A method for detecting abnormality of shock-absorbing hammer

Abstract

The present invention discloses a method for detecting an abnormality of a shock-absorbing hammer, comprising the following steps: (1) a front-end device collects images in real time; (2) performing target detection on the image and extracting the shock-absorbing hammer area containing the shock-absorbing hammer; (3) extracting the conductor area; (4) setting an acquisition time interval; (5) identifying the position of the shock-absorbing hammer for the first time; and executing step (6) according to the acquisition time interval; (6) identifying the position of the shock-absorbing hammer again; and executing step (7) if the shock-absorbing hammer is detected in the conductor area; (7) detecting whether the shock-absorbing hammer is displaced; and (8) repeating step (6) and step (7). The present invention has high detection efficiency and avoids errors related to various human factors. It can directly collect images of the shock-absorbing hammer installed on the conductor for abnormality detection and accurately judge the abnormal state of the shock-absorbing hammer.

Description

Technical Field

[0001] The present invention relates to the technical field of shock-absorbing hammer abnormality detection, and more particularly to a shock-absorbing hammer abnormality detection method. Background Art

[0002] To reduce vibration caused by external forces on power lines, vibration dampers are used in transmission lines. Transmission lines are typically located high in the air, with large spans. When subjected to external forces, these lines are prone to vibration. Frequent vibration can cause fatigue from periodic bending, leading to strand breakage, hardware wear, and wire disconnection. Vibration dampers are typically installed on either side of the suspension point. These dampers produce a motion that is in phase with the conductor vibration, thereby reducing or eliminating it.

[0003] Transmission lines are typically constructed from multiple strands of metal wire wrapped around an iron core. In practice, conductors often experience creep under load, resulting in a decrease in cross-sectional diameter and an increase in length. This reduction in cross-sectional area inevitably leads to a lack of grip on the shock-absorbing hammer's clamping plate, causing it to loosen. Under the influence of factors such as its own weight and wind, the shock-absorbing hammer often slides from its installed position to the lowest point of the conductor's sag. Displacement or even detachment of the shock-absorbing hammer generally has the following adverse consequences: the shock-absorbing hammer cannot absorb energy and prevent conductor vibration; improperly positioned shock-absorbing hammers increase conductor vibration and accelerate fatigue damage; restoring the shock-absorbing hammer is time-consuming and labor-intensive, placing heavy strain on maintenance workers; and in severe cases, power outages may be required, impacting line reliability. Therefore, displacement is a common anomaly of shock-absorbing hammers, significantly impacting not only their effectiveness but also the operation of the entire line.

[0004] Currently, most methods for detecting abnormalities in shock absorbers are performed manually. This method has low detection efficiency and is prone to errors related to various human factors. The use of machine vision to detect shock absorbers can often only target a certain type of appearance defect and cannot accurately determine the abnormal state of the shock absorber. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for detecting abnormalities of a shock-absorbing hammer, which has high detection efficiency and avoids errors related to various human factors. It can directly capture images of the shock-absorbing hammer installed on the wire for abnormality detection and accurately judge the abnormal state of the shock-absorbing hammer.

[0006] To achieve the above object, a method for detecting anomalies of a shock-absorbing hammer is provided, comprising the following steps:

[0007] The front-end device collects images in real time;

[0008] (2) Detect the target in the image and extract the shock-absorbing hammer area containing the shock-absorbing hammer;

[0009] (3) Extracting the wire area: Using the Hough line detection method, mark the wire position and wire extension direction as the wire area;

[0010] (4) Set the collection time interval;

[0011] (5) Identify the shock-absorbing hammer position for the first time: record the shock-absorbing hammer area in the image where the shock-absorbing hammer is first detected, and determine the position of the shock-absorbing hammer hanging wire in combination with the wire position; execute step (6) according to the acquisition time interval;

[0012] (6) Identify the position of the shock-absorbing hammer again: if the shock-absorbing hammer area is not detected in the area of ​​the conductor, it is considered that the shock-absorbing hammer in the area is detached; if the shock-absorbing hammer is detected in the area of ​​the conductor, execute step (7);

[0013] (7) Detect whether the shock-absorbing hammer has shifted: find a shock-absorbing hammer hanging point and the coordinates of the shock-absorbing hammer hanging point in the area where the conductor area and the shock-absorbing hammer area overlap; obtain the shock-absorbing hammer hanging point and the coordinates of the shock-absorbing hammer hanging point in the next time interval, and perform a differential operation on the coordinates of the two shock-absorbing hammer hanging points. If it is calculated that the shock-absorbing hammer hanging point is offset along the conductor direction, it is considered that the shock-absorbing hammer has shifted; otherwise, it is considered that there is no abnormality in the shock-absorbing hammer;

[0014] (8) Repeat steps (6) to (7).

[0015] Particularly, the front-end device adopts a visible light spherical camera.

[0016] In particular, in step (2), the method for extracting the shock-absorbing hammer area is specifically as follows: identifying the area of ​​the target shock-absorbing hammer by using the YOLOv5 algorithm, taking the above area as the region of interest, and further extracting the shock-absorbing hammer contour in the region of interest using the Sobel edge detection algorithm.

[0017] In particular, in step (3), the specific method for extracting the wire area is to track the intersection points between the curves corresponding to each point in the image, and when the number of curves intersecting at a point exceeds a certain threshold, obtain the parameter pairs represented by all the above intersection points and mark them as a straight line in the image.

[0018] Particularly, the Hough line detection method adopts a standard Hough transform line detection method or a cumulative probability Hough transform line detection method.

[0019] In particular, in step (7), a specific method for finding a shock-absorbing hammer suspension point is to calculate the center point of the area where the shock-absorbing hammer area and the conductor area overlap, and mark the center point as the shock-absorbing hammer suspension point.

[0020] In particular, in step (7), the specific method for obtaining the overlapping area between the wire area and the shock-absorbing hammer area is as follows: the image after filling the shock-absorbing hammer outline is subjected to a Boolean operation with the wire area in the previous section to obtain the overlapping area between the shock-absorbing hammer and the wire.

[0021] Particularly, the collection time interval is 1 hour.

[0022] The beneficial effects of the present invention are as follows:

[0023] This invention proposes a method for detecting anomalies in a shock absorber. This method uses target detection to extract the shock absorber area and the Hough line detection method to extract the conductor area. A collection interval is set to detect the shock absorber area. Whether the shock absorber area is detected determines whether the shock absorber has fallen off. If it has not fallen off, the shock absorber's suspension point is set. The coordinates of the two shock absorber suspension points separated by the collection interval are then differentiated to determine whether the shock absorber suspension point has deviated along the conductor, thereby determining whether the shock absorber has deviated. This method boasts high detection efficiency and avoids errors associated with various human factors. It can directly capture images of the shock absorber installed on the conductor for anomaly detection, enabling accurate determination of shock absorber anomalies.

[0024] This paper uses YOLOv5 and CSPDarknet53 as the baseline network to extract rich information features from the input image. CSPNet solves the gradient information duplication problem in network optimization in other large-scale convolutional neural network frameworks. It integrates gradient changes from beginning to end into the feature map, thereby reducing the number of model parameters and FLOPS, ensuring inference speed and accuracy while reducing model size. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1 Schematic diagram of the structure of the shock-absorbing hammer according to an embodiment of the present invention, wherein (a) is a structural diagram of the shock-absorbing hammer, and (b) is a structural diagram of the shock-absorbing hammer installed on a conductor.

[0027] Figure 2 Flowchart of a method according to an embodiment of the present invention.

[0028] Figure 3 This is the image data of the shock-absorbing hammer area obtained in step (2) of the embodiment of the present invention.

[0029] Figure 4 Image data before and after the conductor area is obtained in step (3) of the embodiment of the present invention, wherein (a) is the original image, (b) is the image data of the marked conductor area obtained by processing with the standard Hough transform line detection method, and (c) is the image data of the marked conductor area obtained by processing with the cumulative probability Hough transform line detection method.

[0030] Figure 5 The image data of the suspension point of the shock-absorbing hammer is obtained in step (7) of the embodiment of the present invention.

[0031] Figure 6 Schematic diagram of identifying a shock-absorbing hammer in a verification experiment according to an embodiment of the present invention.

[0032] Figure 7 Schematic diagram of the displacement of the shock-absorbing hammer according to an embodiment of the present invention, wherein point a is the suspension point of the shock-absorbing hammer 1 before displacement, b1 is the suspension point position of the shock-absorbing hammer 2 before displacement, and b2 is the suspension point position of the shock-absorbing hammer 2 after displacement.

[0033] Figure 1 In the figure, 1. Wire; 2. Protective wire; 3. Shock-absorbing hammer. Implementation Method

[0034] The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more precise definition of the protection scope of the present invention.

[0035] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0036] It should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" and the like, indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, or are the orientations or positional relationships in which the inventive product is typically placed when in use. These terms are intended solely to facilitate the description of the present invention and simplify the description, and are not intended to indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third," etc., are used solely for distinction and description, and should not be construed as indicating or implying relative importance.

[0037] Furthermore, terms such as "horizontal," "vertical," and "overhanging" do not necessarily imply that a component must be absolutely horizontal or overhanging, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but rather that it can be slightly tilted.

[0038] like Figure 1 FIG2 is a structural diagram of a shock-absorbing hammer used in an embodiment of the present invention. The shock-absorbing hammer 3 is installed on the conductor 1 through the protective bar 2.

[0039] like Figure 2 As shown, a method for detecting an abnormality of a shock-absorbing hammer according to an embodiment of the present invention includes the following steps:

[0040] The front-end device collects images in real time; the front-end device adopts a visible light spherical camera.

[0041] There are two common types of noise in image data: Gaussian noise and salt and pepper noise. Gaussian noise occurs at fixed locations and has random amplitudes, while salt and pepper noise occurs at random locations and has fixed amplitudes.

[0042] Image smoothing and denoising, or filtering, can largely eliminate random noise introduced during image acquisition or portions not needed for image analysis. The research method of the embodiment of the present invention employs a combination of mean filtering and median filtering for image preprocessing. Mean filtering can be used to filter Gaussian noise, while median filtering can effectively filter salt and pepper noise. This embodiment first performs image processing and then proceeds to steps (2) and (3) to extract the anti-vibration hammer area and the wire area.

[0043] (2) Target detection is performed on the image to extract the shock-absorbing hammer area containing the shock-absorbing hammer. In step (2), the method for extracting the shock-absorbing hammer area is as follows: the area of ​​the target shock-absorbing hammer is identified by the YOLOv5 algorithm, the above area is used as the region of interest, and the contour of the shock-absorbing hammer is further extracted in the region of interest using the Sobel edge detection algorithm. The image data of the shock-absorbing hammer area is obtained as follows: Figure 3 shown.

[0044] The edge detection process of the image in the anti-vibration hammer area is as follows:

[0045] Step 2-1. Convert the color image into a grayscale image;

[0046] Step 2-2. Filter the grayscale image to remove noise;

[0047] Step 2-3. Perform Sobel processing with an attenuation factor on the grayscale image to obtain a grayscale edge map;

[0048] Step 2-4. Perform Sobel processing with an attenuation factor on the obtained grayscale edge image;

[0049] Step 2-5. Subtract the Sobel processing result from the grayscale edge map, and then change the values ​​of the edge points corresponding to the negative parts to 0 to obtain the refined edge map;

[0050] Step 2-6. Extract the geometric features of the image according to the manually designed rules, analyze the extracted features, and determine whether there is any abnormality in the target shock absorber.

[0051] (3) Extracting the wire area: Using the Hough line detection method, the wire position and the wire extension direction are marked as the wire area; In step (3), the specific method of extracting the wire area is to track the intersections between the curves corresponding to each point in the image. When the number of curves intersecting at a point exceeds a certain threshold, the parameter pairs represented by all the above intersections are obtained and marked as a straight line in the image.

[0052] The Hough transform is a feature extraction technique used in image processing. This process calculates the local maximum of the cumulative results in a parameter space, yielding a set of results that conform to a specific shape. This method is not very sensitive to incomplete or noisy data. Common Hough transforms include the Standard Hough Transform (SHT), the Multi-Scale Hough Transform (MSHT), and the Progress Probabilistic Hough Transform (PPHT).

[0053] The Hough line detection method of this embodiment adopts the standard Hough transform line detection method or the cumulative probability Hough transform line detection method. Figure 4 As shown, due to the comparison, it is found that the standard Hough transform line detection method is sufficient to identify the wires in the image. Therefore, this embodiment adopts the standard Hough transform line detection method.

[0054] (4) Set the collection time interval to 1 hour.

[0055] (5) Identify the shock-absorbing hammer position for the first time: record the shock-absorbing hammer area in the image where the shock-absorbing hammer is first detected, and determine the position of the shock-absorbing hammer hanging wire in combination with the wire position; execute step (6) according to the acquisition time interval;

[0056] (6) Identify the position of the shock-absorbing hammer again: if the shock-absorbing hammer area is not detected in the area of ​​the conductor, it is considered that the shock-absorbing hammer in the area is detached; if the shock-absorbing hammer is detected in the area of ​​the conductor, execute step (7);

[0057] (7) Detect whether the shock-absorbing hammer has shifted: find a shock-absorbing hammer hanging point and the coordinates of the shock-absorbing hammer hanging point in the area where the conductor area and the shock-absorbing hammer area overlap; obtain the shock-absorbing hammer hanging point and the coordinates of the shock-absorbing hammer hanging point in the next time interval, and perform a differential operation on the coordinates of the above two shock-absorbing hammer hanging points. If it is calculated that the shock-absorbing hammer hanging point is offset along the conductor direction, it is considered that the shock-absorbing hammer has shifted; otherwise, it is considered that there is no abnormality in the shock-absorbing hammer.

[0058] like Figure 5 As shown, in step (7), the specific method for finding a shock-absorbing hammer hanging point is to calculate the center point of the area where the shock-absorbing hammer area and the wire area overlap, and mark the center point as the shock-absorbing hammer hanging point. In step (7), the specific method for obtaining the area where the wire area and the shock-absorbing hammer area overlap is to perform a Boolean operation on the image after filling the shock-absorbing hammer outline and the wire area in the previous section to obtain the overlapping area of ​​the shock-absorbing hammer and the wire.

[0059] The verification experiment for the embodiment of the present invention is as follows:

[0060] This embodiment of the present invention reduces the detection of shock absorber detachment and displacement anomalies to the identification of shock absorbers. This identification is performed using a YOLOv5-based object detection algorithm. Using this method, images of the shock absorber obtained in its normal state are identified using a model trained with the algorithm and their positions marked. Subsequently, monitoring is performed every hour while maintaining the same camera angle. Each image is then used for object detection using the same model and parameters. The detection results obtained in the normal state are compared with the marked positions. If no shock absorber is detected at the same location, the shock absorber is considered to have detached. If the detected shock absorber suspension point is offset along the conductor from the initially detected shock absorber, the shock absorber in that area is considered to have shifted.

[0061] The input image is detected using the model trained based on the YOLOv5 algorithm, as shown in Figure 6.

[0062] Figure 7 is a schematic diagram of the displacement of the shock-absorbing hammer, where point a is the suspension point of the shock-absorbing hammer 1 before displacement, b1 is the suspension point position of the shock-absorbing hammer 2 before displacement, and b2 is the suspension point position of the shock-absorbing hammer 2 after displacement.

[0063] Although the embodiments of the present invention are described in conjunction with the accompanying drawings, the patent owner may make various changes or modifications within the scope of the appended claims. As long as they do not exceed the scope of protection described in the claims of the present invention, they should be within the scope of protection of the present invention.

Claims

1. A method for detecting abnormality of a shock-absorbing hammer, characterized by: The following steps are involved: (1) The front-end device collects images in real time; (2) Target detection is performed on the image to extract the shock-absorbing hammer area containing the shock-absorbing hammer. The specific method for extracting the shock-absorbing hammer area is as follows: the target shock-absorbing hammer area is identified by the YOLOv5 algorithm, and the above area is used as the region of interest. The contour of the shock-absorbing hammer is further extracted in the region of interest using the Sobel edge detection algorithm. The edge detection processing process of the shock-absorbing hammer area in the image is as follows: Step 2-1. Convert the color image into a grayscale image; Step 2-2. Filter the grayscale image to remove noise; Step 2-3. Perform Sobel processing with an attenuation factor on the grayscale image to obtain a grayscale edge map; Step 2-4. Perform Sobel processing with an attenuation factor on the obtained grayscale edge image; Step 2-5. Subtract the Sobel processing result from the grayscale edge map, and then change the values ​​of the edge points corresponding to the negative parts to 0 to obtain the refined edge map; Step 2-6. Extract the geometric features of the image based on manually designed rules, analyze the extracted features, and determine whether the target shock absorber has any abnormalities; (3) Extracting the wire area: Using the Hough line detection method, mark the wire position and wire extension direction as the wire area; (4) Set the collection time interval; (5) Identify the shock-absorbing hammer position for the first time: record the shock-absorbing hammer area in the image where the shock-absorbing hammer is first detected, and determine the position of the shock-absorbing hammer hanging wire in combination with the wire position; execute step (6) according to the acquisition time interval; (6) Identify the position of the shock-absorbing hammer again: if the shock-absorbing hammer area is not detected in the area of ​​the conductor, it is considered that the shock-absorbing hammer in the area is detached; if the shock-absorbing hammer is detected in the area of ​​the conductor, execute step (7); (7) Detect whether the shock-absorbing hammer has shifted: find a shock-absorbing hammer hanging point and the coordinates of the shock-absorbing hammer hanging point in the area where the conductor area and the shock-absorbing hammer area overlap; obtain the shock-absorbing hammer hanging point and the coordinates of the shock-absorbing hammer hanging point in the next time interval, and perform a differential operation on the coordinates of the two shock-absorbing hammer hanging points. If it is calculated that the shock-absorbing hammer hanging point is offset along the conductor direction, it is considered that the shock-absorbing hammer has shifted; otherwise, it is considered that there is no abnormality in the shock-absorbing hammer; (8) Repeat steps (6) to (7).

2. The method for detecting anomalies of a shock-absorbing hammer according to claim 1, wherein: The front-end device adopts a visible light spherical camera.

3. The method for detecting anomalies of a shock-absorbing hammer according to claim 1, wherein: In step (3), the specific method of extracting the wire area is to track the intersection points between the curves corresponding to each point in the image, and when the number of curves intersecting at a point exceeds a certain threshold, obtain the parameter pairs represented by all the above intersection points and mark them as a straight line in the image.

4. The method for detecting anomalies of a shock-absorbing hammer according to claim 1, wherein: The Hough line detection method adopts a standard Hough transform line detection method or a cumulative probability Hough transform line detection method.

5. The method for detecting anomalies of a shock-absorbing hammer according to claim 1, wherein: In step (7), a specific method for finding a suspension point of the anti-vibration hammer is to calculate the center point of the area where the anti-vibration hammer area and the conductor area overlap, and mark the center point as the suspension point of the anti-vibration hammer.

6. The method for detecting anomalies of a shock-absorbing hammer according to claim 1, wherein: In step (7), the specific method for obtaining the overlapping area between the wire area and the shock-absorbing hammer area is as follows: the image after filling the shock-absorbing hammer outline is subjected to a Boolean operation with the wire area in the previous section to obtain the overlapping area between the shock-absorbing hammer and the wire.

7. The method for detecting anomalies of a shock-absorbing hammer according to claim 1, wherein: The collection time interval is 1 hour.

Images