Power distribution network target identification and tracking method and system based on intelligent unmanned aerial vehicle

Abstract

The invention belongs to the technical field of unmanned aerial vehicle inspection control, and particularly relates to a power distribution network target identification and tracking method and system based on an intelligent unmanned aerial vehicle. Severe inspection environment identification is carried out in the inspection process, and multi-source inspection is carried out by using a multi-source inspection terminal in combination with different types of sensor data in the severe inspection environment; according to the method, the target recognition capability is enhanced, effective detection can be carried out even under the condition that vision is limited, meanwhile, multi-source inspection data comprises operation temperature data, comprehensive and effective inspection in a severe inspection environment is achieved, and in addition, the obstacle recognition and the dynamic tracking of the inspection result are carried out in the severe inspection environment, so that the inspection efficiency is improved. Therefore, obstacle recognition and tradeoff adjustment of the inspection frequency are carried out, so that resources and time can be more concentrated on places needing to be concerned actually, and the resource utilization efficiency, the data processing capability, the fault response capability and the system stability of inspection work can be remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle inspection control, and in particular relates to a distribution network target identification and tracking method and system based on intelligent unmanned aerial vehicles. Background Art

[0002] With the growth of electricity demand and the advancement of technology, the scale and complexity of distribution networks are also increasing, so regular inspections of distribution networks have become more important. However, distribution network lines are usually numerous and wide, and often cross complex terrains. Traditional manual inspections have hidden dangers in inspection efficiency and inspection safety. In this case, drones have the advantages of efficient inspections, no terrain restrictions, and no direct contact with live equipment, which greatly compensates for the shortcomings of manual inspections and become the first choice for distribution network inspections.

[0003] There are also technical solutions in the prior art that use drones to inspect distribution networks. For example, the Chinese invention patent with publication number CN117873130A proposes a low-cost autonomous inspection method and drone for distribution network drone inspection. By determining the drone's waypoints and inspection and photo-taking points, the drone's depth camera is used to capture video streams at the inspection and photo-taking points to generate a three-dimensional point cloud for power line point cloud recognition, and then the power line point cloud is used to plan the drone's flight route; at the same time, obstacle detection and avoidance are performed based on the three-dimensional point cloud. This solution can realize autonomous drone inspections for distribution networks without the need for human control. However, this solution uses the depth camera to inspect distribution network lines only in normal inspection environments, and is not suitable for inspections in harsh environments such as fog and sandstorms. In harsh inspection environments, the camera has limited vision, resulting in poor quality of captured images and videos, which greatly affects its effectiveness in distribution network inspections. In addition, camera inspections alone can only obtain visual operating indicators of the distribution network, such as fallen wires, cracks, dirt, etc., and lack detection of temperature distribution in the distribution network, resulting in overly single inspection indicators and affecting the ability to warn of inspection anomalies, because temperature anomalies are often a precursor to impending equipment failure.

[0004] In addition, due to the limited vision in harsh inspection environments, inspection and obstacle recognition need to be carried out simultaneously, which will generate a large amount of data that needs to be processed. However, the power reserve of the drone continues to decrease as the navigation progresses. In this case, it is difficult to maintain efficient data processing. However, the current inspection and obstacle recognition working frequencies are usually fixed, which can easily exceed the capabilities of the current processing system, resulting in untimely data processing or data loss, which not only affects the normal decision-making of obstacle avoidance and inspection, but also causes invalid data collection and waste of resources. Summary of the invention

[0005] To this end, one purpose of an embodiment of the present application is to provide a distribution network target identification and tracking method and system based on intelligent drones, by conducting multi-source information inspections of the distribution network in harsh inspection environments, and tracking obstacle manifestation trends and distribution network operation risk trends during the inspection process, thereby flexibly adjusting obstacle identification and inspection frequency, thereby effectively solving the problems mentioned in the background technology.

[0006] The objective of the present invention can be achieved through the following technical solutions: The first aspect of the present invention provides a distribution network target identification and tracking method based on an intelligent unmanned aerial vehicle, comprising the following steps: (1) configuring an environmental detection terminal, a multi-source inspection terminal, an obstacle identification terminal and a control center on the unmanned aerial vehicle.

[0007] (2) During the inspection of distribution network lines by drones, the environmental detection terminal is used in real time to detect visibility, dust particle concentration and wind speed, thereby identifying harsh inspection environments.

[0008] (3) When a harsh inspection environment is identified, the control center is used to adjust the inspection speed, and the obstacle identification terminal is used to collect obstacle information, where the obstacle information includes the location of the obstacle and the size and outline of the obstacle. At the same time, the multi-source inspection terminal is used to capture the multi-source operation information of the distribution network.

[0009] (4) Extract obstacle features based on the collected obstacle information, and extract distribution network operation features based on the multi-source operation information of the distribution network.

[0010] (5) The obstacle characteristics obtained within the set inspection cycle are constructed into a time series set. Similarly, the distribution network operation characteristics obtained within the set inspection cycle are constructed into a time series set, thereby performing obstacle performance trend analysis and distribution network operation risk trend analysis.

[0011] (6) The control center uses the obstacle performance trend and distribution network operation risk trend analysis results to make a trade-off adjustment between obstacle identification and inspection frequency.

[0012] The second aspect of the present invention proposes a distribution network target identification and tracking system based on an intelligent unmanned aerial vehicle, comprising the following modules: an inspection equipment setting module for configuring an environmental detection terminal, a multi-source inspection terminal, an obstacle identification terminal and a control center on the unmanned aerial vehicle.

[0013] The harsh inspection environment identification module is used to use the environmental detection terminal to detect visibility, dust particle concentration and wind speed in real time during the distribution network line inspection by the drone, thereby identifying the harsh inspection environment.

[0014] The inspection speed adjustment module is used to adjust the inspection speed using the control center when a harsh inspection environment is identified.

[0015] The obstacle collection module is used to collect obstacle information using the obstacle recognition terminal when a harsh inspection environment is identified, where the obstacle information includes the location of the obstacle and the size and outline of the obstacle.

[0016] The multi-source inspection module is used to capture multi-source operation information of the distribution network using the multi-source inspection terminal when a harsh inspection environment is identified.

[0017] The time series set construction module is used to extract obstacle features based on the collected obstacle information, and to extract distribution network operation features based on the multi-source operation information of the distribution network. At the same time, the obstacle features and distribution network operation features obtained within the set inspection cycle are constructed into a time series set.

[0018] The trend analysis module is used to analyze the obstacle performance trend and distribution network operation risk trend based on the constructed obstacle feature time series set and distribution network operation feature time series set.

[0019] The frequency trade-off adjustment module is used to use the control center to make trade-off adjustments in obstacle identification and inspection frequency based on the analysis results of obstacle trends and distribution network operation risk trends.

[0020] The first cloud database is used to store ideal visibility, ideal dust particle concentration and safe wind speed range.

[0021] The second cloud database is used to store normal line shape contour images and safe operating temperatures at different inspection locations.

[0022] Compared with the prior art, the advantages of the present invention are as follows: 1. The present invention identifies harsh inspection environments during the inspection process, and uses multi-source inspection terminals to combine different types of sensor data to perform multi-source inspections in harsh inspection environments, thereby enhancing the ability to identify targets and being able to perform effective detection even under visually limited conditions, thereby achieving effective inspections in harsh inspection environments and ensuring that inspections are not limited by environmental conditions.

[0023] 2. The inspection data obtained by the present invention using multi-source inspection terminals includes operating temperature data, which can provide more comprehensive distribution network status information, is conducive to early discovery of operating risks of the distribution network, can respond to risk issues more quickly, and shorten risk processing time. At the same time, the combination of multi-source data reduces the impact of data distortion that may occur in single data and improves the overall reliability of the data.

[0024] 3. The present invention obtains obstacle manifestation trends and distribution network operation risk trends by dynamically tracking obstacle identification and inspection results in harsh inspection environments, thereby making trade-off adjustments between obstacle identification and inspection frequency, so that resources and time can be more concentrated on areas that actually need attention, avoiding data over-limit processing when electricity is continuously consumed, and facilitating real-time matching of current processing resources to ensure that data processing capabilities are consistent with actual needs, preventing system overload and performance degradation, achieving stable inspection operation, and minimizing invalid data collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0026] Figure 1 The present invention is a flowchart of the steps for implementing the method.

[0027] Figure 2 This is a system module connection diagram of the present invention.

[0028] Figure 3 This is a schematic diagram of the UAV inspection equipment in the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Example 1

[0031] See also Figure 1 As shown, the present invention provides a distribution network target identification and tracking method based on an intelligent unmanned aerial vehicle, comprising the following steps: configuring an environment detection terminal, a multi-source inspection terminal, an obstacle identification terminal and a control center on the unmanned aerial vehicle.

[0032] In the specific implementation of the above scheme, the environmental detection terminal consists of a visibility sensor, a dust particle sensor and a wind speed sensor, wherein the visibility sensor is used to measure the degree of scattering of light by particles in the air to evaluate the visibility of the atmosphere, the dust particle sensor is used to detect and measure the concentration of suspended particles in the air, and the wind speed sensor is used to measure the speed of air flow.

[0033] In another implementation of the above scheme, the multi-source inspection terminal is composed of laser sensors and infrared sensors. The laser sensor can penetrate haze and sand, and capture the precise position and shape of wires and other structures through laser reflection, which is not affected by light conditions, and then provide real-time data feedback. It can quickly generate a three-dimensional point cloud model to help evaluate the status and potential problems of the distribution network. The infrared sensor detects by detecting infrared radiation emitted by objects. It performs well in low visibility conditions because infrared radiation can penetrate haze and sand and provide information about the existence and temperature of objects.

[0034] In another implementation of the above scheme, the obstacle recognition terminal is composed of a laser radar and an ultrasonic sensor, wherein the laser radar constructs a three-dimensional map of the surrounding environment by emitting a laser beam and measuring the time of reflected light to identify the size and outline of the obstacle, and the ultrasonic sensor measures the position and distance of the obstacle by emitting and receiving ultrasonic waves.

[0035] In other implementations of the above scheme, the control center is used to dynamically adjust the inspection speed and make a trade-off adjustment between the obstacle identification frequency and the inspection frequency based on the data collection results of the environment detection terminal, the multi-source inspection terminal, and the obstacle identification terminal.

[0036] The above mentioned environment detection terminal, multi-source inspection terminal, obstacle identification terminal and control center are connected as shown in Figure 3 shown.

[0037] (2) During the inspection of the distribution network lines by the UAV, the environmental detection terminal is used in real time to detect visibility, dust particle concentration and wind speed, thereby identifying the harsh inspection environment. The specific identification process is as follows: the detected visibility is compared with the ideal visibility stored in the first cloud database to obtain the comparison difference between the ideal visibility and the detected visibility, and then compared with the ideal visibility to obtain the first visual limitation.

[0038] The detected dust particle concentration and wind speed are compared with the ideal dust particle concentration and the upper limit wind speed in the safe wind speed range stored in the cloud database, and the second visual limitation is calculated in combination with the wind speed. The specific calculation formula is: Where q represents the detected atmospheric particle concentration, q 0 represents the ideal atmospheric particle concentration, ws represents the wind speed, and ws 上限 Indicates the upper limit wind speed in the safe wind speed range.

[0039] It is important to know that ideal visibility refers to the minimum visibility standard in which the line of sight in the environment is far enough to ensure that the sensor can accurately capture the target when the drone is performing inspection tasks. It is usually expressed in horizontal line of sight (such as meters or kilometers). The ideal visibility can be set by the model specifications of the multi-source inspection terminal and obstacle identification terminal carried by the drone. For example, the visibility should be at least 1,000 meters.

[0040] The ideal dust particle concentration refers to the minimum safety standard for the concentration of suspended particles in the air that does not affect the normal navigation of the drone. The ideal dust particle concentration can set the particle concentration standard based on the tolerance capacity of the multi-source inspection terminal and obstacle identification terminal carried by the drone.

[0041] The safe wind speed range refers to the conditions within which the drone can fly safely and stably. The lower limit wind speed in the safe wind speed range: below this wind speed, the stability and positioning accuracy of the drone may be affected, especially in light wind conditions, where a wind speed that is too low may cause the drone to drift or fail to remain stable. The upper limit wind speed: exceeding this wind speed, the drone may be difficult to control, and the flight performance may degrade, which may cause the drone to lose control, be damaged, or crash. The safe wind speed range can be found in the drone's user manual or technical specifications. Manufacturers usually provide recommended maximum and minimum wind speed ranges. Based on this information, the operating conditions of the drone within the safe wind speed range can be determined.

[0042] In the supplementary explanation of the above scheme, when dust particles are present during the flight of the UAV, the visual limitation analysis is not directly performed based on the concentration of the dust particles, but a comprehensive analysis is performed in combination with the wind speed. This is because when the wind speed is high, the strong wind will pick up more ground dust and disperse it into the air, which will increase the dust concentration in the air, thereby aggravating the visual limitation. When the wind speed is low, the dust particles on the ground will not be lifted or moved in large quantities, and the dust concentration is relatively stable. It can be seen that relying solely on the dust concentration may ignore the effect of wind speed on dust distribution. When the wind speed is high, even if the dust concentration itself is low, the wind will lift a large amount of dust, resulting in aggravated visual limitation. In addition, when the wind speed is high, the dust concentration may increase sharply, and the concentration data itself may not be able to reflect this change immediately, resulting in insufficient prediction of the actual visual limitation. Therefore, by combining the wind speed for visual limitation analysis, the visual limitation during UAV flight can be more accurately evaluated, avoiding the prediction bias caused by relying solely on the dust particle concentration.

[0043] The sum of the first visual limitation and the second visual limitation is compared with a set threshold. If the sum is higher than the set threshold, a harsh inspection environment is identified; otherwise, a harsh inspection environment is not identified.

[0044] (3) When a harsh inspection environment is identified, the control center is used to adjust the inspection speed, and the obstacle identification terminal is used to collect obstacle information, where the obstacle information includes the location of the obstacle and the size and outline of the obstacle. At the same time, the multi-source inspection terminal is used to capture the multi-source operation information of the distribution network.

[0045] As a preferred implementation of the above scheme, the inspection speed adjustment is implemented by using the control center as follows: the current inspection speed is retrieved and combined with the first visual limitation and the second visual limitation to enter the formula The current required inspection speed v′ is obtained, where v represents the current inspection speed, VL1 represents the first visual limitation, and VL 阈 A set threshold indicating the degree of visual limitation.

[0046] It is important to understand that when a harsh inspection environment is identified, the inspection speed needs to be reduced to improve flight stability and reduce collision risks, which helps multi-source inspection terminals to perform target inspections more accurately, thereby improving the inspection efficiency and safety of drones in complex environments.

[0047] Based on the current wind speed, the appropriate inspection speed corresponding to the current wind speed is extracted from the appropriate corresponding relationship between the wind speed and the inspection speed formed according to historical data.

[0048] Applied to the above embodiment, in order to establish a suitable corresponding relationship between wind speed and the inspection speed of the UAV, the historical flight data of the UAV under different wind speed conditions can be collected, including wind speed, inspection speed, and flight stability. The relationship model between wind speed and inspection speed is established through the collected data. Specifically, linear regression, polynomial regression or other data fitting methods can be used to establish a suitable corresponding relationship between wind speed and inspection speed.

[0049] Compare the current demand inspection speed with the appropriate inspection speed corresponding to the current wind speed. If the current demand inspection speed is higher than the appropriate inspection speed corresponding to the current wind speed, adjust the inspection speed according to the appropriate inspection speed corresponding to the current wind speed. If the current demand inspection speed is lower than the appropriate inspection speed corresponding to the current wind speed, adjust the inspection speed according to the current demand inspection speed. Of course, if the current demand inspection speed is consistent with the appropriate inspection speed corresponding to the current wind speed, adjust it according to the consistent inspection speed.

[0050] It needs to be explained that when the required inspection speed based on the visual limitation analysis is higher than the required inspection speed based on the wind speed analysis, the inspection speed is adjusted according to the required inspection speed based on the wind speed analysis; when the required inspection speed based on the visual limitation analysis is lower than the required inspection speed based on the wind speed analysis, the inspection speed is adjusted according to the required inspection speed based on the visual limitation analysis. This can help ensure flight stability and safety when the wind speed requires a lower speed and reduce flight control problems caused by excessive wind speed. When vision limitations require a lower speed, adjustment according to the visual limitation analysis can ensure the clarity and accuracy of data collection and avoid low inspection collection quality due to excessive speed. Therefore, by giving priority to the most restrictive factors, it is possible to balance flight stability and inspection effects while ensuring safety and data quality.

[0051] When the present invention identifies that the inspection speed needs to be reduced in a harsh inspection environment, the inspection speed adjustment value is determined comprehensively from two aspects: the visual limitation of visibility and dust particle statistics and the appropriate correspondence between wind speed and inspection speed established based on historical data. The inspection speed adjusted based on the visual limitation angle is for the inspection quality, while the inspection speed adjusted based on the wind speed is for flight stability. The combination of the two helps to formulate a more accurate inspection speed.

[0052] As a further preferred implementation of the above scheme, the multi-source operation information capture of the distribution network is implemented using a multi-source inspection terminal as follows: the laser sensor in the multi-source inspection terminal is used to collect laser data of the distribution network line to generate a three-dimensional point cloud model of the current inspection position, and the line shape contour is extracted from the model.

[0053] The infrared sensor in the multi-source inspection terminal is used to collect infrared images of the lines in the three-dimensional point cloud model, and the infrared images are divided into color distribution areas, and the line operating temperature is determined based on the chromaticity of each color distribution area.

[0054] It is important to know that different chromaticities on the infrared image correspond to different temperatures. The corresponding relationship between chromaticity and temperature can be used to determine the line operating temperature based on the color distribution area of ​​the infrared image. Specifically, when there is only one color in the infrared image, the temperature corresponding to the chromaticity of the color can be directly used as the line operating temperature. When there are multiple colors in the infrared image, the highest temperature among the temperatures corresponding to the chromaticities of various colors can be taken as the line operating temperature.

[0055] (4) Extract obstacle features based on the collected obstacle information, and extract distribution network operation features based on the multi-source operation information of the distribution network.

[0056] In the specific implementation of the above scheme, obstacle features are extracted based on the collected obstacle information as follows: the obstacle location is extracted from the collected obstacle information, and compared with the flight position of the UAV to obtain the required obstacle avoidance distance.

[0057] The obstacle size contour is extracted from the collected obstacle information to obtain the obstacle volume.

[0058] In a further specific implementation of the above scheme, the distribution network operation characteristics are extracted based on the multi-source operation information of the distribution network as follows: based on the current inspection position, the normal line shape contour of the corresponding inspection position is retrieved from the second cloud database.

[0059] The line shape contour of the current inspection position is overlapped and compared with the normal line shape contour to calculate the line contour overlap degree. Specifically, the line contour overlap degree can be obtained by obtaining the overlap contour size and comparing it with the normal line shape contour.

[0060] It should be understood that when a line breaks, a clear fracture is usually formed at the break, destroying the continuity of the line. The broken part may cause a clear interruption or change in the line profile. When the line falls, part or all of the line is separated from its supporting structure, which may cause the line position to shift or form a sagging profile. This fall will also make the line profile inconsistent with the normal shape profile. Therefore, when the line breaks or falls, the change in its profile will cause it to not coincide with the normal line profile.

[0061] The line operating temperature is subtracted from the safe operating temperature stored in the second cloud database, and the subtraction result is divided by the safe operating temperature to obtain the line operating temperature overscale.

[0062] (5) The obstacle characteristics obtained within the set inspection cycle are constructed into a time series set. Similarly, the distribution network operation characteristics obtained within the set inspection cycle are constructed into a time series set, thereby performing obstacle performance trend analysis and distribution network operation risk trend analysis.

[0063] It is important to know that the purpose of setting the inspection cycle is to form an inspection time window for corresponding trend analysis, which will facilitate subsequent obstacle identification and periodic adjustment of the inspection frequency, help to detect potential problems and risks early, and take preventive measures in advance. The inspection cycle can be divided by predicting the inspection duration.

[0064] As a specific implementation, the obstacle performance trend analysis in the above scheme refers to the following process: the obstacle performance coefficient statistics are performed on the required obstacle avoidance distance and obstacle volume corresponding to each obstacle feature in the constructed obstacle feature time series set. The specific statistical formula is: Wherein ψ represents the obstacle representation coefficient, R represents the set constant, and R>1, illustratively, R=4, wherein the reference obstacle avoidance distance and the reference obstacle volume are initially set in order to eliminate the dimensions of the corresponding parameters in the obstacle representation coefficient statistics.

[0065] A two-dimensional coordinate system is constructed with inspection time as the horizontal axis and obstacle performance coefficient as the vertical axis, and an obstacle performance change curve is formed in the constructed two-dimensional coordinate system for the obstacle performance coefficient corresponding to each obstacle feature in the obstacle feature time series set.

[0066] The curve change rate is extracted from the obstacle performance change curve, wherein the curve change rate is the overall change rate of the curve. For example, the curve change rate can be calculated by curve fitting, curve integration, numerical differentiation, etc., and then the curve change rate is used as the obstacle performance trend.

[0067] As a specific implementation, the distribution network operation risk trend analysis process is as follows: the line contour coincidence and line operation temperature over-scale corresponding to each distribution network operation feature in the constructed distribution network operation feature time series set are used to perform distribution network operation risk coefficient statistics. The specific statistical formula is:

[0068] In the formula Represents the distribution network operation risk factor.

[0069] A two-dimensional coordinate system is constructed with inspection time as the horizontal axis and distribution network operation risk coefficient as the vertical axis. A distribution network operation risk change curve is formed in the constructed two-dimensional coordinate system for the distribution network operation risk coefficient corresponding to each distribution network operation feature in the distribution network operation feature time series set.

[0070] Similarly, the curve change rate is extracted from the distribution network operation risk change curve to obtain the distribution network operation risk trend.

[0071] (6) The control center adjusts the associated frequency of obstacle identification and inspection based on the analysis results of obstacle trend and distribution network operation risk trend. The specific adjustment is as follows: extract the obstacle performance trend symbol and distribution network operation risk trend symbol calculated in the current inspection cycle. If the obstacle performance trend symbol and distribution network operation risk trend symbol are different, substitute them into the one-way adjustment model. Adjust the frequency of fault identification and inspection to the next inspection cycle In the model Indicates increasing the frequency of obstacle recognition. Indicates reducing the inspection frequency. Indicates reducing the frequency of obstacle recognition. Indicates increasing the inspection frequency, OT and DO represent the obstacle performance trend and distribution network operation risk trend respectively, and ∧ indicates and.

[0072] If the obstacle performance trend sign and the distribution network operation risk trend sign are the same, they are entered into the coordinated adjustment model Obtain the next inspection cycle fault identification and inspection frequency adjustment direction In the model, AS represents the power reserve adequacy. 0 It represents the set standard power reserve adequacy. For example, the standard power reserve adequacy is 0.8, and the power reserve adequacy is specifically obtained as follows: during the UAV inspection, real-time power detection is performed to obtain the power time series set of the current inspection cycle, and the power consumption rate is predicted using the power time series set. At the same time, the last power value is extracted from the power time series set as the remaining power value of the current inspection cycle, and then the full power value is subtracted from the remaining power value of the current inspection cycle and divided by the full power to obtain the full power proximity of the current inspection cycle. At the same time, the full power proximity is combined with the power consumption rate to calculate the power reserve adequacy.

[0073] In particular, the power consumption rate prediction using the power time series set is as follows: the power data in the power time series is combined with the previous power data and the next power data and then divided by the previous power data to obtain the consumption rate of adjacent power data, and the consumption rates of adjacent power data are averaged to obtain the predicted power consumption rate.

[0074] Furthermore, the calculation formula for the power reserve adequacy is:

[0075]

[0076] It should be noted that in harsh inspection environments, when both the obstacle performance trend and the distribution network operation risk increase, but the power reserve resources are limited, prioritizing the increase in the obstacle identification frequency can detect potential problems as early as possible, reduce the impact on the entire system, and prevent large-scale power outages or other serious problems due to equipment failures.

[0077] Example 2

[0078] Reference Figure 2 As shown, the present invention proposes a distribution network target identification and tracking system based on an intelligent unmanned aerial vehicle, comprising the following modules: an inspection equipment setting module for configuring an environmental detection terminal, a multi-source inspection terminal, an obstacle identification terminal and a control center on the unmanned aerial vehicle.

[0079] The harsh inspection environment identification module is connected to the inspection equipment setting module and is used to use the environmental detection terminal to detect visibility, dust particle concentration and wind speed in real time during the distribution network line inspection by the drone, thereby identifying the harsh inspection environment.

[0080] The inspection speed adjustment module is connected to the harsh inspection environment identification module and is used to use the control center to adjust the inspection speed when a harsh inspection environment is identified.

[0081] The obstacle collection module is connected to the harsh inspection environment identification module and the inspection equipment setting module respectively, and is used to use the obstacle identification terminal to collect obstacle information when a harsh inspection environment is identified, wherein the obstacle information includes the location of the obstacle and the size outline of the obstacle.

[0082] The multi-source inspection module is connected to the harsh inspection environment identification module and the inspection equipment setting module respectively, and is used to use the multi-source inspection terminal to capture multi-source operation information of the distribution network when a harsh inspection environment is identified.

[0083] The time series set construction module is connected to the obstacle collection module and the multi-source inspection module. It is used to extract obstacle features based on the collected obstacle information, and to extract distribution network operation features based on the multi-source operation information of the distribution network. At the same time, the obstacle features and distribution network operation features obtained within the set inspection period are constructed into a time series set.

[0084] The trend analysis module is connected to the time series set construction module and is used to analyze the obstacle performance trend and distribution network operation risk trend based on the constructed obstacle feature time series set and distribution network operation feature time series set.

[0085] The frequency trade-off adjustment module is connected to the trend analysis module and is used to use the control center to perform trade-off adjustment of obstacle identification and inspection frequency based on the obstacle trend and distribution network operation risk trend analysis results.

[0086] The first cloud database is connected to the harsh inspection environment identification module and is used to store ideal visibility, ideal dust particle concentration and safe wind speed range.

[0087] The second cloud database is connected to the time series set building module and is used to store normal line shape contour images and safe operating temperatures at different inspection locations.

[0088] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A distribution network target identification and tracking method based on intelligent drones, characterized in that: The following steps are involved: (1) Configure the UAV with an environmental detection terminal, a multi-source inspection terminal, an obstacle identification terminal, and a control center; (2) When the UAV is inspecting the distribution network lines, the environmental detection terminal is used to detect visibility, atmospheric particle concentration, and wind speed in real time, thereby identifying the harsh inspection environment; (3) When a harsh inspection environment is identified, the control center is used to adjust the inspection speed, and the obstacle identification terminal is used to collect obstacle information, where the obstacle information includes the location of the obstacle and the size of the obstacle. At the same time, the multi-source inspection terminal is used to capture the multi-source operation information of the distribution network; (4) Extracting obstacle features based on the collected obstacle information, and extracting distribution network operation features based on the multi-source operation information of the distribution network; (5) The obstacle characteristics obtained within the set inspection cycle are constructed into a time series set. Similarly, the distribution network operation characteristics obtained within the set inspection cycle are constructed into a time series set, thereby performing obstacle performance trend analysis and distribution network operation risk trend analysis; (6) The control center uses the obstacle performance trend and distribution network operation risk trend analysis results to make a trade-off adjustment between obstacle identification and inspection frequency.

2. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 1, characterized in that: The harsh inspection environment identification process is as follows: Comparing the detected visibility with the ideal visibility stored in the first cloud database to obtain a comparison difference between the ideal visibility and the detected visibility, and comparing it with the ideal visibility to obtain a first visual limitation; The detected atmospheric particle concentration and wind speed are compared with the ideal atmospheric particle concentration and the upper limit wind speed in the safe wind speed range stored in the first cloud database, and the second visual limitation is calculated in combination with the wind speed. The specific calculation formula is: Where q represents the detected atmospheric particle concentration, q0 represents the ideal atmospheric particle concentration, ws represents the wind speed, and ws 上限 Indicates the upper limit wind speed in the safe wind speed range; The sum of the first visual limitation and the second visual limitation is compared with a set threshold. If the sum is higher than the set threshold, a harsh inspection environment is identified; otherwise, a harsh inspection environment is not identified.

3. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 2, characterized in that: The inspection speed adjustment by the control center is implemented as follows: Retrieve the current inspection speed and combine it with the first visual limitation and the second visual limitation to import the formula The current required inspection speed v′ is obtained, where v represents the current inspection speed, VL1 represents the first visual limitation, and VL 阈 A set threshold indicating the degree of visual limitation; Based on the current wind speed, extract the appropriate inspection speed corresponding to the current wind speed from the appropriate corresponding relationship between wind speed and inspection speed formed based on historical data; Compare the current demand inspection speed with the appropriate inspection speed corresponding to the current wind speed. If the current demand inspection speed is higher than the appropriate inspection speed corresponding to the current wind speed, adjust the inspection speed according to the appropriate inspection speed corresponding to the current wind speed. If the current demand inspection speed is lower than the appropriate inspection speed corresponding to the current wind speed, adjust the inspection speed according to the current demand inspection speed.

4. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 1, characterized in that: The multi-source operation information capture of the distribution network using the multi-source inspection terminal is implemented as follows: The laser sensor in the multi-source inspection terminal is used to collect laser data of the distribution network line to generate a three-dimensional point cloud model of the current inspection position, and the line shape contour is extracted from the model; The infrared sensor in the multi-source inspection terminal is used to collect infrared images of the lines in the three-dimensional point cloud model, and the infrared images are divided into color distribution areas, and the line operating temperature is determined based on the chromaticity of each color distribution area.

5. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 1, characterized in that: The obstacle feature extraction based on the collected obstacle information is as follows: The obstacle location is extracted from the collected obstacle information and compared with the flight position of the drone to obtain the required obstacle avoidance distance; The obstacle size contour is extracted from the collected obstacle information to obtain the obstacle volume.

6. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 1, characterized in that: The extraction of distribution network operation characteristics based on multi-source operation information of the distribution network is as follows: Based on the current inspection position, retrieve the normal line shape contour of the corresponding inspection position from the second cloud database; Compare the line shape contour of the current inspection position with the normal line shape contour to calculate the line contour overlap degree; The line operating temperature is subtracted from the safe operating temperature stored in the second cloud database, and the subtraction result is divided by the safe operating temperature to obtain the line operating temperature overscale.

7. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 1, characterized in that: The obstacle performance trend analysis is as follows: Obstacle performance coefficients are calculated for the required obstacle avoidance distance and obstacle volume corresponding to each obstacle feature in the constructed obstacle feature time series set; A two-dimensional coordinate system is constructed with the inspection time as the horizontal axis and the obstacle performance coefficient as the vertical axis. An obstacle performance change curve is formed in the constructed two-dimensional coordinate system for the obstacle performance coefficient corresponding to each obstacle feature in the obstacle feature time series set. The curve change rate is extracted from the obstacle performance change curve, and then the curve change rate is used as the obstacle performance trend.

8. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 1, characterized in that: The distribution network operation risk trend analysis process is as follows: The line profile coincidence and line operating temperature over-scale corresponding to each distribution network operation feature in the constructed distribution network operation feature time series set are used to perform distribution network operation risk coefficient statistics; A two-dimensional coordinate system is constructed with the inspection time as the horizontal axis and the distribution network operation risk coefficient as the vertical axis, and a distribution network operation risk change curve is formed in the constructed two-dimensional coordinate system for the distribution network operation risk coefficient corresponding to each distribution network operation feature in the distribution network operation feature time series set; Similarly, the curve change rate is extracted from the distribution network operation risk change curve to obtain the distribution network operation risk trend.

9. The distribution network target identification and tracking method based on intelligent unmanned aerial vehicle according to claim 8, characterized in that: The trade-off between obstacle identification and inspection frequency is implemented as follows: Extract the obstacle performance trend symbol and distribution network operation risk trend symbol calculated in the current inspection cycle. If the obstacle performance trend symbol and distribution network operation risk trend symbol are different, substitute them into the one-way adjustment model. Obtain the next inspection cycle fault identification and inspection frequency adjustment direction In the model Indicates increasing the frequency of obstacle recognition. Indicates reducing the inspection frequency. Indicates reducing the frequency of obstacle recognition. Indicates increasing inspection frequency, OT and DO represent obstacle performance trend and distribution network operation risk trend respectively, ∧ indicates and; If the obstacle performance trend sign and the distribution network operation risk trend sign are the same, they are entered into the coordinated adjustment model Obtain the next inspection cycle fault identification and inspection frequency adjustment direction In the model, AS represents the power reserve adequacy, and AS0 represents the set standard power reserve adequacy. The power reserve adequacy is specifically obtained as follows: During the drone inspection process, real-time power detection is performed to obtain the power time series set of the current inspection cycle, and the power consumption rate is predicted using the power time series set. At the same time, the last power value is extracted from the power time series set as the remaining power value of the current inspection cycle. The full power value is then subtracted from the remaining power value of the current inspection cycle and divided by the full power value to obtain the full power proximity of the current inspection cycle. At the same time, the full power proximity is combined with the power consumption rate to calculate the power reserve adequacy.

10. The distribution network target identification and tracking system based on intelligent drones is characterized by: Includes the following modules: Inspection equipment setting module, used to configure the environment detection terminal, multi-source inspection terminal, obstacle identification terminal and control center on the drone; The harsh inspection environment identification module is used to detect visibility, atmospheric particle concentration and wind speed in real time using the environmental detection terminal during the inspection of the distribution network line by the drone, thereby identifying the harsh inspection environment; Inspection speed adjustment module, used to adjust the inspection speed using the control center when a harsh inspection environment is identified; The obstacle collection module is used to collect obstacle information using the obstacle identification terminal when a harsh inspection environment is identified, where the obstacle information includes the location of the obstacle and the size and outline of the obstacle; A multi-source inspection module is used to capture multi-source operation information of the distribution network using a multi-source inspection terminal when a harsh inspection environment is identified; A time series set construction module is used to extract obstacle features based on the collected obstacle information, and to extract distribution network operation features based on the multi-source operation information of the distribution network, and to construct a time series set of obstacle features and distribution network operation features obtained within a set inspection cycle; Trend analysis module, used to analyze obstacle performance trend and distribution network operation risk trend based on the constructed obstacle feature time series set and distribution network operation feature time series set; The frequency trade-off adjustment module is used to use the control center to perform trade-off adjustment of obstacle identification and inspection frequency based on the analysis results of obstacle trend and distribution network operation risk trend; The first cloud database is used to store ideal visibility, ideal atmospheric particle concentration and safe wind speed range; The second cloud database is used to store normal line shape contour images and safe operating temperatures at different inspection locations.

Images