recently in ascending order of frame numbers, using the frame number of each real array as an independent variable and the distance value as a dependent variable, to build a real-time amplitude model of the marker;

[0108] The fitting sequence construction module 240 is configured to: input the target frame number of the real array currently recorded most recently in the real sequence into the real-time amplitude model to obtain the fitting distance value corresponding to the target frame number, and form a corresponding fitting array with the target frame number and its corresponding fitting distance value, and record it in the fitting sequence;

[0109] The trend determination module 250 is configured to: determine the current vibration trend of the marker according to the second set threshold number of fitting arrays newly recorded when the second total number of fitting arrays included in the fitting sequence reaches the second set threshold value for the first time, or when the second total number has reached the second set threshold value and the fitting arrays included in the fitting sequence are updated;

[0110] The inflection point determination module 260 is configured to: when the current vibration trend is different from the previous vibration trend, determine the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend; wherein the type of the trend inflection point is a peak or a trough;

[0111] The wire dancing determination module 270 is configured to: when the determined trend turning point includes a peak and a trough, determine the dancing information of the wire according to the fitting distance value and / or frame number corresponding to the latest peak and trough respectively; the dancing information includes vibration frequency and / or amplitude.

[0112] In some embodiments, the process of determining the current vibration trend of the marker by the trend determination module 250 according to the second set threshold number of fitting arrays recorded most recently is configured as follows:

[0113] When it is determined that the corresponding trend curve includes an upward trend and a downward trend according to the second set threshold number of fitting arrays recorded in the latest record, the total number of arrays corresponding to the upward trend and the total number of arrays corresponding to the downward trend are determined respectively, and the trend with the larger total number of corresponding arrays is taken as the current vibration trend of the marker;

[0114] When it is determined that the corresponding trend curve is a monotonous rising curve according to the second set threshold fitting array of the latest record, determining that the current vibration trend of the marker is an rising trend;

[0115] When it is determined that the corresponding trend curve is a monotonically decreasing curve according to the second set threshold denoising value of the latest record, it is determined that the current vibration trend of the marker is a decreasing trend.

[0116] In some embodiments, the process of the inflection point determination module 260 determining the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend is configured as follows:

[0117] When the current vibration trend is an upward trend and the previous vibration trend is a downward trend, determining the type of the corresponding trend inflection point is a peak, and obtaining the frame number and fitting distance value corresponding to the peak;

[0118] When the current vibration trend is a downward trend and the previous vibration trend is an upward trend, the type of the corresponding trend inflection point is determined to be a trough, and the frame number and the fitting distance value corresponding to the trough are obtained.

[0119] In some embodiments, the dancing information includes vibration frequency;

[0120] Accordingly, the process of the wire galloping determination module 270 determining the galloping information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs is configured as follows:

[0121] Calculate the difference between the frame number corresponding to the latest peak and the frame number corresponding to the latest trough;

[0122] The ratio of the number of frames collected per second when collecting images to the difference is calculated to obtain the vibration frequency of the wire.

[0123] In some embodiments, the dancing information includes amplitude;

[0124] Accordingly, the process of the wire galloping determination module 270 determining the galloping information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs is configured as follows:

[0125] The difference between the fitting distance value corresponding to the latest crest and the fitting distance value corresponding to the latest trough is calculated to obtain the amplitude of the wire.

[0126] In some embodiments, the dancing information includes vibration frequency and amplitude;

[0127] Accordingly, the process of the wire galloping determination module 270 determining the galloping information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs is configured as follows:

[0128] Calculate the difference between the frame number corresponding to the latest peak and the frame number corresponding to the latest trough;

[0129] Calculating the ratio of the number of frames collected per second when collecting images to the difference value to obtain the vibration frequency of the wire;

[0130] The difference between the fitting distance value corresponding to the latest crest and the fitting distance value corresponding to the latest trough is calculated to obtain the amplitude of the wire.

[0131] In some embodiments, the wire dancing detection device 200 further includes:

[0132] The alarm module is configured to send an alarm message to a server or a bound client when the vibration frequency of the wire is greater than a set frequency threshold, or the amplitude of the wire is greater than a set amplitude threshold.

[0133] In some embodiments, the wire dancing detection device 200 further includes:

[0134] The marker determination module is configured to:

[0135] Process each frame of the wire area image sequence of a set time length to obtain the center position of each spacer bar on the wire in each frame of the image;

[0136] For each spacer bar, determining the maximum displacement of the spacer bar during the set time period according to all center positions of the spacer bar during the set time period;

[0137] Among all the spacers, the one with the largest maximum displacement value is used as a marker.

[0138] In some embodiments, the process of determining the maximum displacement of the spacer bar in the set time period according to all center positions of the spacer bar in the set time period by the marker determination module is configured as follows:

[0139] The distance between each two center positions of the spacer bar during the set time period is calculated, and the maximum distance therebetween is used as the maximum displacement of the spacer bar during the set time period.

[0140] In some embodiments, the wire dancing detection device 200 further includes:

[0141] The reference position determination module is configured to:

[0142] According to all the center positions of the marker in the set time period, a corresponding circumscribed circle is determined, and the center of the circumscribed circle is used as the reference position of the marker; or,

[0143] The mean values ​​of the horizontal coordinates and the vertical coordinates of all the center positions of the marker in the set time period are calculated to obtain the reference position of the marker.

[0144] In some embodiments, the wire dancing detection device 200 further includes:

[0145] The calling module is configured to: when the image processing module 210 processes the current frame image collected in real time and fails to identify the marker, trigger the marker determination module to obtain a new marker; and update the predetermined marker to the new marker.

[0146] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units.

[0147] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0148] Corresponding to the embodiment of the aforementioned wire dancing detection method, the present invention further provides an electronic device, the electronic device comprising:

[0149] processor;

[0150] a memory for storing a computer program executable by the processor;

[0151] Wherein, when the processor executes the program, the steps of the wire dancing detection method in any of the aforementioned method embodiments are implemented.

[0152] The embodiment of the wire dancing detection device provided by the present invention can be applied to electronic devices. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of the electronic device in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory and running them. From the hardware level, if Figure 3 As shown, Figure 3is a hardware structure diagram of an electronic device in which a wire dancing detection device is located according to an exemplary embodiment of the present invention, except Figure 3 In addition to the processor 310, memory 330, interface 320, and non-volatile memory 340 shown, the electronic device where the wire dancing detection device 200 is located in the embodiment may also include other hardware according to the actual function of the electronic device, which will not be described in detail.

[0153] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of the wire galloping detection method in any of the aforementioned method embodiments are implemented.

[0154] The present invention may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer-readable storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0155] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A wire galloping detection method, It is characterized in that include: Processing the current frame image collected in real time to obtain the center position of a predetermined marker in the current frame image; wherein the marker is a spacer rod on the conductor; The frame number of the current frame image and the distance value between the center position of the marker in the current frame image and the reference position form a corresponding real array and record it in the real sequence; When the first total number of real arrays included in the real sequence reaches the first set threshold for the first time, or when the first total number has reached the first set threshold and the real arrays included in the real sequence are updated, a linear fitting process is performed on the first set threshold number of real arrays that are most recently recorded in ascending order of frame numbers, with the frame number of each real array as an independent variable and the distance value as a dependent variable, to construct a real-time amplitude model of the marker; Input the target frame number of the real array currently recorded most recently in the real sequence into the real-time amplitude model to obtain the fitting distance value corresponding to the target frame number, and form a corresponding fitting array with the target frame number and its corresponding fitting distance value, and record it in the fitting sequence; When a second total number of fitting arrays included in the fitting sequence reaches a second set threshold for the first time, or when the second total number has reached the second set threshold and the fitting arrays included in the fitting sequence are updated, determining a current vibration trend of the marker according to the second set threshold number of fitting arrays newly recorded; When the current vibration trend is different from the previous vibration trend, determining the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend; wherein the type of the trend inflection point is a peak or a trough; When the determined trend turning points include crests and troughs, the dancing information of the wire is determined according to the fitting distance values ​​and / or frame numbers corresponding to the latest crests and troughs respectively; the dancing information includes vibration frequency and / or amplitude.

2. The method according to claim 1, It is characterized in that The process of determining the marker is also included, and the determination process includes: Process each frame of the wire area image sequence of a set time length to obtain the center position of each spacer bar on the wire in each frame of the image; For each spacer bar, determining the maximum displacement of the spacer bar during the set time period according to all center positions of the spacer bar during the set time period; Among all the spacers, the one with the largest maximum displacement value is used as a marker.

3. The method according to claim 2, It is characterized in that The method further includes a process of determining a reference position of the marker, the determination process comprising: According to all the center positions of the marker in the set time period, a corresponding circumscribed circle is determined, and the center of the circumscribed circle is used as the reference position of the marker; or, The mean values ​​of the horizontal coordinates and the vertical coordinates of all the center positions of the marker in the set time period are calculated to obtain the reference position of the marker.

4. The method according to claim 2, It is characterized in that In the process of processing the current frame image collected in real time, if the marker cannot be identified, the method further includes: Returning to execute the process of determining the marker to obtain a new marker; The predetermined markers are updated to the new markers.

5. The method according to any one of claims 1 to 4, It is characterized in that The step of determining the current vibration trend of the marker according to the second set threshold fitting array of the latest record comprises: When it is determined that the corresponding trend curve includes an upward trend and a downward trend according to the second set threshold number of fitting arrays recorded in the latest record, the total number of arrays corresponding to the upward trend and the total number of arrays corresponding to the downward trend are determined respectively, and the trend with the larger total number of corresponding arrays is taken as the current vibration trend of the marker; When it is determined that the corresponding trend curve is a monotonous rising curve according to the second set threshold fitting array of the latest record, determining that the current vibration trend of the marker is an rising trend; When it is determined that the corresponding trend curve is a monotonically decreasing curve according to the second set threshold denoising value of the latest record, it is determined that the current vibration trend of the marker is a decreasing trend.

6. The method according to claim 1, It is characterized in that The step of determining the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend comprises: When the current vibration trend is an upward trend and the previous vibration trend is a downward trend, determining the type of the corresponding trend inflection point is a peak, and obtaining the frame number and fitting distance value corresponding to the peak; When the current vibration trend is a downward trend and the previous vibration trend is an upward trend, the type of the corresponding trend inflection point is determined to be a trough, and the frame number and the fitting distance value corresponding to the trough are obtained.

7. The method according to claim 6, It is characterized in that The dancing information includes vibration frequency; The step of determining the dancing information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs respectively comprises: Calculate the difference between the frame number corresponding to the latest peak and the frame number corresponding to the latest trough; Calculating the ratio of the number of frames collected per second when collecting images to the difference value to obtain the vibration frequency of the wire; and / or The dancing information includes amplitude; The step of determining the dancing information of the wire according to the fitting distance values ​​and / or frame numbers corresponding to the latest peaks and troughs respectively comprises: The difference between the fitting distance value corresponding to the latest crest and the fitting distance value corresponding to the latest trough is calculated to obtain the amplitude of the wire.

8. A wire dancing detection device, It is characterized in that The device comprises: The image processing module is configured to: process the current frame image collected in real time to obtain the center position of the predetermined marker in the current frame image; wherein the marker is a spacer on the wire; The real sequence construction module is configured to: form a corresponding real array from the frame number of the current frame image and the distance value between the center position of the marker in the current frame image and the reference position and record it in the real sequence; The amplitude model building module is configured to: when the first total number of real arrays included in the real sequence reaches the first set threshold for the first time, or when the first total number has reached the first set threshold and the real arrays included in the real sequence are updated, perform linear fitting processing on the first set threshold number of real arrays that have been recorded most recently in ascending order of frame numbers, use the frame number of each real array as an independent variable and the distance value as a dependent variable, and build a real-time amplitude model of the marker; The fitting sequence building module is configured to: input the target frame number of the real array currently recorded most recently in the real sequence into the real-time amplitude model to obtain the fitting distance value corresponding to the target frame number, and form a corresponding fitting array with the target frame number and its corresponding fitting distance value, and record it in the fitting sequence; a trend determination module, configured to: determine the current vibration trend of the marker according to the second set threshold number of fitting arrays newly recorded when the second total number of fitting arrays included in the fitting sequence reaches the second set threshold value for the first time, or when the second total number has reached the second set threshold value and the fitting arrays included in the fitting sequence are updated; The inflection point determination module is configured to: when the current vibration trend is different from the previous vibration trend, determine the type and frame number of the trend inflection point according to the current vibration trend and the previous vibration trend; wherein the type of the trend inflection point is a peak or a trough; The wire dancing determination module is configured to: when the determined trend turning point includes a peak and a trough, determine the dancing information of the wire according to the fitting distance value and / or frame number corresponding to the latest peak and trough respectively; the dancing information includes vibration frequency and / or amplitude.

9. An electronic device, It is characterized in that It comprises a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor is used to execute the machine executable instructions to implement the wire dancing detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the wire galloping detection method according to any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Data detection method and system

    CN115859200A

  • Line galloping monitoring and predicting method fusing computer vision and inertial sensor

    CN116753981A

  • Power transmission line windage yaw galloping early warning method and system based on twin network

    CN118157324A