AI holder intelligent monitoring and early warning system based on power transmission channel

Abstract

The invention discloses an AI holder intelligent monitoring and early warning system based on a power transmission channel, relates to the technical field of holder line monitoring, and solves the problems that only a power transmission line is monitored, and related characteristics of the power transmission line are not analyzed. Identifying the corresponding line image based on the corresponding image pixel expression features, locking the associated feature point location in the corresponding line image based on the trend of the internal point location of the corresponding line image, and locking the spatial vector based on the point where the holder is located and the specific position of the feature point location. And determining whether the corresponding line is abnormal or not by adopting a consistency check comparison mode aiming at the space vectors associated with the same line based on the previous group of monitoring periods, so that the abnormal line can be quickly and effectively locked by adopting the processing mode, and a better monitoring effect is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of PTZ line monitoring, and in particular to an AI PTZ intelligent monitoring and early warning system based on a power transmission channel. Background Art

[0002] The transmission channel refers to the physical channel composed of a series of facilities used to transmit electric energy from the power plant to the substation and then to the user. It is a key component of the power system to ensure that electric energy can be transmitted safely and efficiently. For the relevant monitoring process of the transmission channel, time intervals are generally set in the PTZ to monitor the transmission channel.

[0003] The application with publication number CN111105395A relates to the field of transmission line monitoring technology, specifically to an AI smart pan-tilt for transmission operation monitoring; it is applied to a transmission line monitoring system, the transmission line monitoring system includes a monitoring center and several AI smart pan-tilts distributed on the transmission line, the AI ​​smart pan-tilt includes an image acquisition module, an intelligent recognition module and an expert module; the intelligent recognition module analyzes the acquired image, the expert module processes multiple continuous analysis results, and updates the algorithm of the intelligent recognition module according to the processing results; this technical solution collects, analyzes and processes image information through the front end, without the need to transmit useless repeated information to the monitoring center, reducing the demand for network bandwidth and improving the response speed; at the same time, this technical solution continuously trains AI data models and algorithms through the expert system, so that its recognition accuracy can be improved with the length of use.

[0004] The intelligent monitoring process of the transmission channel is generally based on the monitoring images generated within the corresponding preset period, and the corresponding monitoring images are transmitted to the designated terminal in real time. However, this processing method does not analyze the relevant characteristics of the transmission line, and only conducts a monitoring and early warning process. This will result in an incomplete analysis of the corresponding line characteristics and a failure to achieve a more comprehensive monitoring and early warning effect. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an AI pan-tilt intelligent monitoring and early warning system based on the transmission channel, which solves the problem of simply monitoring the transmission line without analyzing the relevant characteristics of the transmission line.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an AI PTZ intelligent monitoring and early warning system based on a power transmission channel, comprising: The PTZ monitoring terminal is a monitoring PTZ installed above the transmission rack. A monitoring cycle is set inside the PTZ, and the monitoring cycle is 30 minutes. The monitoring PTZ has preset angles and focal lengths in advance, and the set angles include rotation angles and shooting angles. The relevant images are collected and transmitted to the image analysis terminal. The database stores different pixel features associated with different light intensities, wherein the pixel features are pixel features of the line and pixel features of the insulator, and the pixel features are all preset features; The image analysis end performs image analysis and confirmation on the relevant images collected by the PTZ monitoring end, locks the line image from the collected relevant images based on the pixel value characteristics of the corresponding points in the relevant images, and transmits the locked line image to the associated line analysis end. The specific method is as follows: Based on the light intensity associated with the corresponding image acquisition time, the light intensity is collected by the light sensor set in the monitoring platform, and the associated light intensity is calibrated as G i , where i represents different related images, the pixel features of the lines associated with the corresponding light intensity are extracted from the database and used as the main features, and the main features are the pixel value intervals; Then, the pixel parameters of different points in the relevant image are confirmed, and the associated points belonging to the main feature are marked as line points. Then, the line image is confirmed based on the image areas associated with several groups of line points, and the confirmed line image is transmitted to the associated line analysis terminal; The associated line analysis end processes the line image monitored by the corresponding monitoring PTZ in the current monitoring period, determines the spatial feature vector associated with the corresponding line image, and then compares the spatial feature vectors confirmed by the same line image in the previous monitoring period for consistency. Based on the consistency comparison results, it identifies whether the corresponding line has an abnormality. The specific method is as follows: Based on the determined line image, confirm the curvature direction of the line image: confirm the horizontal angles of the adjacent points of the corresponding line from the line image, confirm the line with the next point from the previous point, and then confirm the angle between the line and the horizontal plane, and calibrate the angle as the horizontal angle of the corresponding adjacent point. Based on the horizontal angles confirmed in sequence, calibrate the point with the lowest horizontal angle parameter as the feature point, use the location of the monitoring PTZ as the initial point, and use the confirmed feature point as the end point to confirm the spatial feature vector associated with the current line image; Based on the specific line associated with this line image, the spatial feature vector confirmed by this line in the previous set of monitoring cycles is identified, and whether the spatial feature vectors generated by this line in two consecutive monitoring cycles are completely consistent. If not, this specific line is marked as an abnormal line, and the marked abnormal line is directly displayed in the early warning display terminal; if the spatial feature vectors are completely consistent, no calibration is required; The insulator analysis end, based on the line image confirmed during the monitoring period, takes the line image as the central axis, locks a group of annular areas, and then determines the insulator area from this annular area based on the pixel point features of the insulator preset in the database, and then confirms the associated vector of the insulator area, identifies whether several groups of associated vectors are consistent during feature verification, and evaluates whether there is an abnormality in the insulator. The specific method is as follows: Based on the confirmed line image, the internal line is identified, and then a set of annular areas with a radius of R1 is determined with the internal line as the central axis, where R1 is a preset value, and the image associated with the annular area is calibrated as an annular area image; Confirm the pixel parameters of different points in the annular area image and lock the insulator area: based on the confirmed light intensity G i , the pixel features of the insulator corresponding to the light intensity are locked from the database, and based on the confirmed pixel parameters of different points, the points belonging to the insulator pixel features are calibrated as insulator points; Marking the image area associated with the insulator point as the insulator area; Based on the edge contour of the insulator area, the center point of the insulator area is confirmed: the edge contour is placed in a set of three-dimensional coordinate systems, and each group of contour points in the edge contour has three-dimensional coordinates, and then the confirmed groups of three-dimensional coordinates are averaged to determine the mean coordinates, and the corresponding coordinate points are locked based on the determined mean coordinates, and are synchronously calibrated in the insulator area to determine the center point of the insulator area; Based on the location of the monitoring PTZ and the confirmed center point, the location is used as the initial point and the confirmed center point is used as the end point to confirm the associated vector belonging to the insulator area; Based on the correlation vector confirmed in the corresponding insulator area in the previous monitoring cycle, the correlation vector confirmed in the previous monitoring cycle is compared with that in this monitoring cycle for consistency to identify whether the two sets of correlation vectors are completely consistent. If not, the insulator is marked as an abnormal insulator, and the marked abnormal insulator is directly displayed on the early warning display terminal; if the correlation vectors are completely consistent, no calibration is required.

[0007] Preferably, the early warning display terminal generates a line abnormality signal based on the received abnormal line and directly displays it; and generates an insulator abnormality signal based on the received abnormal insulator and displays it.

[0008] The present invention provides an AI PTZ intelligent monitoring and early warning system based on power transmission channels. Compared with the prior art, it has the following beneficial effects: The present invention analyzes the image monitored by the AI ​​PTZ, identifies the corresponding line image based on the corresponding image pixel performance characteristics, and then locks the characteristic points associated with the corresponding line image based on the trend of the internal points of the corresponding line image. Then, based on the location of the PTZ and the specific location of the characteristic points, the spatial vector is locked. Then, based on the spatial vector associated with the same line in the previous set of monitoring cycles, a consistency check comparison method is used to determine whether the corresponding line is abnormal. This processing method can quickly and effectively lock the line with abnormalities, achieving better monitoring effects. Based on the confirmed relevant lines, the specific direction of the corresponding lines is used as the internal center axis, and then the annular area is set based on the preset radius, and the annular image of the corresponding annular area is determined, and the area of ​​the insulator is confirmed for the annular image. The same method of confirming the corresponding vector is used to confirm the associated vector of the corresponding insulator, and the consistency check is performed on the associated vectors determined in the previous and subsequent cycles to identify whether there are abnormalities in the corresponding insulator and issue a warning in time. The relevant methods of intelligent monitoring and early warning processing can not only quickly lock the relevant lines of the corresponding AI gimbal, but also monitor the corresponding insulators simultaneously. Its intelligent monitoring is more comprehensive and the effect is better, which can achieve better monitoring and early warning effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a schematic diagram of the principle framework of the present invention; Figure 2 It is a schematic diagram of determining the spatial feature vector of the present invention. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0011] First embodiment See also Figure 1, the present application provides an AI PTZ intelligent monitoring and early warning system based on a power transmission channel, including a PTZ monitoring terminal, an image analysis terminal, a database, an associated line analysis terminal, an early warning display terminal, and an insulator analysis terminal, wherein the PTZ monitoring terminal and the database are electrically connected to the image analysis terminal input node, and the image analysis terminal is electrically connected to the associated line analysis terminal input node, wherein the associated line analysis terminal is electrically connected to the early warning display terminal or the insulator analysis terminal input node, respectively, wherein the database, the insulator analysis terminal, and the early warning display terminal are electrically connected from the output node to the input node in sequence; Among them, the PTZ monitoring terminal is a monitoring PTZ installed above the transmission rack, which is provided with a monitoring cycle of 30 minutes, that is, a set of monitoring processes is executed every 30 minutes, and the monitoring PTZ has preset setting angles and focal lengths in advance, and the setting angles include rotation angles and shooting angles (the shooting angles can be understood as floating angles), and there are generally four sets of setting angles, which collect relevant images from four different directions, and transmit the collected relevant images to the image analysis terminal. Specifically, the monitoring PTZ executes a set of monitoring processes in 30 minutes, and each set of monitoring processes will generate corresponding monitoring images. When acquiring the monitoring images, it is necessary to ensure that the corresponding monitoring PTZ is executed at the set angle and focal length; The database internally stores different pixel features associated with different light intensities, wherein the pixel features are the pixel features of the line and the pixel features of the insulator, and the pixel features are all preset features, which are prepared in advance by relevant operators based on experience; The image analysis end performs image analysis and confirmation on the relevant images collected by the PTZ monitoring end, locks the line image from the collected relevant images based on the pixel value characteristics of the corresponding points in the relevant images, and transmits the locked line image to the associated line analysis end. The specific method of locking the line image is as follows: Based on the light intensity associated with the corresponding image acquisition time, the light intensity is collected by the light sensor set in the monitoring platform, and the associated light intensity is calibrated as G i , where i represents different related images, and the pixel features of the lines associated with the corresponding light intensity are extracted from the database and used as the main features (the main features are the pixel value intervals); Then, the pixel parameters of different points in the relevant image are confirmed, and the associated points belonging to the main feature are marked as line points. Then, the line image is confirmed based on the image areas associated with several groups of line points, and the confirmed line image is transmitted to the associated line analysis terminal; Specifically, after the corresponding angle and focal length have been set in the gimbal, a clear image of the specified line in front will be obtained, and the obtained clear image of the line will be specifically analyzed. There are obvious difference characteristics in its pixel points, and the corresponding pixel parameter range, that is, the main feature, has been preset in the database. The line image can be locked based on the main feature and the pixel performance of the corresponding image.

[0012] Among them, the associated line analysis end processes the line image monitored by the corresponding monitoring PTZ in the current monitoring period, determines the spatial feature vector associated with the corresponding line image, and then performs consistency comparison on the spatial feature vector confirmed by the same line image in the previous set of monitoring periods. Based on the consistency comparison result, it is identified whether the corresponding line has an abnormality. The specific method of performing consistency comparison is as follows: Combination Figure 2 , based on the determined line image, confirm the curvature direction of the line image: confirm the horizontal angles of the adjacent points of the corresponding line from the line image, confirm the line with the next point from the previous point, and then confirm the angle between the line and the horizontal plane, and calibrate the angle as the horizontal angle of the corresponding adjacent point. Based on the horizontal angles confirmed in sequence, calibrate the point with the lowest horizontal angle parameter as the feature point, take the location of the monitoring pan / tilt as the initial point, take the confirmed feature point as the end point, and confirm the spatial feature vector associated with the current line image; Based on the specific line associated with this line image, identify the spatial feature vector confirmed by this line in the previous set of monitoring cycles, and identify whether the spatial feature vectors generated by this line in two consecutive monitoring cycles are completely consistent (that is, completely overlapped). If they are completely consistent, no calibration is required. If they are not completely consistent, this specific line is marked as an abnormal line, and the marked abnormal line is directly displayed in the early warning display terminal; Its early warning display terminal generates a line abnormality signal based on the received abnormal line and displays it directly, and simultaneously issues an early warning to inform relevant maintenance personnel that such lines have relevant abnormalities.

[0013] Specifically, after the image of the corresponding line is acquired, the lowest point of the corresponding line can be identified based on the specific performance status of the corresponding line, that is, based on the trend of adjacent points in the corresponding line, the specific point with the lowest angle change can be locked. Subsequently, the corresponding spatial vector is determined based on the position of the corresponding pan-tilt head and the determined specific point, so that the spatial vector of the corresponding line can be confirmed in each monitoring cycle, and the confirmed spatial vector of the same line can be checked and calibrated, and the spatial feature vector associated with the same line can be locked for consistency verification and comparison. Based on the corresponding verification and comparison results, the corresponding abnormal line can be locked.

[0014] Second embodiment In the specific implementation process, compared with the first embodiment, this embodiment mainly performs correlation monitoring on the insulators around the line, identifies whether there are related problems between the corresponding insulators, and transmits the identification processing results to the corresponding early warning display terminal; The insulator analysis end, based on the line image confirmed during the monitoring period, takes the line image as the central axis, locks a group of annular areas, determines the insulator area from the annular area based on the pixel features of the insulator preset in the database, and then confirms the associated vector of the insulator area, identifies whether several groups of associated vectors are consistent when performing feature verification, and evaluates whether the insulator has abnormalities. The specific method of evaluation is as follows: Based on the confirmed line image, the internal line is identified, and then a set of annular areas with a radius of R1 is determined with the internal line as the central axis, where R1 is a preset value, and its specific value is determined by the operator based on experience, and the image associated with this annular area is calibrated as an annular area image; Confirm the pixel parameters of different points in the annular area image and lock the insulator area: based on the confirmed light intensity G i , the pixel features of the insulator corresponding to the light intensity are locked from the database, and based on the confirmed pixel parameters of different points, the points belonging to the insulator pixel features are calibrated as insulator points; Marking the image area associated with the insulator point as the insulator area; Based on the edge contour of the insulator area, the center point of the insulator area is confirmed: the edge contour is placed in a set of three-dimensional coordinate systems, and each group of contour points in the edge contour has three-dimensional coordinates, and then the confirmed groups of three-dimensional coordinates are averaged to determine the mean coordinates, and the corresponding coordinate points are locked based on the determined mean coordinates, and are synchronously calibrated in the insulator area to determine the center point of the insulator area; Based on the location of the monitoring PTZ and the confirmed center point, the location is used as the initial point and the confirmed center point is used as the end point to confirm the associated vector belonging to the insulator area; Based on the correlation vector confirmed in the corresponding insulator area in the previous monitoring cycle, the correlation vectors confirmed in the previous monitoring cycle and the current monitoring cycle are compared for consistency to identify whether the two sets of correlation vectors are completely consistent. If they are completely consistent, no calibration is required. If they are not completely consistent, the insulator is calibrated as an abnormal insulator, and the calibrated abnormal insulator is directly displayed in the early warning display terminal; Among them, the early warning display terminal generates and displays an insulator abnormality signal based on the received abnormal insulator, and simultaneously issues an early warning to inform the corresponding maintenance personnel that such insulator has relevant abnormalities; Specifically, the corresponding insulator is confirmed based on the corresponding line ring area, and then the relevant insulator area is confirmed based on the confirmed pixel features of the relevant insulator and the pixel values ​​in the corresponding image, and the confirmed insulator area is confirmed by contour analysis, and the center point associated with the corresponding insulator area is locked. The corresponding vector is determined based on the location of the corresponding monitoring gimbal and the corresponding center point. If the corresponding insulator has corresponding anomalies in the previous and subsequent monitoring cycles, it will be displayed that the corresponding insulator has inconsistencies before and after, and monitoring and early warning are required to inform specific relevant personnel.

[0015] Third embodiment The specific implementation process of this embodiment includes the entire implementation process of the above two groups of embodiments.

[0016] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0017] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The AI ​​PTZ intelligent monitoring and early warning system based on power transmission channel is characterized by: include: The PTZ monitoring terminal is a monitoring PTZ installed above the transmission rack. A monitoring cycle is set inside the PTZ, and the monitoring cycle is 30 minutes. The monitoring PTZ has preset angles and focal lengths in advance, and the set angles include rotation angles and shooting angles. The relevant images are collected and transmitted to the image analysis terminal. The database stores different pixel features associated with different light intensities, wherein the pixel features are pixel features of the line and pixel features of the insulator, and the pixel features are all preset features; The image analysis end performs image analysis and confirmation on the relevant images collected by the PTZ monitoring end, locks the line image from the collected relevant images based on the pixel value characteristics of the corresponding points in the relevant images, and transmits the locked line image to the associated line analysis end; The associated line analysis end processes the line image monitored by the corresponding monitoring PTZ in the current monitoring cycle, determines the spatial feature vector associated with the corresponding line image, and then performs consistency comparison on the spatial feature vector confirmed in the previous monitoring cycle of the same line image. Based on the consistency comparison results, it identifies whether the corresponding line has an abnormality. The insulator analysis end, based on the line image confirmed during the monitoring period, takes the line image as the central axis, locks a group of annular areas, and then determines the insulator area within this annular area based on the pixel features of the insulator preset in the database, and then confirms the associated vector of the insulator area. Depending on whether several groups of associated vectors are consistent during feature verification, it is assessed whether there is any abnormality in the insulator.

2. The AI ​​PTZ intelligent monitoring and early warning system based on power transmission channel according to claim 1 is characterized in that: The specific method of locking the line image at the image analysis end is: Based on the light intensity associated with the corresponding image acquisition time, the light intensity is collected by the light sensor set in the monitoring platform, and the associated light intensity is calibrated as G i , where i represents different related images, the pixel features of the lines associated with the corresponding light intensity are extracted from the database and used as the main features, and the main features are the pixel value intervals; Then, the pixel parameters of different points in the relevant image are confirmed, and the associated points belonging to this main feature are marked as line points. Then, the line image is confirmed based on the image area associated with several groups of line points, and the confirmed line image is transmitted to the associated line analysis terminal.

3. The AI ​​PTZ intelligent monitoring and early warning system based on power transmission channel according to claim 2 is characterized in that: The specific method of the associated line analysis end for performing consistency comparison on the confirmed spatial feature vector is: Based on the determined line image, confirm the curvature direction of the line image: confirm the horizontal angles of the adjacent points of the corresponding line from the line image, confirm the line with the next point from the previous point, and then confirm the angle between the line and the horizontal plane, and calibrate the angle as the horizontal angle of the corresponding adjacent point. Based on the horizontal angles confirmed in sequence, calibrate the point with the lowest horizontal angle parameter as the feature point, use the location of the monitoring PTZ as the initial point, and use the confirmed feature point as the end point to confirm the spatial feature vector associated with the current line image; Based on the specific line associated with this line image, the spatial feature vector confirmed by this line in the previous set of monitoring cycles is identified, and whether the spatial feature vectors generated by this line in two consecutive monitoring cycles are completely consistent. If not, this specific line is marked as an abnormal line, and the marked abnormal line is directly displayed in the early warning display terminal.

4. The AI ​​PTZ intelligent monitoring and early warning system based on power transmission channel according to claim 3 is characterized in that: If the spatial eigenvectors are completely consistent, no calibration is required.

5. The AI ​​PTZ intelligent monitoring and early warning system based on power transmission channel according to claim 3 is characterized in that: The specific method of identifying a plurality of groups of associated vectors for feature verification at the insulator analysis end is as follows: Based on the confirmed line image, the internal line is identified, and then a set of annular areas with a radius of R1 is determined with the internal line as the central axis, where R1 is a preset value, and the image associated with the annular area is calibrated as an annular area image; Confirm the pixel parameters of different points in the annular area image and lock the insulator area: based on the confirmed light intensity G i , the pixel features of the insulator corresponding to the light intensity are locked from the database, and based on the confirmed pixel parameters of different points, the points belonging to the insulator pixel features are calibrated as insulator points; Marking the image area associated with the insulator point as the insulator area; Based on the edge contour of the insulator area, the center point of the insulator area is confirmed: the edge contour is placed in a set of three-dimensional coordinate systems, and each group of contour points in the edge contour has three-dimensional coordinates, and then the confirmed groups of three-dimensional coordinates are averaged to determine the mean coordinates, and the corresponding coordinate points are locked based on the determined mean coordinates, and are synchronously calibrated in the insulator area to determine the center point of the insulator area; Based on the location of the monitoring PTZ and the confirmed center point, the location is used as the initial point and the confirmed center point is used as the end point to confirm the associated vector belonging to the insulator area; Based on the correlation vector confirmed in the corresponding insulator area in the previous monitoring cycle, the correlation vector confirmed in the previous monitoring cycle is compared with that in this monitoring cycle for consistency to identify whether the two sets of correlation vectors are completely consistent. If not, the insulator is marked as an abnormal insulator, and the marked abnormal insulator is directly displayed on the early warning display terminal.

6. The AI ​​PTZ intelligent monitoring and early warning system based on power transmission channel according to claim 5 is characterized in that: If the correlation vectors are exactly the same, no calibration is required.

7. The AI ​​PTZ intelligent monitoring and early warning system based on power transmission channel according to claim 5 is characterized in that: The early warning display terminal generates a line abnormality signal based on the received abnormal line and directly displays it; and generates an insulator abnormality signal based on the received abnormal insulator and displays it.

Images