Detection system of current collection circuit

Abstract

The invention discloses a detection system of a current collection line, and relates to the technical field of power line detection. The system is used for detecting a current collection line and comprises an image acquisition unit, an abnormity identification unit and a wireless communication unit, the image acquisition unit is arranged on the unmanned aerial vehicle and is used for acquiring a visible light image and an infrared image of the current collection circuit and sending the visible light image and the infrared image to the abnormity identification unit; the abnormity identification unit is used for determining the line state of the current collection line based on the visible light image and the infrared image and sending the line state to the wireless communication unit, and the line state comprises a line normal state and a line abnormal state; and the wireless communication unit is used for sending the line state to a far-end upper computer. According to the technical scheme, the detection efficiency of the current collection circuit in the power system can be remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of power line detection, and in particular to a detection system for a collector line. Background Art

[0002] With the rapid growth of social, economic and cultural levels, the demand for electricity is also increasing. In order to meet the increasing demand for electricity, the power system is becoming more and more complex. Among them, the collector line is the core component of the power system, and its operating status is directly related to the safety and stability of the entire power grid. In order to ensure the safety of the power system, it is particularly important to regularly inspect the collector line.

[0003] At present, the inspection of power collection lines mainly relies on manual inspections, where inspectors inspect the transmission lines by walking, driving or climbing. Although this manual inspection method is still effective in some cases, it also has many shortcomings that cannot be ignored. Especially in the context of the continuous expansion of modern power networks, the problems of low efficiency, high cost and high safety risks of manual inspection methods are gradually emerging.

[0004] When inspecting the collector line through manual inspection, personnel need to climb to the location of the collector line to inspect the collector line, which makes the manual inspection slow. At the same time, the workload of the inspectors is large and limited, resulting in a long inspection period for the collector line. Especially in complex geographical environments, the coverage of manual inspection is very limited, resulting in potential hidden dangers that may not be discovered in time. At the same time, since many collector lines are distributed in complex environments such as high altitudes, cliffs, and mountains, the staff of manual inspections face high safety risks. In addition, high-altitude operations, harsh outdoor environments, and uncertain factors such as weather increase the possibility of injury to personnel, making manual inspection a high-risk and arduous task. At the same time, manual inspections are easily affected by the experience of inspectors, visual fatigue, environmental factors, etc., resulting in missed inspections and false inspections. For example, subtle equipment failures or environmental changes may be ignored, thereby causing potential safety hazards and increasing the risk of power system failures. Therefore, the existing detection methods for collector lines have low detection efficiency. Summary of the invention

[0005] In view of this, the present application provides a detection system for a collector line, the main purpose of which is to solve the technical problem of low detection efficiency of the existing detection method for the collector line.

[0006] According to a first aspect of the present invention, there is provided a detection system for a collector line, for detecting a collector line, the detection system for the collector line comprising an image acquisition unit, an abnormality recognition unit and a wireless communication unit;

[0007] The image acquisition unit is disposed at the drone, and is used to acquire a visible light image and an infrared image of the collector line, and send the visible light image and the infrared image to the abnormality recognition unit;

[0008] The abnormality recognition unit is used to determine the line state of the collector line based on the visible light image and the infrared image, and send the line state to the wireless communication unit, wherein the line state includes a normal line state and an abnormal line state;

[0009] The wireless communication unit is used to send the line status to a remote host computer.

[0010] In an optional embodiment, the image acquisition unit includes an image collector and an infrared thermal imager, the image collector is used to collect visible light images of the collector line, and the infrared thermal imager is used to collect infrared images of the collector line, wherein the image acquisition ranges of the image collector and the infrared thermal imager are the same; the abnormality recognition unit determines the line state of the collector line based on the visible light image and the infrared image, comprising: the abnormality recognition unit recognizes the device image of the line device to be detected of the collector line in the visible light image, and determines the device infrared image of the line device to be detected in the infrared image based on the position of the device image in the visible light image; determines whether the line device to be detected is damaged based on the device image, and determines whether the line device to be detected has a temperature abnormality based on the device infrared image; when the line device to be detected is damaged or the line device to be detected has a temperature abnormality, determines that the line state is the line abnormal state.

[0011] In an optional embodiment, the abnormality identification unit determines whether the line device to be detected has a temperature abnormality based on the infrared image of the device, including: the abnormality identification unit determines the temperature value of the line device to be detected based on the infrared image of the device; compares the temperature value with a normal temperature range value preset for the line device to be detected, and determines whether the temperature value is within the normal temperature range; if the temperature value is not within the normal temperature range, it is determined that the line device to be detected has a temperature abnormality.

[0012] In an optional embodiment, the abnormality identification unit determines the line status of the collector line based on the visible light image and the infrared image, and also includes: the abnormality identification unit inputs the device image and the device infrared image into a pre-trained abnormality detection neural network model to obtain the detection result of the line equipment to be detected, wherein the detection result includes a normal result and an abnormal result; when the detection result is the abnormal result, the line status is determined to be the line abnormal status.

[0013] In an optional embodiment, the anomaly detection neural network model includes a temperature anomaly detection sub-model, a crack detection sub-model, a corrosion detection sub-model and a looseness detection sub-model; the anomaly recognition unit inputs the device image and the device infrared image into the pre-trained anomaly detection neural network model to obtain the detection result of the line device to be detected, including: inputting the device infrared image into the temperature anomaly detection sub-model to obtain the temperature detection result of the line device to be detected, wherein the temperature detection result includes a normal temperature result and an abnormal temperature result; inputting the device image into the crack detection sub-model to obtain the crack detection result of the line device to be detected, wherein The crack detection result includes a crack result and a non-crack result; the device image is input into the corrosion detection sub-model to obtain the corrosion detection result of the line device to be detected, wherein the corrosion detection result includes a corrosion result and a non-corrosion result; the device image is input into the looseness detection sub-model to obtain the looseness detection result of the line device to be detected, wherein the looseness detection result includes a looseness result and a non-looseness result; when the temperature detection result is the temperature abnormality result, or the crack detection result is the crack result, or the corrosion detection result is the corrosion result, or the looseness detection result is the looseness result, it is determined that the detection result is the abnormal result.

[0014] In an optional embodiment, it is characterized in that a gimbal is provided at the drone; the image collector and the infrared thermal imaging instrument are arranged at the gimbal.

[0015] In an optional embodiment, the image acquisition unit is also used to respond to a position detection instruction, collect an environmental image of the location of the collector line, and send the environmental image and the position detection instruction to the abnormality identification unit; the abnormality identification unit is also configured to perform the following processing: the abnormality identification unit determines the geodetic coordinates of the UAV in response to the position detection instruction, and determines the relative position of the collector line and the UAV based on the environmental image, and determines the collector line geodetic coordinates of the collector line based on the geodetic coordinates and the relative position of the line; determines an obstacle in the environmental image, and determines the relative position of the obstacle between the obstacle and the UAV based on the environmental image, and determines the obstacle geodetic coordinates of the obstacle based on the obstacle relative position and the geodetic coordinates; determines the interval distance between the collector line and the obstacle based on the collector line geodetic coordinates and the obstacle geodetic coordinates, and compares the interval distance with a preset distance threshold. When the interval distance is less than or equal to the distance threshold, controls the wireless communication unit to send an alarm prompt message to the host computer.

[0016] In an optional embodiment, the image acquisition unit is also used to respond to a loose line detection instruction, collect an environmental image of the location of the collector line, and send the environmental image and the loose line detection instruction to the abnormality identification unit; the abnormality identification unit is also configured to perform the following processing: the abnormality identification unit determines the geodetic coordinates of the UAV in response to the loose line detection instruction, and determines the relative position of the collector line and the UAV based on the environmental image; determines the geodetic coordinates of the collector line based on the geodetic coordinates and the relative position of the line; determines whether the collector line has loose wires based on the collector line geodetic coordinates, and when the collector line has loose wires, controls the wireless communication unit to send a loose wire alarm message to the host computer.

[0017] In an optional embodiment, the collector line includes a line device; the detection system of the collector line also includes a sensor, each of which is arranged on the line device of the collector line; each of the sensors is connected to the abnormality identification unit, and is used to send the line parameters of the line device collected by the sensor to the abnormality identification unit; the abnormality identification unit is also used to obtain the line parameters, and compare the line parameters with the normal data range preset for the line equipment, and determine whether the line parameters are within the normal data range. If the line parameters are not within the normal data range, the wireless communication unit is controlled to send an abnormal alarm message to the host computer.

[0018] In an optional embodiment, the collector line detection system further includes a power supply, which includes a solar panel and a battery; the solar panel is used to charge the battery, and the battery is used to power the abnormality identification unit and the wireless communication unit.

[0019] The present invention provides a detection system for a collector line, which can fly to the collector line through an unmanned aerial vehicle, and collect visible light images and infrared images of the collector line through an image acquisition unit arranged on the unmanned aerial vehicle, and send the visible light image and infrared image to an abnormality recognition unit; further, the abnormality recognition unit determines the line state of the collector line based on the visible light image and infrared image of the collector line, determines whether the collector line is abnormal, and remotely sends the abnormality recognition result to a remote host computer via a wireless communication unit, so that the staff at the host computer can know the detection result in time. The technical solution provided by the present application eliminates the process of manual detection of the collector line, makes the detection work not limited by the complex geographical environment, and significantly enhances the detection efficiency of the collector line in the power system.

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0022] Figure 1 A schematic structural diagram of a detection system for a current collection line provided by an embodiment of the present invention is shown;

[0023] Figure 2 A schematic structural diagram of another current collection line detection system provided by an embodiment of the present invention is shown;

[0024] Figure 3 A schematic diagram of a current collection line and line detection points provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0025] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0026] In one embodiment, Figure 1As shown, a detection system for a collector circuit is provided, and the detection of the collector circuit is described by taking the system as an example, and the detection system for the collector circuit includes an image acquisition unit 100, an abnormality recognition unit 200, and a wireless communication unit 300. Among them, the image acquisition unit 100 can be a camera and an infrared thermal imager, which is used to collect visible light images and infrared images of the collector circuit (not shown in the figure), further, the abnormality recognition unit 200 can be a computer device such as a single-chip microcomputer and a digital signal processor, and the wireless communication unit 300 can be a wireless communication device such as a radio communication device.

[0027] Furthermore, the image acquisition unit 100 is arranged on a drone (not shown in the figure), the abnormality recognition unit 200 and the wireless communication unit 300 can be arranged in a detection device on the ground, and the image acquisition unit 100 and the abnormality recognition unit 200 can communicate with each other through wireless signals. Here, the drone can carry the image acquisition unit 100 for aerial inspections, and the drone has automatic flight, path planning and self-charging functions to enhance the inspection efficiency and coverage. In the actual monitoring process, the drone can be made to fly along the collector line so that the image acquisition unit 100 can collect visible light images and infrared images of the collector line along the way.

[0028] Specifically, the image acquisition unit 100 may send the visible light image and the infrared image to the abnormality recognition unit 200 .

[0029] Furthermore, the abnormality identification unit 200 is used to determine the line status of the collector line based on the visible light image and the infrared image, and send the line status to the wireless communication unit 300, wherein the line status includes a normal line status and an abnormal line status.

[0030] Specifically, a pre-trained abnormality recognition model may be provided in the abnormality recognition unit 200, and the abnormality recognition model is trained based on historical visible light images and historical infrared images of the collector line as samples, and the line states corresponding to the historical visible light images and historical infrared images are used as labels. When the collector line is actually detected, the visible light image and infrared image of the collector line can be input into the abnormality recognition model to obtain the line state output by the abnormality recognition model.

[0031] Furthermore, the wireless communication unit 300 is used to send the line status to a remote host computer 400. The remote host computer 400 may be a computer terminal used by relevant staff, and the relevant staff may determine the line status of the collector line at the host computer 400 to detect the collector line.

[0032] The detection system for the collector line provided in this embodiment can quickly inspect long-distance transmission lines through the detection system for the collector line with a drone, and collect images of the collector line in real time by equipping a high-definition camera and an infrared thermal imager to determine whether the collector line is abnormal. With the flexibility and efficiency of drone operations, especially in remote and dangerous areas, the efficiency of collector line detection is significantly improved, and it can replace manual work to complete high-risk tasks, effectively reducing the safety risks of workers, thereby improving inspection efficiency, reducing labor costs, reducing safety risks, and improving the accuracy of abnormal detection. It can also combine visual recognition, drones, the Internet of Things, and artificial intelligence technologies to greatly improve the coverage, accuracy, timeliness, and reliability of detection, and further ensure the safe and stable operation of the collector line. At the same time, it can also help power companies reduce inspection costs, reduce equipment maintenance costs, and provide technical guarantees for future intelligent power system management.

[0033] In an optional embodiment, if Figure 2 As shown, the image acquisition unit 100 includes an image collector 110 and an infrared thermal imager 120, the image collector 110 is used to collect visible light images of the collector circuit, and the infrared thermal imager 120 is used to collect infrared images of the collector circuit, wherein the image acquisition ranges of the image collector 110 and the infrared thermal imager 120 are the same, so that the visible light image collected by the image collector 110 and the infrared image collected by the infrared thermal imager 120 have the same image acquisition angle, and the visible light image and the infrared image have the same coverage range.

[0034] Furthermore, the abnormality identification unit 200 determines the line state of the collector line based on the visible light image and the infrared image, including:

[0035] First, the abnormality identification unit 200 identifies the device image of the circuit equipment to be detected of the collector circuit in the visible light image, and determines the device infrared image of the circuit equipment to be detected in the infrared image based on the position of the device image in the visible light image.

[0036] Here, when the abnormality identification unit 200 receives the visible light image and the infrared image, it can first perform denoising and enhancement processing on the visible light image and the infrared image to reduce the noise in the visible light image and the infrared image and improve the clarity of the visible light image and the infrared image. Further, feature extraction is performed on the visible light image to determine the line equipment to be detected in the visible light image, wherein the line equipment to be detected can be a fixture and power equipment in the collector line, including bolts, joints, cables, and transformers, etc. By determining whether the line equipment to be detected on the collector line is abnormal, it can be detected whether the collector line has a line abnormality.

[0037] Specifically, the visible light image can be input into a pre-trained first convolutional neural network model, which is trained based on the line equipment to be detected contained in the collector line and the pictures of the line equipment to be detected. Based on the received visible light image containing the line equipment to be detected, the type of the line equipment to be detected and the position of the device image of the line equipment to be detected in the visible light image can be determined, that is, the device image of the line equipment to be detected is framed in the visible light image. Furthermore, after the position of the device image of the line equipment to be detected in the visible light image is determined, because the visible light image and the infrared image have the same coverage range, the position and image range of the device image in the visible light image can be determined, and the image at the same position and image range in the infrared image can be determined as the device infrared image.

[0038] Then, it is determined whether the line device to be detected has equipment damage based on the equipment image, and it is determined whether the line device to be detected has temperature abnormality based on the equipment infrared image.

[0039] Specifically, the device image can be input into a pre-trained second convolutional neural network model for feature extraction to determine whether the line device to be detected is damaged; here, the second convolutional neural network model is trained by a common damaged image of the line device to be detected, and can determine whether the line device to be detected in the device image is damaged based on the device image. Furthermore, the temperature value of the line device to be detected can be determined based on the device infrared image of the line device to be detected, and the temperature value can be compared with the normal temperature range value preset for the line device to be detected to determine whether the temperature value is within the normal temperature range value; if the temperature value is not within the normal temperature range value, it is determined that the line device to be detected has a temperature abnormality. Here, the numerical range of the normal temperature range value can be determined according to actual conditions.

[0040] Here, after the type of line equipment to be detected is determined through the visible light image, the temperature value of the line equipment to be detected determined from the device infrared image can be compared with the normal temperature range value preset for the type of line equipment to be detected to determine whether the line equipment to be detected has a temperature abnormality.

[0041] Finally, when the line device to be detected is damaged or the line device to be detected has a temperature abnormality, the line state is determined to be the line abnormal state, and a device abnormality alarm can be sent to the host computer 400 via the wireless communication unit 300; conversely, when the line device to be detected is not damaged and the line device to be detected has no temperature abnormality, the line state is determined to be the line normal state.

[0042] The embodiments provided in the present application can quickly determine whether the circuit equipment to be inspected on the collector circuit is damaged or has abnormal temperature based on the visible light image and infrared image of the collector circuit, thereby significantly improving the detection efficiency of the collector circuit detection system.

[0043] In an optional embodiment, the abnormality identification unit determines the line state of the collector line based on the visible light image and the infrared image, further comprising:

[0044] First, the abnormality recognition unit inputs the device image and the device infrared image into a pre-trained abnormality detection neural network model to obtain the detection result of the line device to be detected, wherein the detection result includes a normal result and an abnormal result.

[0045] Among them, the anomaly detection neural network model is based on the historical equipment images and historical equipment infrared images of the line equipment to be detected on the collector circuit as samples, and the detection results of the line equipment to be detected are obtained by label training, among which the historical equipment images are the previous equipment images of the line equipment to be detected, and the historical equipment infrared images are the previous equipment infrared images of the line equipment to be detected.

[0046] Finally, when the detection result is the abnormal result, the line state is determined to be the abnormal state of the line, and an abnormal device alarm can be sent to the upper computer via the wireless communication unit; conversely, when the detection result of the line device to be detected is a normal result, the line state is determined to be the normal state of the line. In the embodiment provided by the present application, after determining that there is no equipment damage or temperature abnormality in the line device to be detected, the line device to be detected is detected again based on the abnormality detection neural network model to perform a more detailed detection of the line device to be detected, thereby improving the detection effectiveness of the detection system of the collector line.

[0047] In an optional embodiment, the anomaly detection neural network model includes a temperature anomaly detection sub-model, a crack detection sub-model, a corrosion detection sub-model and a looseness detection sub-model.

[0048] Among them, the temperature anomaly detection sub-model is pre-trained based on the historical equipment infrared images of the line equipment to be detected as samples and the temperature detection results of the line equipment to be detected as labels. It can determine whether the temperature detection result of the line equipment to be detected is a normal temperature result or a temperature abnormality result based on the equipment infrared images of the line equipment to be detected obtained during the detection process.

[0049] Furthermore, the crack detection sub-model is pre-trained based on historical equipment images of the line equipment to be detected as samples and the crack detection results of the line equipment to be detected as labels. It can determine whether cracks appear in the line equipment to be detected based on the equipment images of the line equipment to be detected obtained during the detection process. When cracks appear in the line equipment to be detected, the crack detection result is a crack result, and when no cracks appear in the line equipment to be detected, the crack detection result is a non-crack result.

[0050] Furthermore, the corrosion detection sub-model is pre-trained based on historical equipment images of the line equipment to be detected as samples and the corrosion detection results of the line equipment to be detected as labels. It can determine whether corrosion occurs in the line equipment to be detected based on the equipment images of the line equipment to be detected obtained during the detection process. When corrosion occurs in the line equipment to be detected, the corrosion detection result is a corrosion result, and when no corrosion occurs in the line equipment to be detected, the corrosion detection result is a non-corrosion result.

[0051] Furthermore, the looseness detection sub-model is pre-trained based on historical equipment images of the line equipment to be detected as samples and the looseness detection results of the line equipment to be detected as labels. It can determine whether the line equipment to be detected has loose connectors based on the equipment images of the line equipment to be detected obtained during the detection process. When the line equipment to be detected has loose connectors, the looseness detection result is a looseness result; when the line equipment to be detected has no loose connectors, the looseness detection result is a non-looseness result.

[0052] Furthermore, the abnormality recognition unit inputs the device image and the device infrared image into a pre-trained abnormality detection neural network model to obtain the detection result of the line device to be detected, including:

[0053] The abnormality identification unit inputs the infrared image of the device into the temperature abnormality detection sub-model to obtain the temperature detection result of the line device to be detected, wherein the temperature detection result includes a normal temperature result and an abnormal temperature result.

[0054] Furthermore, the device image is input into the crack detection sub-model to obtain a crack detection result of the line device to be detected, wherein the crack detection result includes a crack result and a non-crack result.

[0055] Furthermore, the device image is input into the corrosion detection sub-model to obtain the corrosion detection result of the circuit device to be detected, wherein the corrosion detection result includes a corrosion result and a non-corrosion result.

[0056] Furthermore, the device image is input into the loose detection sub-model to obtain a loose detection result of the line device to be detected, wherein the loose detection result includes a loose result and a non-loose result.

[0057] Further, when the temperature detection result is the abnormal temperature result, or the crack detection result is the crack result, or the corrosion detection result is the corrosion result, or the looseness detection result is the looseness result, the detection result is determined to be the abnormal result.

[0058] The embodiments provided in the present application can derive the types of abnormalities that occur in the line equipment to be detected, detect the temperature, cracks, corrosion and looseness of the line equipment to be detected respectively, and determine whether there are hidden fault hazards in the line equipment to be detected, thereby improving the functionality of the detection system of the collector line.

[0059] In an optional embodiment, a gimbal is provided at the drone; the image collector and the infrared thermal imaging instrument are provided on the table top of the gimbal. Furthermore, the drone can controllably adjust the angle of the gimbal to change the image acquisition angle of the image collector and the infrared thermal imaging instrument. Here, the image collector can work stably under different lighting conditions, and can automatically adjust acquisition parameters according to changes in ambient light, such as the aperture and sensitivity of the camera of the image collector. The embodiment provided in the present application can change the image acquisition angle of the image collector and the infrared thermal imaging instrument by operating the gimbal to adjust the detection object of the detection system of the collector line, thereby improving the operability of the detection system of the collector line.

[0060] In an optional embodiment, the detection system of the collector line also includes a power supply, and the power supply includes a solar panel and a battery; the power supply can be arranged at the detection device and connected to the abnormality identification unit and the wireless communication unit. Furthermore, the solar panel is connected to the battery for charging the battery; further, the battery is connected to the abnormality identification unit and the wireless communication unit for powering the abnormality identification unit and the wireless communication unit. The embodiment provided in the present application can power the abnormality identification unit and the wireless communication unit based on a power supply with a solar charging function and a power storage function, thereby improving the power supply stability of the detection system of the collector line.

[0061] In an optional embodiment, the image acquisition unit is also used to respond to the position detection instruction, collect the environmental image of the location of the collector line, and send the environmental image and the position detection instruction to the abnormality recognition unit; wherein the environmental image is a panoramic image collected by the image acquisition unit when the drone is located near the collector line, and the panoramic image includes an image of the collector line. Here, the image acquisition unit is provided with a wireless communication device, and the relevant staff can send the position detection instruction to the image acquisition unit based on the computer terminal, so that the image acquisition unit collects the environmental image of the location of the collector line, and sends the environmental image and the position detection instruction to the abnormality recognition unit.

[0062] Furthermore, the abnormality identification unit is further configured to perform the following processing:

[0063] First, the abnormality identification unit determines the geodetic coordinates of the UAV in response to the position detection instruction, determines the relative position of the collector line and the UAV based on the environmental image, and determines the collector line geodetic coordinates of the collector line based on the geodetic coordinates and the relative position of the line.

[0064] Specifically, when the abnormality recognition unit receives the position detection instruction and the environmental image, the longitude and latitude and altitude of the UAV are determined by the positioning device set in the unmanned place, and the geodetic coordinates of the UAV in the geodetic coordinate system are determined by the longitude and latitude and the flight altitude of the UAV. Furthermore, the relative position of the collector line and the UAV is determined by the collector line in the environmental image. Here, the collector line in the environmental image can be determined based on the pre-trained collector line recognition convolution model, and then the relative position of the collector line and the UAV can be determined based on the environmental image. Specifically, as Figure 3 As shown, multiple line detection points can be determined on the power collection line in the environment image, and the relative position between each line detection point and the drone can be determined. The more line detection points selected on the power collection line, the more accurate the relative position of the line can be determined.

[0065] Furthermore, after determining the relative position between the UAV and multiple line detection points on the collector line, the geodetic coordinates of each line detection point can be determined, and the geodetic coordinates of the multiple line detection points can be connected into a curve, and the geodetic coordinates of the curve in the geodetic coordinate system can be calculated to obtain the geodetic coordinates of the collector line.

[0066] Then, an obstacle is determined in the environment image, and a relative position of the obstacle between the obstacle and the drone is determined based on the environment image, and obstacle geodetic coordinates of the obstacle are determined based on the relative position of the obstacle and the geodetic coordinates.

[0067] Here, there must be no obstacles within a certain range around the collector line to ensure that the collector line is in an open space to ensure the safety of the collector line. In actual environments, trees and buildings often invade the vicinity of the collector line, causing safety hazards to the collector line.

[0068] In response to the above problems, the abnormality recognition unit determines obstacles in the environmental image based on the environmental image. Specifically, the environmental image can be input into a pre-trained tree recognition convolution model and a building recognition convolution model to determine the trees and buildings in the environmental image. Among them, the tree recognition convolution model is trained based on pictures of trees, and can determine the trees in the picture from the input picture, while the building recognition convolution model is trained based on pictures of buildings, and can determine the buildings in the picture from the input picture. Furthermore, after the trees and buildings in the environmental image are determined, the trees and buildings are uniformly determined as obstacles, and the relative position of the obstacle between the obstacle and the drone is determined based on the environmental image, and the obstacle geodetic coordinates of the obstacle are determined based on the relative position of the obstacle and the geodetic coordinates.

[0069] Finally, based on the geodetic coordinates of the collector line and the geodetic coordinates of the obstacle, the interval distance between the collector line and the obstacle is determined, and the interval distance is compared with a preset distance threshold. When the interval distance is less than or equal to the distance threshold, the wireless communication unit is controlled to send an alarm prompt message to the host computer.

[0070] Specifically, the interval distance between the collector line and the obstacle can be determined based on the collector line geodetic coordinates of the collector line and the geodetic coordinates of the obstacle, wherein the collector line geodetic coordinates are a set of geodetic coordinates of each point on the curve of the collector line. The point distance between the geodetic coordinates of each point and the geodetic coordinates of the obstacle can be determined, and the shortest point distance among them is determined as the interval distance.

[0071] Furthermore, the interval distance is compared with a preset distance threshold, wherein the distance threshold is determined through experiments or tests. When the distance between the collector line and the obstacle is greater than a certain value, the obstacle will not cause a safety hazard to the collector line. This value can be determined based on actual conditions.

[0072] Furthermore, when the interval distance is less than or equal to the distance threshold, the wireless communication unit is controlled to send an alarm prompt message to the host computer, wherein the alarm prompt message may include the location information of the drone where the abnormal identification unit is located, so that relevant personnel can determine where the collector line is at risk of being too close to an obstacle.

[0073] The embodiments provided in the present application can achieve accurate measurement of the collector circuit and its surrounding environment, avoid potential risks to the collector circuit caused by external objects, and improve the detection capability of the collector circuit detection system.

[0074] In an optional embodiment, the image acquisition unit is also used to respond to the loose line detection instruction, collect the environmental image of the location of the collector line, and send the environmental image and the loose line detection instruction to the abnormality recognition unit. The environmental image is a panoramic image collected by the image acquisition unit when the drone is located near the collector line, and the panoramic image includes an image of the collector line. Here, the relevant staff can send a loose line detection instruction to the image acquisition unit based on a computer terminal, so that the image acquisition unit collects the environmental image of the location of the collector line, and sends the environmental image and the loose line detection instruction to the abnormality recognition unit.

[0075] Furthermore, the abnormality identification unit is further configured to perform the following processing:

[0076] First, the abnormality identification unit determines the geodetic coordinates of the UAV in response to the loose line detection instruction, and determines the relative position of the line between the collector line and the UAV based on the environmental image.

[0077] Here, the collector line in the environmental image can be determined, and then the relative position of the line between the collector line and the drone can be determined based on the environmental image. In the actual detection process, multiple line detection points can be determined on the collector line, and the relative position of the line between the collector line and the drone can be determined by determining the relative position between each line detection point and the drone. Among them, the more line detection points selected on the collector line, the more accurate the relative position of the line is determined.

[0078] Then, based on the geodetic coordinates and the relative position of the line, the geodetic coordinates of the collector line of the collector line are determined. Specifically, after the relative position between the drone and multiple line detection points on the collector line is determined, the geodetic coordinates of each line detection point can be determined, and the geodetic coordinates of the multiple line detection points are connected into a curve, and the geodetic coordinates of the curve in the geodetic coordinate system are calculated to obtain the geodetic coordinates of the collector line of the collector line.

[0079] Finally, it is determined whether the collector line has loose wires based on the geodetic coordinates of the collector line, and when the collector line has loose wires, the wireless communication unit is controlled to send loose wire alarm information to the host computer.

[0080] Specifically, the curvature of the curve of the collector line can be calculated through the geodetic coordinates of the collector line, and the curvature can be compared with a preset curvature range to determine whether the curvature is within the curvature range, wherein the curvature range is the curvature range of the collector line under normal conditions determined by experiments or tests, and the value of the curvature range can be determined based on actual conditions. Further, when the curvature is not within the curvature range, it is determined that the collector line has loose wires, and conversely, when the curvature is within the curvature range, it is determined that the collector line has no loose wires.

[0081] The embodiment provided in the present application can determine whether the collector circuit is loose based on the coordinates of the collector circuit, and issue an alarm message in a timely manner when the collector circuit is loose, thereby improving the detection capability of the collector circuit detection system.

[0082] Furthermore, the detection system of the collector line also includes a set of self-diagnosis functions, which can monitor the working status of the detection system of the collector line in real time, including battery power, communication quality and sensor status, and issue an alarm when a fault occurs. Furthermore, the image acquisition unit can also acquire a tower image of the tower and send the tower image to the abnormality recognition unit, so that the abnormality recognition unit can determine whether the tower is tilted by judging the angle between the tower and the ground in the tower image, and send an alarm message to the upper computer in time when the tower is tilted.

[0083] In an optional embodiment, the collector line includes line equipment, which may be a fixing device and power equipment in the collector line, including bolts, joints, cables, and transformers.

[0084] Furthermore, the detection system of the collector line also includes sensors, and the number of sensors can be multiple, and each of the sensors is arranged on the line equipment of the collector line; wherein the sensors include temperature sensors, humidity sensors, and vibration sensors, etc., and the above sensors can be arranged on the line equipment on the collector line, and the temperature sensors, humidity sensors, and vibration sensors can respectively collect line parameters such as temperature parameters, humidity parameters, and vibration parameters of the line equipment. The line equipment where the sensors are installed, the installation positions of the sensors, and the number of installations can be determined according to actual conditions. Furthermore, the detection system of the collector line stores the line equipment where each sensor is installed. Specifically, the detection system of the collector line maps and stores the identification information of the sensor with the identification information of the line equipment where the sensor is installed.

[0085] Furthermore, each of the sensors is connected to the abnormality identification unit to send the line parameters of the line equipment collected by the sensor to the abnormality identification unit. Specifically, a wireless communication device may be provided at the sensor to establish a wireless communication connection between the sensor and the abnormality identification unit.

[0086] Furthermore, the abnormality identification unit is also used to obtain the line parameters, compare the line parameters with the normal data range preset for the line equipment, and determine whether the line parameters are within the normal data range. If the line parameters are not within the normal data range, the wireless communication unit is controlled to send an abnormal alarm message to the host computer.

[0087] Specifically, after receiving the line parameters, the abnormality identification unit determines the identification information of each sensor that sends the line parameters, and based on the identification information of each sensor, determines the sensors installed on the same line device among multiple sensors, and divides the sensors installed on the same line device into sensor groups. Further, the line device corresponding to the sensor group is determined, and then the normal data range corresponding to the line device is obtained.

[0088] As an example, if the sensor group includes a temperature sensor and a humidity sensor, and the normal data range of the line equipment corresponding to the sensor group includes a normal temperature data range and a normal humidity data range, then the line parameters obtained are the temperature value collected by the temperature sensor and the humidity value collected by the humidity sensor; further, determine whether the temperature value is within the normal temperature data range and whether the humidity value is within the normal humidity data range. If the temperature value is within the normal temperature data range and the humidity value is within the normal humidity data range, then determine that the line parameter is within the normal data range; conversely, if the temperature value is not within the normal temperature data range or the humidity value is not within the normal humidity data range, then determine that the line parameter is not within the normal data range, and control the wireless communication unit to send an abnormal alarm message to the host computer, and the abnormal alarm message includes the identification information of the line equipment.

[0089] The embodiments provided by the present application realize real-time monitoring and data transmission of equipment status by installing sensors at key components of the collector circuit, and store and analyze data through cloud servers, which can help engineers remotely evaluate the operation of equipment, timely discover potential problems of equipment and issue early warnings. In addition, the detection system of the collector circuit can also be remotely controlled and intelligently managed through the Internet of Things, such as remote switch operation and parameter adjustment, which improves the flexibility and controllability of the detection system of the collector circuit.

[0090] Furthermore, the line parameters collected by the above-mentioned sensors from the line equipment can also be stored, so as to discover hidden dangers such as aging and damage of the equipment in advance through data analysis based on the stored route parameters. Specifically, the line parameters of the historical line equipment from the time of commissioning to the aging and failure of the equipment can be recorded, and the above-mentioned line parameters can be used to generate line parameter trend data for characterizing the changes of line parameters over time in chronological order, wherein the line parameter trend data includes the manifestation of line parameters when various faults and abnormalities occur in the historical line equipment. In the subsequent detection process, the line parameters collected from the line equipment within a certain time range can be compared with the line parameter trend data to determine whether the line parameters collected from the line equipment conform to the manifestation of the line parameter trend data within a certain time range. When the line parameters collected from the line equipment conform to the manifestation of the line parameter trend data within a certain time range, the operation trend of the current line equipment and the hidden dangers such as the degree of aging and damage can be inferred based on the position of the time range in the entire life cycle of the historical line equipment and the faults and abnormalities of the historical line equipment during the time range, and the equipment maintenance plan and maintenance timing can be optimized accordingly to reduce downtime and maintenance costs.

[0091] Furthermore, for the above-mentioned line parameters, based on the combination of data analysis and AI technology, valuable information can be extracted from a large number of line parameters, and through machine learning, deep learning and other methods, a health model of the collector line can be established to predict the operating trend and potential failures of the collector line, thereby providing a scientific basis for the early formulation of maintenance plans. This not only improves the accuracy of the prediction, but also effectively reduces the unplanned downtime of the equipment. Furthermore, through the analysis of historical inspection data and image data, AI can automatically identify the status of the equipment and provide processing suggestions, such as identifying potential problems of line equipment in advance and performing preventive maintenance before failures occur. These measures have greatly improved the operating efficiency of the equipment and reduced operation and maintenance costs.

[0092] The detection system of the collector line provided in this embodiment can fly to the collector line through an unmanned aerial vehicle, and collect the visible light image and infrared image of the collector line through the image acquisition unit set on the unmanned aerial vehicle, and send the visible light image and infrared image to the abnormal recognition unit; further, the abnormal recognition unit determines the line state of the collector line based on the visible light image and infrared image of the collector line, determines whether the collector line is abnormal, and remotely sends the abnormal recognition result to the remote host computer via the wireless communication unit, so that the staff at the host computer can know the detection result in time; at the same time, it can detect whether the collector line has loose wires and whether the pole tower is tilted, and send an alarm message in time when it is detected that the collector line has loose wires and the pole tower is tilted. Further, it can also detect the environment where the collector line is located to avoid potential risks to the line caused by external objects. The technical solution provided by this application eliminates the process of manual detection of the collector line, so that the detection work is not restricted by the complex geographical environment, and significantly enhances the detection efficiency of the collector line in the power system.

[0093] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. In addition, the numbers corresponding to the various steps in the above embodiments only serve as identification and are not used as a limitation on the order in which the steps are executed. The order in which the steps are executed in each embodiment can be set according to actual conditions.

[0094] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.

Claims

1. A detection system for a collector line, used for detecting a collector line, characterized in that: The system includes an image acquisition unit, an abnormality recognition unit and a wireless communication unit; The image acquisition unit is disposed at the drone, and is used to acquire a visible light image and an infrared image of the collector line, and send the visible light image and the infrared image to the abnormality recognition unit; The abnormality recognition unit is used to determine the line state of the collector line based on the visible light image and the infrared image, and send the line state to the wireless communication unit, wherein the line state includes a normal line state and an abnormal line state; The wireless communication unit is used to send the line status to a remote host computer.

2. The detection system for the collector line according to claim 1, characterized in that: The image acquisition unit includes an image collector and an infrared thermal imager, wherein the image collector is used to acquire a visible light image of the collector circuit, and the infrared thermal imager is used to acquire an infrared image of the collector circuit, wherein the image acquisition range of the image collector and the infrared thermal imager is the same; The abnormality identification unit determines the line state of the collector line based on the visible light image and the infrared image, including: The abnormality recognition unit recognizes a device image of a circuit device to be detected of the collector circuit in the visible light image, and determines a device infrared image of the circuit device to be detected in the infrared image based on a position of the device image in the visible light image; Determine whether the line device to be detected is damaged based on the device image, and determine whether the line device to be detected is temperature abnormal based on the device infrared image; When the line device to be detected is damaged or the line device to be detected has a temperature abnormality, the line state is determined to be the line abnormal state.

3. The detection system for the collector line according to claim 2, characterized in that: The abnormality identification unit determines whether the temperature of the line device to be detected is abnormal based on the infrared image of the device, including: The abnormality identification unit determines the temperature value of the circuit equipment to be detected based on the infrared image of the equipment; Compare the temperature value with a normal temperature range value preset for the circuit device to be detected, and determine whether the temperature value is within the normal temperature range value; If the temperature value is not within the normal temperature range, it is determined that the circuit device to be detected has a temperature abnormality.

4. The detection system for the collector line according to claim 2, characterized in that: The abnormality identification unit determines the line state of the collector line based on the visible light image and the infrared image, and further includes: The abnormality recognition unit inputs the device image and the device infrared image into a pre-trained abnormality detection neural network model to obtain a detection result of the line device to be detected, wherein the detection result includes a normal result and an abnormal result; When the detection result is the abnormal result, the line state is determined to be the line abnormal state.

5. The detection system for the collector line according to claim 4, characterized in that: The anomaly detection neural network model includes a temperature anomaly detection sub-model, a crack detection sub-model, a corrosion detection sub-model and a looseness detection sub-model; The abnormality recognition unit inputs the device image and the device infrared image into a pre-trained abnormality detection neural network model to obtain a detection result of the line device to be detected, including: Inputting the infrared image of the device into the temperature anomaly detection sub-model to obtain the temperature detection result of the line device to be detected, wherein the temperature detection result includes a normal temperature result and an abnormal temperature result; Inputting the device image into the crack detection sub-model to obtain a crack detection result of the line device to be detected, wherein the crack detection result includes a crack result and a non-crack result; Inputting the device image into the corrosion detection sub-model to obtain the corrosion detection result of the circuit device to be detected, wherein the corrosion detection result includes a corrosion result and a non-corrosion result; Inputting the device image into the loose detection sub-model to obtain a loose detection result of the line device to be detected, wherein the loose detection result includes a loose result and a non-loose result; When the temperature detection result is the abnormal temperature result, or the crack detection result is the crack result, or the corrosion detection result is the corrosion result, or the looseness detection result is the looseness result, it is determined that the detection result is the abnormal result.

6. The detection system for the current collecting line according to claim 2, characterized in that: The drone is provided with a gimbal; The image collector and the infrared thermal imaging instrument are arranged on the pan / tilt platform.

7. The detection system for the collector line according to claim 1, characterized in that: The image acquisition unit is also used to collect an environmental image of the location of the collector line in response to the location detection instruction, and send the environmental image and the location detection instruction to the abnormality recognition unit; The abnormality identification unit is further configured to perform the following processing: The abnormality identification unit determines the geodetic coordinates of the UAV in response to the position detection instruction, determines the relative position of the line between the collector line and the UAV based on the environmental image, and determines the collector line geodetic coordinates of the collector line based on the geodetic coordinates and the line relative position; Determine an obstacle in the environment image, determine a relative position of the obstacle between the obstacle and the drone based on the environment image, and determine the obstacle geodetic coordinates of the obstacle based on the relative position of the obstacle and the geodetic coordinates; Based on the geodetic coordinates of the collector line and the geodetic coordinates of the obstacle, the interval distance between the collector line and the obstacle is determined, and the interval distance is compared with a preset distance threshold. When the interval distance is less than or equal to the distance threshold, the wireless communication unit is controlled to send an alarm prompt message to the host computer.

8. The detection system for the current collecting line according to claim 1, characterized in that: The image acquisition unit is also used to collect an environmental image of the location of the collector line in response to a line loose detection instruction, and send the environmental image and the line loose detection instruction to the abnormality recognition unit; The abnormality identification unit is further configured to perform the following processing: The abnormality identification unit determines the geodetic coordinates of the UAV in response to the line loosening detection instruction, and determines the relative position of the line between the collector line and the UAV based on the environmental image; Determining the collector line geodetic coordinates of the collector line based on the geodetic coordinates and the relative position of the line; Based on the geodetic coordinates of the collector line, it is determined whether the collector line has loose wires, and when the collector line has loose wires, the wireless communication unit is controlled to send loose wire alarm information to the host computer.

9. The detection system for the collector line according to claim 1, characterized in that: The current collecting line includes a line device; the detection system of the current collecting line also includes a sensor, each of the sensors is arranged on the line device of the current collecting line; Each of the sensors is connected to the abnormality identification unit and is used to send the line parameters of the line equipment collected by the sensor to the abnormality identification unit; The abnormality identification unit is also used to obtain the line parameters, compare the line parameters with the normal data range preset for the line equipment, and determine whether the line parameters are within the normal data range. If the line parameters are not within the normal data range, the wireless communication unit is controlled to send an abnormal alarm message to the host computer.

10. The detection system for the current collecting line according to claim 1, characterized in that: The detection system of the collector circuit also includes a power supply, and the power supply includes a solar panel and a battery; The solar panel is used to charge the storage battery, and the storage battery is used to supply power to the abnormality identification unit and the wireless communication unit.

Images