Transmission line ice removal method, deicing robot and monitoring management system

Abstract

The present invention provides a method for removing ice from a power transmission line, an ice-removing robot and a monitoring and management system. The method is applied to a power transmission line monitoring and management system. The system includes multiple monitoring devices, an edge server, an ice-removing robot and a control center. The multiple monitoring devices are used to perform real-time monitoring of different sections of power transmission lines in the same monitoring area. The method includes the following steps: the edge server sends each of the image monitoring data and the target monitoring device identifier to the control center; the control center performs lens cleanliness analysis based on a lens cleanliness analysis model and the monitoring image collected by the target monitoring device to obtain the lens cleanliness of the target monitoring device; if the lens cleanliness does not meet the preset requirements, the temperature control module of the target monitoring device is marked as to be repaired, thereby realizing automatic monitoring of the operating status of the temperature control module of the monitoring device and being able to detect abnormalities in time.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method for removing ice from a power transmission line, an ice removal robot and a monitoring and management system. Background Art

[0002] The maintenance of power transmission equipment is an important part of ensuring the safe and reliable operation of the power system. Realizing the intelligentization of transmission line status monitoring, strengthening the research and application of transmission line status maintenance, full life cycle management and intelligent disaster prevention technology are of great significance to improving the reliability of power grid operation and improving the operating efficiency of enterprises.

[0003] Transmission line icing is one of the major issues that affect the safe operation of power grids. Major accidents such as conductor dancing, tower collapse, line breakage and insulator flashover caused by icing seriously threaten the stability and safe operation of the power system. At present, the principles and devices for monitoring icing on transmission lines have initially formed the specifications of power companies, but there are still some problems, such as how to realize automatic evaluation of the operating status of the monitoring system to ensure the stable and reliable operation of the system. Summary of the invention

[0004] In order to realize automatic evaluation of the operating status of a transmission line monitoring system to ensure stable and reliable operation of the system, an embodiment of the present application provides a method for removing ice from transmission lines. The method is applied to a transmission line monitoring and management system. The system includes multiple monitoring devices, an edge server, an ice removal robot and a control center. The multiple monitoring devices are used to perform real-time monitoring of different sections of transmission lines in the same monitoring area. The method includes the steps of: the edge server receives image monitoring data and temperature control data sent by each of the monitoring devices, performs temperature control anomaly analysis on each of the monitoring devices based on the image monitoring data and the temperature control data; determines the monitoring device with an abnormal temperature control anomaly analysis result as a target monitoring device; sends each of the image monitoring data and the temperature control data The target monitoring device is identified to the control center; wherein the image monitoring data includes a monitoring image; the control center performs a lens cleanliness analysis based on a lens cleanliness analysis model and the monitoring image collected by the target monitoring device to obtain the lens cleanliness of the target monitoring device; if the lens cleanliness does not meet the preset requirements, the temperature control module of the target monitoring device is marked as to be repaired, a temperature control instruction and a re-collection instruction are sent to the target monitoring device, and when the lens cleanliness indicated by the monitoring image in the target image monitoring data returned by the target monitoring device meets the preset requirements, a first deicing task is generated based on the target image monitoring data, and the first deicing task is sent to the deicing robot; the deicing robot receives and executes the first deicing task.

[0005] Based on the above technical solution, the edge server first performs preliminary abnormality identification on the temperature control module of the monitoring device based on the image monitoring data and temperature control data sent by each monitoring device, and analyzes the lens cleanliness based on the monitoring image of the target monitoring device through the control center, and then determines whether the lens of the target monitoring module is covered with ice or frost according to the analysis results. If so, the temperature control command is directly sent to the target monitoring device to detect whether the temperature control module of the target monitoring device can operate normally, and then the operating status of the temperature control module of each monitoring device is monitored in real time, so that the operation and maintenance personnel can be notified in time to handle it when an abnormality occurs in the temperature control module. In addition, the first deicing task generated is generated based on the image monitoring data whose lens cleanliness meets the preset requirements, which can ensure the accuracy of the first deicing task.

[0006] In one implementation, the method also includes: if the cleanliness of the lens meets the preset requirements, the control center generates a second deicing task based on each of the image monitoring data, and sends the second deicing task to the deicing robot, wherein the second deicing task includes detecting the actual ice thickness of the target transmission line monitored by the target monitoring device and sending it to the control center; the control center compares the actual ice thickness with the ice recognition result in the image monitoring data corresponding to the target monitoring device; if the comparison result indicates that the ice recognition result of the target monitoring device does not meet the preset requirements, the ice recognition model of the target monitoring device is marked as to be updated; if the comparison result indicates that the ice recognition result of the target monitoring device meets the preset requirements, the abnormal analysis program of the edge server is marked as to be updated.

[0007] Based on the above technical solution, the actual ice thickness can be obtained with the help of the de-icing robot when performing de-icing tasks, and then the recognition accuracy of the ice recognition model of the target monitoring equipment can be evaluated, and then the ice recognition model of the target monitoring equipment and the abnormal analysis program of the edge server can be fault located. In this way, a comprehensive analysis of the accuracy of each recognition technology on the monitoring equipment and the edge server can be achieved, and the deficiencies of the program algorithm can be discovered in time, thereby providing an update direction for software upgrades.

[0008] In one implementation, the temperature control anomaly analysis of each monitoring device based on each image monitoring data and each temperature control data includes: sorting the ice cover thickness in each image monitoring data to determine the first ranking of each monitoring device; sorting each temperature control data to determine the second ranking of each monitoring device; determining the monitoring device whose difference between the first ranking and the second ranking exceeds a preset ranking threshold as a candidate device; and determining whether the candidate device has temperature control anomaly based on the thickness change trend in the image monitoring data sent by the candidate device.

[0009] In one implementation, determining whether the candidate device has abnormal temperature control according to the thickness change trend in the image monitoring data sent by the candidate device includes: determining the climate change situation within a period according to the thickness change trend, when the thickness change trend indicates that the climate continues to decline or drops suddenly within the period, if the ranking difference of the candidate device indicates that the number of heating times or the total duration is insufficient, it is determined that the temperature control of the candidate device is abnormal; if the ranking difference of the candidate device indicates that the number of heating times or the total duration is too much, it is determined that the temperature control of the candidate device is normal; when the thickness change trend indicates that the climate continues to rise or warms overall within the period, if the ranking difference of the candidate device indicates that the number of heating times or the total duration is insufficient, it is determined that the temperature control of the candidate device is normal; if the ranking difference of the candidate device indicates that the number of heating times or the total duration is too much, it is determined that the temperature control of the candidate device is abnormal; when the thickness change trend indicates that the climate is relatively stable within the period, it is determined that the temperature control of the candidate device is abnormal.

[0010] Based on the above technical solution, the edge server can directly perform preliminary temperature control anomaly analysis on each monitoring device based on the received image monitoring data and temperature control data, and can accurately locate monitoring devices with obvious abnormalities. It has small calculation amount, fast response, and low hardware requirements for the equipment, thereby saving equipment investment costs.

[0011] In one embodiment, the method further includes, when the lens cleanliness indicated by the monitoring image in the target image monitoring data returned by the target monitoring device does not meet the preset requirements, determining whether the target temperature control data is consistent with the temperature control instruction; if so, determining that the heating element in the temperature control module is faulty; otherwise, determining that the control element in the temperature control module is faulty.

[0012] Based on the above technical solution, it is possible to further locate the faults of various components in the temperature control module, thereby providing more specific maintenance directions for operation and maintenance personnel, saving manpower input, and assisting in the rapid recovery of faults.

[0013] An embodiment of the present application also provides a deicing robot, which includes a communication module and a control module, wherein the control module receives a second deicing task sent by a control center through the communication module, and before executing the second deicing task, obtains the actual ice thickness of the transmission line and sends it to the control center, wherein the second deicing task is generated by the above method.

[0014] In one embodiment, the de-icing robot also includes a clamping mechanism and a pressure sensor, wherein a method for obtaining an actual ice thickness of the transmission line includes the clamping mechanism determining whether it is clamped to the transmission line based on pressure data between the clamping mechanism and the transmission line detected by the pressure sensor, and when the clamping is confirmed, determining the ice thickness of the transmission line based on control data.

[0015] Based on the same inventive concept, the present application also provides a power transmission line monitoring and management system, which is used to implement the above method.

[0016] In addition, an embodiment of the present application also provides an electronic device, which includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the method steps performed by the edge server or the control center in the above method.

[0017] The embodiment of the present application further provides a computer-readable storage medium storing a computer program, wherein the computer program implements the above method when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of the structure of a power transmission line monitoring and management system provided in an embodiment of the present application is shown.

[0019] Figure 2 A schematic diagram of the structure of a monitoring device provided in an embodiment of the present application is shown.

[0020] Figure 3 A flow chart of a lens foreign body identification method provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0022] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0023] Please refer to Figure 1 The power transmission line monitoring and management system provided in the embodiment of the present application includes a plurality of monitoring devices 11, an edge server 12 and a control center 13. The edge server 12 is respectively connected to each monitoring device 11 and the control center for communication. Each monitoring device 11 and the edge server 12 are installed in the same monitoring area.

[0024] The monitoring equipment is used to collect image monitoring data of the transmission line. It can be installed on the transmission line or on the line tower. In one example, please refer to Figure 2 The monitoring device 11 includes a control module 111, an image acquisition module 112, a data analysis module 113 and a temperature control module 114, wherein the control module 111 is respectively connected to the other modules for communication and managing the operation status of the other modules.

[0025] The image acquisition module 112 includes a camera for photographing the transmission line to obtain monitoring images, wherein the monitoring images can be in picture format or video format, and the image acquisition method can be online real-time monitoring or snapshot at the snapshot frequency issued by the system.

[0026] The image acquisition module 112 sends the acquired monitoring image to the data analysis module 113, so that the data analysis module 113 can determine the ice thickness of the transmission line based on the recognition of the monitoring image. In one example, the data analysis module 113 can recognize the monitoring image based on an ice recognition model to determine the ice thickness of the transmission line, wherein the ice recognition model can be implemented based on a support vector machine (SVM), a random forest, a deep learning model, etc., and the present application is not limited thereto.

[0027] The data analysis module 113 periodically sends image monitoring data to the edge server, wherein the image monitoring data includes monitoring images used to identify the ice thickness of the transmission line and ice thickness data obtained by identification. In implementation, the data analysis module 113 may identify the monitoring images based on the ice thickness identification model to determine the ice thickness. The ice thickness identification model may be a pre-trained AI model based on a deep learning framework, such as ResNet (deep residual network) or VGG16.

[0028] In one example, a method for generating an ice thickness recognition model may include introducing image preprocessing algorithms, such as bilateral filtering and adaptive histogram equalization, to improve image quality; using semi-supervised learning to expand the labeled data set to cover various ice conditions from mild to severe; constructing ResNet to effectively alleviate the gradient vanishing problem by adding skip connections and improve the learning efficiency of deep structures; dividing the data set and controlling the proportional distribution to ensure that each subset is representative and independent of each other, providing effective support for model training, parameter adjustment and evaluation; selecting a cross-entropy loss function with L2 regularization to guide the model training process, combining the Dropout strategy to suppress the risk of overfitting, and periodically saving the optimal weight configuration.

[0029] The temperature control module 114 can be used to heat the key components of the monitoring device when the temperature is too low to prevent the key components from being unstable or having a shortened service life due to the low temperature. The key components include but are not limited to electronic components such as chips, capacitors, resistors, batteries, lenses, etc. In the embodiment of the present application, the temperature control module 114 can also heat the outer lens of the lens to remove foreign matter such as ice, snow, and frost attached to the lens.

[0030] In one implementation, the temperature control module 114 includes a control element and a heating element, wherein the heating element is a ring structure and is installed around the lens. The control element can perform lens foreign body recognition based on the real-time monitoring image captured by the image acquisition module 112 to determine whether foreign bodies such as ice, snow, and frost are attached to the lens. When the recognition result indicates that ice, snow, or frost is attached to the lens, a control instruction is sent to the heating element to instruct the heating element to heat the lens, thereby removing the ice, snow, or frost attached to the lens.

[0031] In an example, see Figure 3 The method for the control element to identify foreign matter on the lens based on the monitoring image specifically includes the following steps:

[0032] S301, obtaining historical monitoring images collected within a preset time period before the monitoring screen.

[0033] In implementation, the historical monitoring images collected within a preset time before the current monitoring image collection moment can be read from the local storage. The historical monitoring images can be videos or multiple snapshots. The preset time can be an empirical value configured by the system.

[0034] S302: Create a video to be identified based on the monitoring screen and historical monitoring images.

[0035] In implementation, the monitoring screen is a picture, and the historical monitoring image may be a video or multiple pictures. When the historical monitoring image is a video, key frames can be added to the video based on the monitoring screen to create a video to be identified; when the historical monitoring image is multiple pictures, the video to be identified can be generated based on the acquisition time of each picture in the order of the acquisition time.

[0036] S303, calculating the video to be identified based on the foreign object identification program to determine whether the lens is covered with foreign objects and the type of foreign objects.

[0037] Among them, the foreign object recognition program can first identify all foreign objects covering the lens that appear in the video to be recognized, and determine whether the foreign object is ice or snow based on the changing state of the foreign object. For example, if the state of the foreign object changes over time and the state change is continuous, it is determined to be ice or snow or frost.

[0038] Based on this, the control element can identify whether there is ice, snow or frost on the monitoring screen based on a lightweight foreign object recognition program, and thus trigger the control of the heating element according to the recognition result, and promptly deal with the situation where the lens is covered with ice, snow and frost, thereby avoiding the problem of not being able to obtain clear monitoring images due to ice, snow or frost.

[0039] Multiple monitoring devices are installed in the same monitoring area, wherein the area to which the monitoring devices belong is determined by a control center based on the climate similarity and geographical location of each monitoring point. In one example, the method by which the control center determines the area to which the monitoring devices belong includes: dividing the monitoring points located in the same terrain area into the same monitoring area, for example, dividing the monitoring points installed in the same mountain forest area into the same monitoring area.

[0040] An edge server is deployed in an area. The edge server can be implemented based on an independent service device or run on a monitoring device in the area in the form of an agent program.

[0041] The control center is deployed in the cloud and can be used to obtain the icing condition of the transmission line based on the image monitoring data collected by each monitoring device, and determine whether the transmission line needs to be de-iced according to the degree of icing, and then generate a de-icing task; it can also be used to manage each monitoring device and edge server to ensure that each monitoring device and edge server can operate stably and reliably; in addition, the control center can also provide transmission line monitoring data to show users the real-time status of the transmission line.

[0042] Based on the above system, an embodiment of the present application provides a method for removing ice from a power transmission line, and the method specifically includes the following steps.

[0043] S401, the edge server receives image monitoring data and temperature control data sent by each monitoring device.

[0044] During implementation, each monitoring device sends image monitoring data and temperature control data to the edge server based on the data reporting cycle set by the system.

[0045] Among them, the image monitoring data includes the monitoring images collected within the period, and the ice cover recognition results obtained by the monitoring equipment based on the monitoring images. The ice cover recognition results include but are not limited to the ice cover thickness and thickness change trend.

[0046] In one example, the monitoring image is a monitoring video collected within a period, the ice thickness can be identified based on the last key frame in the monitoring video, and the thickness change trend can be identified based on the key frame images at different time points in the monitoring video.

[0047] In another example, the monitoring image may also be a monitoring screen captured at a set frequency within a period. Accordingly, the ice thickness may be identified based on the latest captured screen, and the thickness variation trend may be identified based on each captured screen.

[0048] The temperature control data is generated by the temperature control module on the monitoring device and is used to record the control parameters sent by the control element to the heating element during the cycle, including but not limited to the heating time.

[0049] S402: Perform temperature control anomaly analysis on each monitoring device based on each image monitoring data and each temperature control data.

[0050] S403, determining the monitoring device with abnormal temperature control analysis result as a target monitoring device.

[0051] In one example, a method for performing temperature control anomaly analysis on each monitoring device based on each image monitoring data and each temperature control data specifically includes the following steps.

[0052] S501, sorting the ice thickness in each image monitoring data to determine the first ranking of each monitoring device.

[0053] In practice, the thicker the ice cover, the higher the corresponding ambient temperature and humidity. It is understandable that the monitoring points of the high-voltage line are far apart. Therefore, even if they are located in the same monitoring area, the climate will still differ due to factors such as the terrain, altitude, and surrounding environment. These differences may cause the ice cover thickness of the transmission lines corresponding to each monitoring point to be different. Based on this, the difference in ice cover thickness corresponding to each monitoring point can be used to characterize meteorological differences. Correspondingly, the conditions in which the lens of the corresponding monitoring equipment is covered with ice and snow or frosted will also be different. It is understandable that the climate environment in which the monitoring equipment and the transmission line are located at the same monitoring point is basically the same. Therefore, there is a certain correlation between the ice cover thickness of the transmission line and the condition of the lens being covered with ice and snow or frosted. For example, the thicker the ice cover, the greater the probability that the lens is covered with ice and snow or frosted.

[0054] S502, sorting each temperature control data to determine the second ranking of each monitoring device.

[0055] In one example, when the heating temperature is fixed, the heating data is the total heating time or the total number of heating times. Based on this, sorting the temperature control data includes sorting the various heating times or the total number of heating times, wherein the sorting method is consistent with the sorting method of the ice thickness, for example, both are sorted in ascending or descending order.

[0056] S503: Determine as a candidate device a monitoring device whose difference between the first ranking and the second ranking exceeds a preset ranking threshold.

[0057] It is understandable that, as mentioned above, the thicker the ice coating, the easier it is for the lens to be covered with ice or snow or frost, so the total number of heating times or total duration of the temperature control module will be longer. The corresponding first ranking and second ranking are consistent under ideal circumstances. When the rankings are inconsistent, it indicates to a certain extent that the temperature control module of the monitoring device may be abnormal.

[0058] The preset ranking threshold is used to provide fault tolerance, that is, when the ranking difference is less than or equal to the preset ranking threshold, it can be considered normal, and when the ranking difference is greater than the preset ranking threshold, it indicates that the temperature control module is more likely to be abnormal. Among them, the preset ranking threshold can be based on the experience value set by the user, or it can be updated and set by the system based on historical operation data.

[0059] S504: Determine whether the candidate device has abnormal temperature control according to the thickness change trend in the image monitoring data sent by the candidate device.

[0060] In implementation, the climate change situation within the period can be determined based on the thickness change trend. When the thickness change trend indicates that the climate continues to decline or drops suddenly within the period, if the ranking difference of the candidate device indicates that the number of heating times or the total duration is insufficient, the temperature control of the candidate device is determined to be abnormal; if the ranking difference of the candidate device indicates that the number of heating times or the total duration is too much, the temperature control of the candidate device is determined to be normal. When the thickness change trend indicates that the climate continues to rise or warms overall within the period, if the ranking difference of the candidate device indicates that the number of heating times or the total duration is insufficient, the temperature control of the candidate device is determined to be normal; if the ranking difference of the candidate device indicates that the number of heating times or the total duration is too much, the temperature control of the candidate device is determined to be abnormal. When the thickness change trend indicates that the climate is relatively stable within the period, the temperature control of the candidate device is determined to be abnormal.

[0061] In one example, a change curve graph can be constructed according to the thickness change trend, and the curve graph obtained by horizontally flipping the change curve graph is used as a climate change curve graph, and the climate change situation within the period can be determined based on the analysis of the climate change curve graph.

[0062] The method for determining whether the number of heating times or the total duration is insufficient or excessive based on the ranking difference of the candidate devices includes first determining the ranking difference, and then determining the number of heating times and the total duration based on the sorting method. For example, when sorting in ascending order, if the ranking difference is a positive number, it is determined that the number of heating times and the total duration are excessive, otherwise they are insufficient; and in the case of descending order, the opposite is true.

[0063] S505: Determine the candidate device with abnormal temperature control as the target monitoring device.

[0064] Based on the above technical solution, the anomaly analysis program on the edge server can directly perform anomaly analysis based on each image monitoring data and determine the target monitoring device. The data processing volume is small and rapid identification of anomalies can be achieved.

[0065] S404, sending each image monitoring data and the target monitoring device identification to the control center.

[0066] The target monitoring device identification is unique information used by the system to identify the identity of the target monitoring device, such as a device ID.

[0067] S405: The control center performs lens cleanliness analysis based on the lens cleanliness analysis model and the monitoring image collected by the target monitoring device to obtain the lens cleanliness of the target monitoring device.

[0068] In one implementation, performing lens cleanliness analysis based on a lens cleanliness analysis model and a monitoring image collected by a target monitoring device includes:

[0069] Obtain the target monitoring video collected by the target monitoring device within the target monitoring period.

[0070] Among them, the target monitoring cycle is the collection cycle corresponding to the generation of image monitoring data by the target monitoring device; the control center can directly request all monitoring images collected within the target monitoring cycle from the target monitoring device, and then obtain the target monitoring video.

[0071] In one example, the target monitoring video may be directly acquired by the monitoring device, or may be generated by editing a plurality of monitoring images captured by the target monitoring device in the order of the capturing time.

[0072] The target monitoring video and the monitoring image in the image monitoring data sent by the target monitoring device are used as the input layer to calculate the lens cleanliness analysis model.

[0073] The lens cleanliness analysis model is used to analyze the cleanliness of the lens, and the output layer includes but is not limited to the cleanliness score and the type of foreign matter. In one example, the lens cleanliness analysis model includes a deep learning network model built based on a convolutional neural network (CNN) and a long short-term memory (LSTM), in which CNN performs feature extraction and the LSTM network captures time series features, and finally classifies ice, snow, frost and other coverings through a fully connected layer, and when ice, snow or frost is identified, the coverage area of ​​ice, snow or frost on the monitoring image is calculated, and then the cleanliness score is determined based on the area ratio, wherein the larger the coverage area, the worse the cleanliness and the lower the score. The calculation of the coverage area can be achieved based on image segmentation technology.

[0074] S406, determining whether the cleanliness of the lens meets the preset requirements.

[0075] In one implementation, the system is configured with a minimum threshold for lens cleanliness, which is used to screen out situations where the cleanliness is too low and affects the identification of ice thickness. When the foreign matter type output by the model is ice, snow or frost, the control center can further determine the relationship between the lens cleanliness output by the model and the minimum threshold, and determine whether the lens cleanliness meets the preset requirements. When the lens cleanliness is greater than the minimum threshold, it is determined to meet the preset requirements, otherwise, it does not meet the preset requirements.

[0076] In one example, the minimum threshold of lens cleanliness can be determined according to the clarity requirement of the ice thickness recognition model on the monitoring device for the image to be recognized. Specifically, it can be obtained by testing the recognition accuracy of the model using test samples of different lens cleanliness.

[0077] If the lens cleanliness does not meet the preset requirement, step S407 is executed.

[0078] S407, marking the temperature control module of the target monitoring device as to be repaired, and sending a temperature control instruction and a re-collection instruction to the target monitoring device.

[0079] Among them, the temperature control instruction is used to instruct the temperature control module of the target monitoring device to heat the lens to remove the ice, snow or frost attached to the lens. After receiving the temperature control instruction, the target monitoring device controls the component to obtain the latest monitoring image, and identifies foreign objects on the lens, and determines whether it is necessary to heat the lens to remove the attached ice, snow or frost based on the identification result.

[0080] The re-collection instruction is used to instruct the target monitoring device to re-collect the target image monitoring data after removing the ice or frost attached to the lens, and send the target image monitoring data and target temperature control data to the control center. The target temperature control data is the control parameters generated by the temperature control module when executing the temperature control instruction, including but not limited to the heating temperature and heating time.

[0081] S408, determining whether the cleanliness of the lens indicated by the target image monitoring data meets a preset requirement.

[0082] In this step, the method of determining whether the cleanliness of the lens indicated by the monitoring image in the target image monitoring data meets the preset requirements is the same as that of step S406, and will not be repeated herein.

[0083] If the preset requirements are met, execute step S409; otherwise, execute step S410.

[0084] S409: Generate a first deicing task based on the target image monitoring data.

[0085] Specifically, when the control center determines that the lens cleanliness indicated by the monitoring image in the target image monitoring data meets the preset requirements, it can directly generate the first de-icing task based on the ice recognition results in the target image monitoring data and the image monitoring data of other monitoring devices.

[0086] The control center can plan deicing routes and deicing parameters based on the ice thickness and thickness change trends in each ice recognition result. In one implementation, the thickness adjustment coefficient can be determined based on each thickness change trend, and each ice thickness can be calculated based on each thickness adjustment coefficient to obtain an ice thickness prediction value, and the ice thickness prediction value is determined to determine the risk level of each transmission line, wherein the risk level may include emergency, medium and general, and corresponding deicing tasks are generated for each transmission line with an emergency risk level, and the deicing route is planned with the transmission line with an emergency risk level as the starting point, so that each deicing route can cover all transmission lines that need to be deiced. The mapping relationship between the ice thickness prediction value and the risk level can be pre-configured by the user or determined by the system based on historical data.

[0087] In one example, if the thickness change trend is increasing, the thickness adjustment coefficient is determined according to the maximum increase recorded in the thickness change trend. For example, the thickness change data recorded in the thickness change trend of the monitoring device A is: 10cm, 12cm, 16cm, then the thickness adjustment coefficient is determined to be 1.3, and the calculated ice thickness prediction value is 20.8cm; if the thickness change trend is relatively stable, that is, the ice thickness does not change, or the change range is lower than the preset standard, such as: 10cm, 10.5cm, 10cm, then the thickness adjustment coefficient is determined to be 1; if the thickness change trend is decreasing, then the thickness adjustment coefficient is determined according to the minimum decrease recorded in the thickness change trend. For example, the thickness change data recorded in the thickness change trend of the monitoring device B is: 15cm, 13cm, 10cm, then the thickness adjustment coefficient is determined to be 0.87, and the calculated ice thickness prediction value is 8.7cm.

[0088] When it is determined to generate multiple first deicing tasks, a deicing route can be generated based on the transmission line with an emergency risk level but the smallest predicted value. The process of generating the deicing route includes taking the transmission line as the starting point, determining the transmission line connected to the same line tower as the transmission line, with a non-emergency risk level and the largest predicted value as the next node, and so on, until the deicing task amount reaches the allocated value, thereby obtaining the deicing transmission line and deicing route corresponding to the first deicing task. Then, among the remaining transmission lines, the corresponding deicing transmission lines and deicing routes are determined for the corresponding first deicing tasks in turn based on the same method.

[0089] In one example, the control center can estimate the corresponding deicing time based on the ice thickness value of each transmission line, and then obtain the total time required to complete the deicing tasks of all transmission lines, and use the average value of the total time as the allocated value. It is worth noting that the first deicing task amount in the above method reaches the allocated value, which means that the total deicing time required for the selected transmission line is greater than or equal to the allocated value.

[0090] S410: locating a fault of a temperature control module of a target monitoring device based on target temperature control data.

[0091] In one implementation, the control center can locate the fault by determining whether the target temperature control data matches the temperature control instruction. If so, it is determined that the heating element in the temperature control module is faulty; if not, it is determined that the control element in the temperature control module is faulty.

[0092] Among them, judging whether the target temperature control data is consistent with the temperature control instruction includes: determining whether the target temperature control data carries control parameters and whether the control parameter values ​​are normal. In other words, determining whether the target temperature control data indicates that the control element has sent the correct control parameters to the heating element. If so, it is determined that the target temperature control parameters are consistent with the temperature control instruction, otherwise they are not consistent.

[0093] The control center can record the fault location results and send them to the operation and maintenance personnel.

[0094] Based on the above technical solution, each monitoring device is preliminarily identified for abnormalities through the edge server, and the identification results are synchronized to the control center. The control center then performs lens cleanliness analysis based on the monitoring images collected by the target monitoring device to determine whether the temperature control module of the target monitoring device is working properly. Based on this, real-time automatic monitoring of the operating status of the temperature control module of the monitoring device can be achieved to ensure the accuracy of the image data, thereby ensuring the stable and reliable operation of the system. At the same time, after determining that the temperature control module is abnormal, the first de-icing task is generated based on the re-acquired target image monitoring data that meets the requirements, thereby ensuring the accuracy of the de-icing task.

[0095] The above system also includes a deicing robot for executing the deicing task issued by the control center. In other embodiments of the present application, after executing step S406, if it is determined that the cleanliness of the lens meets the preset requirements, the following steps are also included.

[0096] S601: The control center generates a second deicing task based on each image monitoring data.

[0097] S602: Issue a second deicing task to the deicing robot.

[0098] Among them, the method for generating the second deicing task is basically the same as the method for generating the first deicing task in step S409, the difference lies in the different data basis, and the second deicing task generated in this step further includes detecting the actual ice thickness of the target transmission line monitored by the target monitoring equipment and sending it to the control center.

[0099] Specifically, after receiving the second de-icing task, the de-icing robot can first reach the starting transmission line with the help of a drone or an operator, that is, the transmission line with the thickest ice coverage among the transmission lines included in the second de-icing task, and before the de-icing operation starts, first detect the actual ice thickness of the transmission line to adaptively adjust the received de-icing operation parameters, and then start the de-icing operation. Based on this, when the de-icing robot performs de-icing operations on the target transmission line, the actual ice thickness of the target transmission line can be recorded and sent to the control center.

[0100] S603, receiving the actual ice thickness returned by the de-icing robot, and comparing the actual ice thickness with the ice recognition result recorded in the image monitoring data corresponding to the target monitoring device.

[0101] In one implementation, the time difference between the collection time of the actual ice thickness and the collection time of the ice thickness in the ice identification result can be determined first. If the time difference is less than a preset time length, the difference between the actual ice thickness and the ice thickness in the ice identification result is calculated. If the difference is within a preset error range, it is determined that it meets the preset requirements; otherwise, it is determined that it does not meet the preset requirements; if the time difference is greater than or equal to the preset time length, then based on the thickness change trend and combined with the meteorological data of the corresponding time period, it is evaluated whether the actual ice thickness falls within the predicted thickness range. If so, it is determined that it meets the preset requirements; otherwise, it is determined that it does not meet the preset requirements.

[0102] In one example, obtaining a predicted thickness range based on the thickness variation trend and in combination with meteorological data of a corresponding period includes predicting the ice variation range of meteorological data of a corresponding period based on an ice variation prediction model, and calculating the ice thickness in the ice recognition result based on the ice variation value to obtain the predicted thickness range. The ice variation prediction model is constructed based on a deep learning algorithm or a machine learning algorithm and is used to predict the ice thickness variation range under different climate conditions.

[0103] If the comparison result indicates that the ice recognition result of the target monitoring device does not meet the preset requirements, step S604 is executed; if it meets the preset requirements, step S605 is executed.

[0104] S604: Mark the ice recognition model of the target monitoring device as to be updated.

[0105] S605: Mark the exception analysis program of the edge server as to be updated.

[0106] Specifically, when the abnormality analysis program of the edge server performs temperature control abnormality analysis on the monitoring device, the data basis is the ice recognition result and the temperature control data. When it is determined in step S406 that the lens cleanliness meets the preset requirements, it indicates that the temperature control module can operate normally; therefore, if the ice recognition result of the monitoring device is also accurate, it indicates that the abnormality analysis program needs to be optimized.

[0107] Based on this, the de-icing robot can be used to obtain the actual ice thickness of the target transmission line during the de-icing task, so as to verify the ice recognition model of the target monitoring equipment and the abnormal analysis program of the edge server, and then locate the specific reason why the target monitoring equipment was misidentified, so as to facilitate the technicians to upgrade the relevant programs, thereby continuously improving the recognition accuracy and operation stability of the system.

[0108] In one implementation, the de-icing robot includes a communication module 71, a control module 72, an execution module 73 and a detection module 74, which are mainly responsible for tasks such as remote control data processing, motion control, sensor data acquisition, and power management.

[0109] In a specific example, the communication module 71 includes a 4G industrial router and a DR16 receiver, which are used to realize signal transmission between the background control terminal and the device body control terminal. After the industrial router is connected to the server through the 4G signal, a communication connection is established on the PC side through a virtual serial port, and various sensor data can be transmitted; after the remote control is paired with the DR16, the instructions of each lever of the remote control can be transmitted back to the development board through the serial port, and each execution module can be operated according to this data.

[0110] The control module 72 is the core part of the device body control end. The core controller uses the STM32 development board and transplants the real-time operating system for task management. The real-time operating system has the characteristics of occupying less system resources and real-time stability. It can efficiently manage multiple tasks, including setting task priorities and obtaining task-related information. The controller analyzes the remote control instructions and then drives the corresponding execution device in the execution module 73.

[0111] The execution module 73 serves as the end of the control system link and mainly executes various instructions of the remote controller, including the walking mechanism, clamping mechanism and de-icing mechanism, which are specifically implemented based on hardware structures such as micro air pumps, motor drivers, relays, linear push rods, and solenoid valves.

[0112] The detection module 74 includes an inclination sensor, a coulomb meter and a pressure sensor. The inclination sensor and the coulomb meter are connected to the controller via an RS485 electrical interface to realize the inclination detection and real-time power detection of the deicing robot. The pressure sensor is installed on the clamping mechanism of the deicing robot to detect the pressure between the transmission lines when the clamping mechanism clamps the transmission line. The clamping mechanism determines whether it is clamped by pressure data, and after clamping, the diameter of the ice-covered transmission line is determined by obtaining the control data of the clamping operation, and then the actual ice thickness of the transmission line is determined in combination with the diameter of the transmission line, and sent to the control module 72.

[0113] In addition, an embodiment of the present application also provides an electronic device, which includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the method steps performed by the edge server or control center in any method embodiment of the present application; wherein the processor may adopt a general-purpose central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), a graphics processing unit (GPU), or one or more integrated circuits, for executing relevant programs to implement the method steps performed by the edge server or control center in any method embodiment of the present application.

[0114] The processor may also be an integrated circuit electronic device with signal processing capability. In the implementation process, the method steps performed by the edge server or control center in any method embodiment of the present application may be completed by hardware integrated logic circuits in the processor or software instructions.

[0115] The above-mentioned processor can also be a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be combined and executed.

[0116] The software module may be located in a random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register or other mature storage media in the art. The storage medium is located in the memory, and the processor reads the information in the memory, and combines its hardware to complete the functions required to be performed by the units included in the data processing device of the embodiment of the present application, or executes the method steps performed by the edge server or control center in any method embodiment of the present application.

[0117] Another embodiment of the present application relates to a computer-readable storage medium storing a computer program, which implements the above method when executed by a processor.

[0118] Those skilled in the art can understand that all or part of the steps in the above-mentioned implementation methods can be completed by instructing related hardware through a program, and the program is stored in a storage medium, including several instructions for making a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) perform all or part of the steps of each implementation method of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0119] The above-mentioned embodiments only express the implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be based on the attached claims.

Claims

1. A method for removing ice from a power transmission line, characterized in that: The method is applied to a power transmission line monitoring and management system, the system comprising a plurality of monitoring devices, an edge server, a deicing robot and a control center, the plurality of monitoring devices being used to perform real-time monitoring of different sections of power transmission lines within the same monitoring area, the method comprising the steps of: The edge server receives the image monitoring data and the temperature control data sent by each monitoring device, and performs temperature control abnormality analysis on each monitoring device based on the image monitoring data and the temperature control data; Determine the monitoring device with abnormal temperature control abnormality analysis result as the target monitoring device; send each of the image monitoring data and the target monitoring device identifier to the control center; wherein the image monitoring data includes a monitoring image; The control center performs lens cleanliness analysis based on the lens cleanliness analysis model and the monitoring image collected by the target monitoring device to obtain the lens cleanliness of the target monitoring device; If the cleanliness of the lens does not meet the preset requirements, mark the temperature control module of the target monitoring device as to be repaired, send a temperature control instruction and a re-collection instruction to the target monitoring device, and when the cleanliness of the lens indicated by the monitoring image in the target image monitoring data returned by the target monitoring device meets the preset requirements, generate a first deicing task based on the target image monitoring data, and send the first deicing task to the deicing robot; The deicing robot receives and executes the first deicing task.

2. The method according to claim 1, characterized in that The method further includes: if the cleanliness of the lens meets the preset requirements, the control center generates a second deicing task based on each of the image monitoring data, and sends the second deicing task to the deicing robot, wherein the second deicing task includes detecting the actual ice thickness of the target transmission line monitored by the target monitoring device and sending it to the control center; The control center compares the actual ice thickness with the ice recognition result in the image monitoring data corresponding to the target monitoring device; If the comparison result indicates that the ice recognition result of the target monitoring device does not meet the preset requirements, then marking the ice recognition model of the target monitoring device as to be updated; If the comparison result indicates that the ice recognition result of the target monitoring device meets the preset requirements, the abnormal analysis program of the edge server is marked as to be updated.

3. The method according to claim 1, characterized in that The performing temperature control abnormality analysis on each monitoring device based on each image monitoring data and each temperature control data comprises: Sorting the ice thickness in each of the image monitoring data to determine a first ranking for each of the monitoring devices; Sorting each of the temperature control data to determine a second ranking of each of the monitoring devices; Determine the monitoring device whose difference between the first ranking and the second ranking exceeds a preset ranking threshold as a candidate device; Determine whether the candidate device has abnormal temperature control according to the thickness change trend in the image monitoring data sent by the candidate device.

4. The method according to claim 3, characterized in that The determining whether the candidate device has abnormal temperature control according to the thickness change trend in the image monitoring data sent by the candidate device includes: The climate change situation within the period is determined according to the thickness change trend. When the thickness change trend indicates that the climate continues to drop or drops suddenly within the period, if the ranking difference of the candidate device indicates that the number of heating times or the total duration is insufficient, it is determined that the temperature control of the candidate device is abnormal; if the ranking difference of the candidate device indicates that the number of heating times or the total duration is too much, it is determined that the temperature control of the candidate device is normal; when the thickness change trend indicates that the climate continues to rise or is generally warming within the period, if the ranking difference of the candidate device indicates that the number of heating times or the total duration is insufficient, it is determined that the temperature control of the candidate device is normal; if the ranking difference of the candidate device indicates that the number of heating times or the total duration is too much, it is determined that the temperature control of the candidate device is abnormal; when the thickness change trend indicates that the climate is relatively stable within the period, it is determined that the temperature control of the candidate device is abnormal.

5. The method according to claim 1, characterized in that The method also includes, when the lens cleanliness indicated by the monitoring image in the target image monitoring data returned by the target monitoring device does not meet the preset requirements, determining whether the target temperature control data is consistent with the temperature control instruction; if consistent, determining that the heating element in the temperature control module is faulty; otherwise, determining that the control element in the temperature control module is faulty.

6. A de-icing robot, characterized in that: The deicing robot includes a communication module and a control module, wherein the control module receives a second deicing task sent by a control center through the communication module, and before executing the second deicing task, obtains the actual ice thickness of the transmission line and sends it to the control center, wherein the second deicing task is generated based on the method described in claim 2.

7. The de-icing robot according to claim 6, characterized in that: The de-icing robot also includes a clamping mechanism and a pressure sensor, wherein a method for obtaining the actual ice thickness of the transmission line includes the clamping mechanism determining whether it is clamped to the transmission line based on pressure data between the clamping mechanism and the transmission line detected by the pressure sensor, and when the clamping is confirmed, determining the ice thickness of the transmission line based on control data.

8. A power transmission line monitoring and management system, characterized in that: The power transmission line monitoring and management system is used to implement the method described in any one of claims 1 to 5.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the method steps performed by the edge server or the control center in the method as described in any one of claims 1 to 5.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images