Fault data prediction method based on multidimensional data

Abstract

The present invention provides a fault data prediction method based on multidimensional data, which receives point acquisition information input by a management end, controls a drone to go to a corresponding wire monitoring point to acquire images based on the point acquisition information, and obtains a monitoring image corresponding to the corresponding wire monitoring point; calls a de-interference strategy to perform de-interference processing on the monitoring image to obtain a de-interference image; segments the de-interference image based on an extracted template to obtain an internal monitoring area and an external monitoring area; calls an internal monitoring strategy to perform internal abnormality judgment on the internal monitoring area, the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy, calls an external monitoring strategy to perform external abnormality judgment on the external monitoring area, the external monitoring strategy includes a redundant monitoring strategy, obtains abnormal monitoring data corresponding to the corresponding wire monitoring point and sends it to the management end.

Description

Technical Field

[0001] The present invention relates to multidimensional data technology, and in particular to a fault data prediction method based on multidimensional data. Background Art

[0002] With the continuous development of society, electricity has become the most important energy supply. Electricity is usually transmitted through power lines. If there is a problem with the power lines, it may cause the power system to collapse. Therefore, monitoring of power lines is very important.

[0003] At present, when monitoring power lines, people are usually dispatched to the corresponding monitoring points to check the corresponding power lines at regular intervals, so that targeted treatment can be carried out when power line faults occur. However, due to the complexity and diversity of power lines, there may be more than one fault problem, such as breakage, covering by attachments, etc. In this way, people are often required to go to the scene to diagnose the corresponding fault problem before targeted treatment can be carried out, so effective measures may not be taken in time to solve the problems of power lines, affecting the normal operation of the power system.

[0004] Therefore, how to automatically monitor power lines in multiple dimensions and carry out targeted processing of power lines in a timely and effective manner through monitoring data has become an urgent problem that needs to be solved today. Summary of the invention

[0005] The embodiment of the present invention provides a fault data prediction method based on multi-dimensional data, which can automatically monitor the power lines in multiple dimensions and carry out targeted processing on the power lines in a timely and effective manner through the monitoring data.

[0006] A first aspect of an embodiment of the present invention provides a fault data prediction method based on multidimensional data, comprising:

[0007] Receive the point acquisition information input by the management end, and control the drone to go to the corresponding wire monitoring point to collect images based on the point acquisition information, so as to obtain the monitoring image corresponding to the corresponding wire monitoring point;

[0008] Retrieving a de-interference strategy to perform de-interference processing on the monitoring image to obtain a de-interference image;

[0009] Segmenting the interference-removed image based on the extracted template to obtain an internal monitoring area and an external monitoring area;

[0010] The internal monitoring strategy is called to perform internal abnormality judgment on the internal monitoring area, and the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy. The external monitoring strategy is called to perform external abnormality judgment on the external monitoring area, and the external monitoring strategy includes a redundant monitoring strategy. The abnormal monitoring data corresponding to the corresponding wire monitoring point is obtained and sent to the management end.

[0011] Optionally, in a possible implementation of the first aspect, receiving point acquisition information input by a management terminal, and controlling a drone to go to a corresponding wire monitoring point to acquire an image based on the point acquisition information, to obtain a monitoring image corresponding to the corresponding wire monitoring point, includes:

[0012] Parsing the point acquisition information to obtain a sequence of wire monitoring points when the drone performs image acquisition, and a collection posture corresponding to each wire monitoring point in the wire monitoring point sequence;

[0013] The drone is controlled to go to the corresponding wire monitoring points in sequence based on the wire monitoring point sequence to collect images according to the corresponding collection postures, so as to obtain monitoring images corresponding to the corresponding wire monitoring points.

[0014] Optionally, in a possible implementation manner of the first aspect, calling a de-interference strategy to perform de-interference processing on the monitoring image to obtain a de-interference image includes:

[0015] Determine a pixel point in the monitoring image corresponding to the electric wire pixel value as an electric wire pixel point, and obtain adjacent electric wire pixel points to generate an electric wire pixel point set;

[0016] The pixel values ​​of the electric wire pixel point set are adjusted to a first preset pixel value, and the pixel values ​​of the pixel points outside the electric wire pixel point set are adjusted to a second preset pixel value, so as to obtain a de-interferenced image.

[0017] Optionally, in a possible implementation manner of the first aspect, segmenting the interference-removed image based on an extraction template to obtain an internal monitoring area and an external monitoring area includes:

[0018] Acquire a calibration area corresponding to the top of the electric pole in the interference-removed image, and retrieve the extraction template, wherein the extraction template includes a positioning area in the middle and a plurality of extraction areas around the positioning area, and the positioning area is a rectangular area;

[0019] Acquire a first center point of the calibration area and a second center point of the positioning area, locate the second center point based on the first center point, and locate the extraction template in the interference-removed image;

[0020] An image area corresponding to the extraction area in the interference-removed image is determined as an internal monitoring area, and an image area in the interference-removed image other than the internal monitoring area and the positioning area is determined as an external monitoring area.

[0021] Optionally, in a possible implementation of the first aspect, an internal monitoring strategy is called to perform internal abnormality judgment on the internal monitoring area, the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy, an external monitoring strategy is called to perform external abnormality judgment on the external monitoring area, the external monitoring strategy includes a redundant monitoring strategy, and abnormal monitoring data corresponding to the corresponding wire monitoring point is obtained and sent to the management end, including:

[0022] Acquire the wire contours in each of the internal monitoring areas, and perform internal abnormality judgment according to the internal monitoring strategy and the internal connection information of the wire contours to obtain internal abnormality data, wherein the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy;

[0023] Acquire the outline of the electric wires in the external monitoring area, and perform external abnormality judgment according to the external monitoring strategy and the external connection information of the electric wire outline to obtain external abnormality data, wherein the external monitoring strategy includes a redundant monitoring strategy;

[0024] Generate abnormal monitoring data according to the internal abnormal data and the external abnormal data and send it to the management end.

[0025] Optionally, in a possible implementation of the first aspect, a wire profile in each of the internal monitoring areas is obtained, and internal abnormality judgment is performed according to the internal monitoring strategy and internal connection information of the wire profile to obtain internal abnormality data, wherein the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy, including:

[0026] Acquire the outline of the electric wires in each of the internal monitoring areas;

[0027] When there is at least one wire contour in the internal monitoring area, the attachment judgment strategy is called to judge, and if the corresponding wire contour only intersects with the positioning area or only with the image edge of the interference-free image, the wire contour is regarded as an abnormal contour;

[0028] Determine the contour center line of each abnormal contour, extend the contour center line of each abnormal contour to obtain an extended line, use the abnormal contour where the extended line overlaps as the attachment wire contour, and output attachment data according to the attachment judgment strategy and the attachment wire contour;

[0029] When there is no wire outline in the internal monitoring area, or the extension line does not overlap with other extension lines, outputting missing data according to the missing judgment strategy;

[0030] Internal abnormal data is obtained according to the attachment data and the missing data.

[0031] Optionally, in a possible implementation manner of the first aspect, determining a contour center line of each abnormal contour includes:

[0032] Acquire contour lines on both sides of each abnormal contour perpendicular to the length direction of the extraction area as edge lines, and determine the middle point of the edge line to obtain a connection point;

[0033] The contour center line is generated according to a line connecting two of the connection points.

[0034] Optionally, in a possible implementation of the first aspect, a wire contour in the external monitoring area is obtained, and external abnormality judgment is performed according to the external monitoring strategy and external connection information of the wire contour to obtain external abnormality data, and the external monitoring strategy includes a redundant monitoring strategy, including:

[0035] Acquire the outline of the electric wire in the external monitoring area, and when the outline of the electric wire exists, call the redundant monitoring strategy to perform redundant judgment on the outline of the electric wire;

[0036] If one end of the wire contour intersects with the positioning area or the extraction area, and the other end of the wire contour intersects with the image edge of the interference-removed image, the wire contour is regarded as a redundant wire contour;

[0037] Redundant data is obtained according to the redundant electric wire contour, and external abnormal data is obtained based on the redundant data.

[0038] Optionally, in a possible implementation manner of the first aspect, the extraction template is obtained by the following steps, including:

[0039] Obtaining a standard image corresponding to each of the wire monitoring points, and extracting a standard wire contour and a support frame top contour in the standard image;

[0040] Generate a positioning area according to the top contour of the support frame, obtain the contour line of each standard wire contour, and determine the direction perpendicular to the contour line as the expansion direction;

[0041] The contour line of the standard wire contour is translated outward according to the preset extension distance and the extension direction to obtain the extraction area corresponding to each of the standard wire contours;

[0042] The extraction template is generated based on the positioning area and the extraction area.

[0043] Optionally, in a possible implementation of the first aspect, the method further includes:

[0044] receiving deletion information or addition information input by the management end to the extraction template, and calling up a transparent layer to be superimposed on the extraction template;

[0045] In response to the deletion information, receiving a first trigger trace input by the management end based on the transparent layer, and obtaining a deletion area according to the first trigger trace;

[0046] Deleting the extraction area within the deletion area to obtain an updated extraction template;

[0047] In response to the adding information, receiving a second trigger trace input by the management end based on the transparent layer, and obtaining a newly added area line according to the second trigger trace;

[0048] A new area is generated according to the new area line, and the extraction template is updated based on the new area to obtain an updated extraction template.

[0049] The beneficial effects of the present invention are as follows:

[0050] 1. The present invention can automatically monitor the power lines in multiple dimensions, and timely and effectively carry out targeted processing of the power lines through monitoring data. When monitoring the power lines, the present invention will dispatch drones to the corresponding power line monitoring points in turn to collect images, and then perform multi-dimensional fault analysis on the collected monitoring images, so that the power line data of multiple power line monitoring points can be automatically analyzed for faults. When performing fault analysis on the monitoring images, the present invention will first perform interference removal processing on the monitoring images, so as to improve the accuracy of fault analysis, and then the present invention will divide the monitoring images into internal monitoring areas and external monitoring areas by extracting templates, and then respectively perform different methods of fault analysis on the fault data in the internal monitoring areas and the external monitoring areas, so as to achieve multi-dimensional analysis of fault data, improve the accuracy of fault analysis, and when abnormal data is identified, the present invention will promptly send the abnormal data to the management end for targeted processing, so as to reduce the damage caused by the power line fault.

[0051] 2. When performing fault analysis on fault data in the internal monitoring area and the external monitoring area, the present invention will adopt an attachment monitoring strategy and / or a missing monitoring strategy to perform fault analysis on the data in the internal monitoring area, and adopt a redundant monitoring strategy to perform fault analysis on the data in the external monitoring area. This is because in actual applications, the standard wires in the internal monitoring area may be broken, missing, or covered by attachments, thereby affecting the normal operation of the standard wires, and the external monitoring area generally cannot be connected to the wires. If wires are found in the external monitoring area, the wires are likely to be illegally connected and are redundant. Therefore, when performing fault analysis on the internal monitoring area and the external monitoring area, the present invention will adopt different judgment strategies to perform fault analysis on them, so that the fault data of the wires can be automatically monitored in multiple dimensions, so that the management end can make targeted processing of abnormal data in a timely manner.

[0052] 3. When generating the extraction template, the present invention will generate the extraction area and positioning area of ​​the extraction template according to the standard wire contour in the standard image and the top contour of the support frame, so that the multiple monitoring wires of the corresponding wire monitoring point can be accurately extracted according to the obtained extraction template, thereby improving the accuracy of fault analysis. And when the standard wires of the corresponding wire monitoring point are added or reduced, the present invention will also update the extraction template according to the addition information or deletion information input by the management end, so that the extraction template can be updated in real time according to the addition and deletion of the standard wires of the wire monitoring point, thereby improving the accuracy of fault analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of the present invention;

[0054] Figure 2 It is a structural schematic diagram of a fault data prediction system based on multidimensional data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0056] See also Figure 1, is a schematic diagram of an application scenario provided by an embodiment of the present invention. The present invention will dispatch drones to the corresponding power line monitoring points in turn to collect images, and then perform multi-dimensional fault analysis on the collected images. When performing fault analysis on the images, the present invention will divide the monitoring images into internal monitoring areas and external monitoring areas by extracting templates, and then respectively analyze the fault data in the internal monitoring areas and the external monitoring areas in different ways, so as to realize automatic monitoring of the power line at the power line monitoring point, and when abnormal data is identified, it can be sent to the management end in time for targeted processing, thereby reducing the damage caused by power line faults. It is worth mentioning that, considering that some behaviors such as private wiring often occur at traditional electric poles in towns and villages, the power line monitoring points of this scheme are configured based on traditional electric poles in towns and villages. For large-scale transmission racks, their supervision is relatively strict, and it is not easy to privately wire, so this scheme is not applicable.

[0057] The execution subject of this application may include but is not limited to at least one of the following: user equipment, network equipment, etc. Among them, user equipment may include but is not limited to computers, smart phones, personal digital assistants (PDA) and the electronic devices mentioned above. Network equipment may include but is not limited to a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. It includes steps S1 to S4, as follows:

[0058] S1, receiving the point acquisition information input by the management end, and controlling the drone to go to the corresponding wire monitoring point to collect images based on the point acquisition information, so as to obtain the monitoring image corresponding to the corresponding wire monitoring point.

[0059] In practical applications, when monitoring power lines, multiple power line monitoring points can be set in advance, and then the drones can be controlled to go to the corresponding power line monitoring points to collect images of the corresponding power lines, so that the corresponding power lines can be automatically monitored in multiple dimensions through the analysis of the monitoring images in the future.

[0060] Based on the above embodiment, the specific implementation of step S1 may be:

[0061] S11, parsing the point acquisition information to obtain a sequence of wire monitoring points when the drone performs image acquisition, and a collection posture corresponding to each wire monitoring point in the wire monitoring point sequence.

[0062] It is understandable that when the drone is collecting data, it may collect data on the wire lines of multiple wire monitoring points. In order to enable the drone to go to each wire monitoring point in an orderly manner to collect data, a wire monitoring point sequence can be set up, so that the drone can go to each wire monitoring point in an orderly manner according to the wire monitoring point sequence to collect data.

[0063] In practical applications, the collection posture of the drone at each power line monitoring point can be the height and direction of the drone when collecting data. For example, the height of the drone when collecting data can be five meters from the top of the power pole, and the direction can be shooting towards the south.

[0064] S12, controlling the drone to go to the corresponding wire monitoring points in sequence based on the wire monitoring point sequence and to collect images in accordance with the corresponding collection postures, so as to obtain monitoring images corresponding to the corresponding wire monitoring points.

[0065] When a drone collects images at a power line monitoring point, the drone's shooting position can be located through the circular area on the top of a traditional utility pole. When the circular area on the top of the utility pole is located at the center of the drone's collection interface, the drone can be controlled to collect monitoring images of the corresponding power line monitoring point.

[0066] Through the above method, the drone can be controlled in an orderly manner to collect monitoring data from each power line monitoring point.

[0067] S2, calling a de-interference strategy to perform de-interference processing on the monitoring image to obtain a de-interference image.

[0068] In practical applications, there may be interference data in the collected monitoring images that may affect the subsequent fault analysis. Therefore, before performing fault analysis, this solution will first perform interference removal on the monitoring images, thereby improving the accuracy of subsequent fault analysis.

[0069] Based on the above embodiment, the specific implementation of step S2 may be:

[0070] S21, determining a pixel point corresponding to a wire pixel value in the monitoring image as a wire pixel point, and acquiring adjacent wire pixel points to generate a wire pixel point set.

[0071] The above-mentioned pixel value of the electric wire can be set accordingly in advance by the staff. For example, if the pixel value of the target electric wire is black, the above-mentioned pixel value of the electric wire can also be set to black accordingly.

[0072] S22, adjusting the pixel values ​​of the electric wire pixel point set to a first preset pixel value, and adjusting the pixel values ​​of the pixel points outside the electric wire pixel point set to a second preset pixel value, to obtain a de-interferenced image.

[0073] It is understandable that adjusting the pixel values ​​of the pixels outside the wire pixel set to the second preset pixel value can distinguish the wire from other objects, thereby eliminating the interference caused by other objects. In practical applications, the first preset pixel value can be set to black, and the second preset pixel value can be set to white.

[0074] Through the above method, the monitoring image can be de-interferenceed to improve the accuracy of subsequent fault analysis.

[0075] S3, segmenting the interference-removed image based on the extracted template to obtain an internal monitoring area and an external monitoring area.

[0076] When performing fault analysis on the de-interferenced image, this solution divides the de-interferenced image into an internal monitoring area and an external monitoring area, and then performs fault analysis on the internal monitoring area and the external monitoring area in different ways respectively.

[0077] Specifically, step S3 can be implemented through steps S31 to S33, which are specifically as follows:

[0078] S31, obtaining a calibration area corresponding to the top of the electric pole in the interference-removed image, and calling the extraction template, wherein the extraction template includes a positioning area in the middle and multiple extraction areas around it, and the positioning area is a rectangular area.

[0079] The calibration area mentioned above refers to the area where the drone is positioned during data collection. The calibration area is obtained so that the extraction template can be positioned through the calibration area later, so that the interference-free image can be divided through the extraction template.

[0080] In some embodiments, the extraction template can be obtained by the following steps:

[0081] S311, obtaining a standard image corresponding to each of the wire monitoring points, and extracting a standard wire contour and a support frame top contour in the standard image.

[0082] The above-mentioned standard image refers to the image collected from each wire monitoring point without any interference. It can be understood that in actual applications, a support frame is usually set on the top of the utility pole to fix the wires. Therefore, after obtaining the standard image, the standard wire contour and the top contour of the support frame in the standard image can be extracted.

[0083] S312, generating a positioning area according to the top contour of the support frame, obtaining contour lines of the contours of each standard wire, and determining a direction perpendicular to the contour lines as an expansion direction.

[0084] In practical applications, the support frame is usually set at the top of the electric pole to fix the wires around the electric pole. Therefore, the electric pole is generally located at the center of the support frame. Therefore, a positioning area can be generated according to the top contour of the support frame, so that the electric pole can be positioned according to the positioning area. It is worth mentioning that the top contour of the support frame in this solution is circular.

[0085] The expansion direction may be a direction perpendicular to both sides of the contour line, and when the contour line is subsequently expanded outward, the expansion may be performed outward toward both sides of the contour line.

[0086] S313, translating the contour lines of the standard wire contours outwards according to the preset extension distance and the extension direction to obtain extraction areas corresponding to the standard wire contours.

[0087] It is understandable that the support frame may be connected to multiple wires in general. In order to monitor each wire, an extraction area corresponding to each standard wire profile may be generated, so that the conditions of the corresponding wires may be monitored based on the extraction area.

[0088] S314: Generate the extraction template based on the positioning area and the extraction area.

[0089] The extraction template generated above can be used to accurately extract multiple monitoring wires at corresponding wire monitoring points, thereby improving the accuracy of fault analysis.

[0090] In addition, based on the above embodiment, this solution also includes the following embodiments to update the extraction template:

[0091] S315, receiving deletion information or addition information input by the management end for the extraction template, and retrieving a transparent layer to be superimposed on the extraction template.

[0092] It is understandable that in actual applications, the standard wires of the corresponding wire monitoring points may be added or reduced. In this case, errors may occur in dividing the de-interference image according to the original extraction template. Therefore, the extraction template corresponding to the corresponding wire monitoring point can be updated in real time according to the deletion information or addition information input by the management end.

[0093] S316, in response to the deletion information, receiving a first trigger trace input by the management terminal based on the transparent layer, and obtaining a deletion area according to the first trigger trace.

[0094] When the deletion information is inputted at the management end, it indicates that the standard wires at the corresponding wire monitoring point may be reduced, and thus the corresponding deletion area can be obtained according to the first trigger trace inputted at the management end.

[0095] S317, deleting the extraction area within the deletion area to obtain an updated extraction template.

[0096] It can be understood that the deleted area contains reduced standard wires, so after obtaining the deleted area, the extracted area in the deleted area can be deleted to obtain an updated extraction template.

[0097] S318, in response to the adding information, receiving a second trigger trace input by the management terminal based on the transparent layer, and obtaining a newly added area line according to the second trigger trace.

[0098] When the added information is inputted at the management end, it indicates that the standard wires at the corresponding wire monitoring point may have been increased, so the newly added area line can be obtained according to the second trigger trace inputted at the management end.

[0099] S319, generating a new area according to the new area line, and updating the extraction template based on the new area to obtain an updated extraction template.

[0100] After obtaining the newly added area line, the newly added area can be obtained according to the newly added area line, thereby obtaining an updated extraction template.

[0101] Through the above method, the extraction template can be updated in real time according to the addition and deletion of standard wires at the wire monitoring point, thereby improving the accuracy of fault analysis.

[0102] S32, obtaining a first center point of the calibration area and a second center point of the positioning area, positioning the second center point based on the first center point, and positioning the extraction template in the interference-removed image.

[0103] It is understandable that, since the center positions of the positioning area and the calibration area correspond to each other, both are the center positions of the top area of ​​the electric pole, when positioning the extraction template, the extraction template can be positioned in the interference-free image according to the first center point of the calibration area and the second center point of the positioning area. It is also understandable that, since the shooting orientation is fixed, the extraction template can be accurately positioned after the center points are aligned.

[0104] S33, determining an image area corresponding to the extraction area in the interference-removed image as an internal monitoring area, and determining an image area in the interference-removed image other than the internal monitoring area and the positioning area as an external monitoring area.

[0105] It can be understood that the extraction area corresponds to the area of ​​standard wires. During monitoring, it is necessary to determine whether the wires in this area are operating normally. Therefore, when dividing, the image area corresponding to the extraction area can be used as the internal monitoring area, and the area outside the extraction area and the positioning area is not the area corresponding to the standard wires. No additional wires can be added in this area. Therefore, the image area outside the internal monitoring area and the positioning area can be used as the external monitoring area.

[0106] In the above manner, the area where the standard wires are located and the area where additional wires cannot be installed can be accurately distinguished by extracting the template, thereby improving the accuracy of fault analysis.

[0107] S4, calling the internal monitoring strategy to perform internal abnormality judgment on the internal monitoring area, the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy, calling the external monitoring strategy to perform external abnormality judgment on the external monitoring area, the external monitoring strategy includes a redundant monitoring strategy, and obtaining the abnormal monitoring data corresponding to the corresponding wire monitoring point and sending it to the management end.

[0108] In actual applications, standard wires may be broken, missing, or covered by attachments. Therefore, when performing fault analysis on the internal monitoring area, this solution will adopt an attachment monitoring strategy and / or a missing monitoring strategy to perform corresponding analysis on the monitoring data of the internal monitoring area. There may be redundant illegal wires in the external monitoring area. Therefore, when performing fault analysis on the external monitoring area, this solution will adopt a redundant monitoring strategy to perform corresponding analysis on the monitoring data in the external monitoring area.

[0109] Based on the above embodiment, the specific implementation of step S4 may be:

[0110] S41, obtaining the wire contours in each of the internal monitoring areas, performing internal abnormality judgment according to the internal monitoring strategy and the internal connection information of the wire contours to obtain internal abnormality data, wherein the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy.

[0111] Specifically, step S41 can be implemented through steps S411 to S413, which are specifically as follows:

[0112] S411, obtaining the outline of the electric wires in each of the internal monitoring areas.

[0113] S412, when there is at least one wire contour in the internal monitoring area, the attachment judgment strategy is called to make a judgment, and if the corresponding wire contour only intersects with the positioning area or only intersects with the image edge of the interference-free image, the wire contour is regarded as an abnormal contour.

[0114] It is understandable that if there are attachments on the wires, when the wire contour of the standard wire is extracted, the wire contour of the standard wire may be separated by the attachments, thereby becoming multiple wire contours. Therefore, when there is at least one wire contour in the internal monitoring area, the attachment judgment strategy can be called to perform attachment judgment on the corresponding internal monitoring area.

[0115] If the wire contour only intersects with the positioning area or only with the image edge of the interference-free image, it means that the corresponding wire contour is likely to be separated by attachments. Therefore, it can be regarded as an abnormal contour, so that further abnormality judgment can be made on it later.

[0116] S413, determining the contour center line of each of the abnormal contours, extending the contour center line of each of the abnormal contours to obtain an extended line, taking the abnormal contours where the extended lines overlap as the attachment wire contour, and outputting the attachment data according to the attachment judgment strategy and the attachment wire contour.

[0117] It is understandable that if the center lines of the abnormal contours overlap after extension, they are likely to be the same wire contour but separated by attachments. Therefore, the corresponding abnormal contour can be used as the attachment wire contour, and the attachment data can be output based on the attachment judgment strategy and the attachment wire contour.

[0118] It is worth mentioning that the above scheme is aimed at making overlapping judgments on the extended contour lines of the electric wire contours in the same internal monitoring area. This scheme will not make overlapping judgments on the extended contour lines in different internal monitoring areas.

[0119] In some embodiments, the contour centerline can be obtained by the following steps:

[0120] The contour lines of both sides of each abnormal contour perpendicular to the length direction of the extraction area are obtained as edge lines, and the middle points of the edge lines are determined to obtain connection points.

[0121] In practical applications, when obtaining the connection point, the number of pixels corresponding to the edge line can be obtained, and then the position point at half the number of pixels in the direction from the upper point to the lower point of the edge line is determined as the connection point.

[0122] The contour center line is generated according to a line connecting two of the connection points.

[0123] S414: when there is no wire outline in the internal monitoring area, or when the extension line does not overlap with other extension lines, output missing data according to the missing judgment strategy.

[0124] S415: Obtain internal abnormal data according to the attachment data and the missing data.

[0125] It is understandable that if there is no wire outline in the internal monitoring area, or the extension line does not overlap with other extension lines, it means that the wire in the internal monitoring area is likely to be broken. Therefore, in this case, missing data can be output according to the missing judgment strategy.

[0126] Through the above method, internal abnormal data can be monitored, so that the management end can be reminded in time when the wires are covered by attachments or missing, so that the management end can make targeted processing.

[0127] S42, obtaining the outline of the electric wires in the external monitoring area, performing external abnormality judgment according to the external monitoring strategy and the external connection information of the electric wire outline to obtain external abnormality data, wherein the external monitoring strategy includes a redundant monitoring strategy.

[0128] Specifically, step S42 can be implemented through steps S421 to S423, which are as follows:

[0129] S421, obtaining the outline of the electric wire in the external monitoring area, and when the outline of the electric wire exists, calling the redundant monitoring strategy to perform redundant judgment on the outline of the electric wire.

[0130] It is understandable that, generally speaking, wires cannot be connected in the external monitoring area. If there is a wire outline in the external monitoring area, it means that the wire outline is likely to be illegally connected. Therefore, a redundant monitoring strategy can be called to make redundant judgments on the wire outline.

[0131] S422: If one end of the wire contour intersects with the positioning area or the extraction area, and the other end of the wire contour intersects with the image edge of the interference-removed image, the wire contour is regarded as a redundant wire contour.

[0132] It can be understood that if one end of the wire contour intersects with the positioning area or the extraction area, and the other end of the wire contour intersects with the image edge of the de-interference image, it means that the corresponding wire contour is likely to be illegally connected from the outside, so the corresponding wire contour can be regarded as a redundant wire contour.

[0133] S423, obtaining redundant data according to the redundant electric wire contour, and obtaining external abnormal data based on the redundant data.

[0134] Through the above method, external abnormal data can be monitored, so that when there are illegally connected wires in the external monitoring area, the management end can be reminded in time, so that the management end can make targeted processing.

[0135] S43: Generate abnormal monitoring data according to the internal abnormal data and the external abnormal data and send the data to the management terminal.

[0136] Through the above method, the abnormal data in the internal monitoring area and the external monitoring area can be monitored respectively, so that the fault data of the power line can be automatically monitored in multiple dimensions, so that the management end can make targeted processing of the abnormal data in a timely manner.

[0137] See also Figure 2 , is a schematic diagram of a structure of a fault data prediction system based on multidimensional data provided by an embodiment of the present invention, the fault data prediction system based on multidimensional data includes:

[0138] The acquisition module is used to receive the point acquisition information input by the management end, and control the drone to go to the corresponding wire monitoring point to collect images based on the point acquisition information, so as to obtain the monitoring image corresponding to the corresponding wire monitoring point;

[0139] The interference removal module is used to call the interference removal strategy to perform interference removal processing on the monitoring image to obtain the interference removal image;

[0140] A segmentation module, used for segmenting the interference-removed image based on the extraction template to obtain an internal monitoring area and an external monitoring area;

[0141] The monitoring module is used to call the internal monitoring strategy to perform internal abnormality judgment on the internal monitoring area, the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy, call the external monitoring strategy to perform external abnormality judgment on the external monitoring area, the external monitoring strategy includes a redundant monitoring strategy, and obtain the abnormal monitoring data corresponding to the corresponding wire monitoring point and send it to the management end.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fault data prediction method based on multidimensional data, characterized in that: include: Receive the point acquisition information input by the management end, and control the drone to go to the corresponding wire monitoring point to collect images based on the point acquisition information, so as to obtain the monitoring image corresponding to the corresponding wire monitoring point; Retrieving a de-interference strategy to perform de-interference processing on the monitoring image to obtain a de-interference image; Segmenting the interference-removed image based on the extracted template to obtain an internal monitoring area and an external monitoring area; Retrieving an internal monitoring strategy to perform internal abnormality judgment on the internal monitoring area, wherein the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy; retrieving an external monitoring strategy to perform external abnormality judgment on the external monitoring area, wherein the external monitoring strategy includes a redundant monitoring strategy; obtaining abnormal monitoring data corresponding to the corresponding wire monitoring point and sending the data to the management terminal; The interference-free image is segmented based on the extracted template to obtain an internal monitoring area and an external monitoring area, including: Acquire a calibration area corresponding to the top of the electric pole in the interference-removed image, and retrieve an extraction template, wherein the extraction template includes a positioning area in the middle and multiple extraction areas around the positioning area, and the positioning area is a rectangular area; Acquire a first center point of the calibration area and a second center point of the positioning area, locate the second center point based on the first center point, and locate the extraction template in the interference-removed image; Determine an image area corresponding to the extraction area in the interference-removed image as an internal monitoring area, and determine an image area in the interference-removed image other than the internal monitoring area and the positioning area as an external monitoring area; Retrieving an internal monitoring strategy to perform internal abnormality judgment on the internal monitoring area, the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy, retrieving an external monitoring strategy to perform external abnormality judgment on the external monitoring area, the external monitoring strategy includes a redundant monitoring strategy, obtaining abnormal monitoring data corresponding to the corresponding wire monitoring point and sending it to the management end, including: Acquire the wire contours in each of the internal monitoring areas, and perform internal abnormality judgment according to the internal monitoring strategy and the internal connection information of the wire contours to obtain internal abnormality data, wherein the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy; Acquire the outline of the electric wires in the external monitoring area, and perform external abnormality judgment according to the external monitoring strategy and the external connection information of the electric wire outline to obtain external abnormality data, wherein the external monitoring strategy includes a redundant monitoring strategy; Generate abnormal monitoring data according to the internal abnormal data and the external abnormal data and send it to the management end; The wire contours in each of the internal monitoring areas are obtained, and internal abnormality judgment is performed according to the internal monitoring strategy and the internal connection information of the wire contours to obtain internal abnormality data, wherein the internal monitoring strategy includes an attachment monitoring strategy and / or a missing monitoring strategy, including: Acquire the outline of the electric wires in each of the internal monitoring areas; When there is at least one wire contour in the internal monitoring area, the attachment monitoring strategy is retrieved for judgment, and if the corresponding wire contour only intersects with the positioning area or only with the image edge of the interference-free image, the wire contour is regarded as an abnormal contour; Determine the contour center line of each abnormal contour, extend the contour center line of each abnormal contour to obtain an extended line, use the abnormal contour where the extended line overlaps as the attachment wire contour, and output attachment data according to the attachment monitoring strategy and the attachment wire contour; When there is no wire outline in the internal monitoring area, or the extension line does not overlap with other extension lines, outputting missing data according to the missing monitoring strategy; Obtaining internal abnormal data according to the attachment data and the missing data; The wire contour in the external monitoring area is obtained, and external abnormality judgment is performed according to the external monitoring strategy and the external connection information of the wire contour to obtain external abnormality data, wherein the external monitoring strategy includes a redundant monitoring strategy, including: Acquire the outline of the electric wire in the external monitoring area, and when the outline of the electric wire exists, call the redundant monitoring strategy to perform redundant judgment on the outline of the electric wire; If one end of the wire contour intersects with the positioning area or the extraction area, and the other end of the wire contour intersects with the image edge of the interference-removed image, the wire contour is regarded as a redundant wire contour; Redundant data is obtained according to the redundant electric wire contour, and external abnormal data is obtained based on the redundant data.

2. The method according to claim 1, characterized in that Receiving point acquisition information input by the management end, and controlling the drone to go to the corresponding wire monitoring point to collect images based on the point acquisition information, and obtaining the monitoring image corresponding to the corresponding wire monitoring point, including: Parsing the point acquisition information to obtain a sequence of wire monitoring points when the drone performs image acquisition, and a collection posture corresponding to each wire monitoring point in the wire monitoring point sequence; The drone is controlled to go to the corresponding wire monitoring points in sequence based on the wire monitoring point sequence to collect images according to the corresponding collection postures, so as to obtain monitoring images corresponding to the corresponding wire monitoring points.

3. The method according to claim 1, characterized in that Retrieving a de-interference strategy to perform de-interference processing on the monitoring image to obtain a de-interference image, including: Determine a pixel point in the monitoring image corresponding to the electric wire pixel value as an electric wire pixel point, and obtain adjacent electric wire pixel points to generate an electric wire pixel point set; The pixel values ​​of the electric wire pixel point set are adjusted to a first preset pixel value, and the pixel values ​​of the pixel points outside the electric wire pixel point set are adjusted to a second preset pixel value, so as to obtain a de-interferenced image.

4. The method according to claim 1, characterized in that: Determining the contour center line of each abnormal contour includes: Acquire contour lines on both sides of each abnormal contour perpendicular to the length direction of the extraction area as edge lines, and determine the middle point of the edge line to obtain a connection point; The contour center line is generated according to a line connecting two of the connection points.

5. The method according to claim 1, characterized in that: The extraction template is obtained by the following steps, including: Obtaining a standard image corresponding to each of the wire monitoring points, and extracting a standard wire contour and a support frame top contour in the standard image; Generate a positioning area according to the top contour of the support frame, obtain the contour line of each standard wire contour, and determine the direction perpendicular to the contour line as the expansion direction; The contour line of the standard wire contour is translated outward according to the preset extension distance and the extension direction to obtain the extraction area corresponding to each of the standard wire contours; The extraction template is generated based on the positioning area and the extraction area.

6. The method according to claim 5, characterized in that Also includes: receiving deletion information or addition information input by the management end to the extraction template, and calling up a transparent layer to be superimposed on the extraction template; In response to the deletion information, receiving a first trigger trace input by the management end based on the transparent layer, and obtaining a deletion area according to the first trigger trace; Deleting the extraction area within the deletion area to obtain an updated extraction template; In response to the adding information, receiving a second trigger trace input by the management end based on the transparent layer, and obtaining a newly added area line according to the second trigger trace; A new area is generated according to the new area line, and the extraction template is updated based on the new area to obtain an updated extraction template.

Images