A distribution network line unmanned aerial vehicle intelligent inspection and dynamic inventory management system

Abstract

This invention relates to the field of power distribution network line inspection and discloses an intelligent UAV inspection and dynamic inventory management system for power distribution network lines. The key technical aspects include a geographic acquisition module, a weather acquisition module, a fault diagnosis module, a maintenance record module, a statistics module, a planning module, and an inventory configuration module. This invention acquires terrain and vegetation conditions through the geographic acquisition module, obtains meteorological information through the weather acquisition module and combines it with historical fault data, and uses the statistics module to calculate the predicted fault probability for each area to determine key inspection areas and the required inventory quantity. It uses weather, actual terrain, and vegetation conditions to predict and mark potential disasters in the area, enabling the UAV to focus on key inspection areas during inspection and fault diagnosis, predict the inventory quantity of each component, and schedule inventory quantities in advance, thereby improving inventory utilization and reducing the possibility of insufficient parts when faults occur.

Description

Technical Field

[0001] This invention relates to the field of power distribution network line inspection, and more specifically to a power distribution network line unmanned aerial vehicle (UAV) intelligent inspection and dynamic inventory management system. Background Technology

[0002] Because over 20% of the power distribution network is located in uninhabited areas, mountainous regions, and other areas with harsh natural conditions, the transmission line equipment operates year-round in complex and variable open-air environments. After prolonged operation, conductors, lightning protection wires, insulators, and hardware in the distribution lines may experience strand breakage, corrosion, overheating, flashover, and other defects, seriously affecting the safe operation of the lines and creating hidden dangers. Existing power distribution network lines utilize unmanned aerial vehicles (UAVs) for intelligent inspection, identifying and judging the status of various components based on the images transmitted by the UAVs. However, existing inspection systems typically inspect along preset paths, failing to distinguish fault-prone points, and the fixed inventory of various parts easily leads to untimely parts distribution or excessive inventory. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the present invention aims to provide an intelligent inspection and dynamic inventory management system for power distribution lines using unmanned aerial vehicles (UAVs), which overcomes the aforementioned deficiencies in existing technologies and enables the system to easily identify key inspection areas and flexibly allocate parts inventory.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A smart unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines, including

[0006] The geographic acquisition module is pre-programmed with terrain information, including elevation, elevation difference, and landform characteristics. This module acquires and recognizes detection image information transmitted back by the drone and outputs vegetation information, including vegetation cover data, occlusion data, and vegetation growth status. The vegetation cover data reflects the types and area of ​​plant cover within the detection image; the occlusion data reflects the degree of occlusion by power distribution lines; and the vegetation growth status reflects the lushness of plant growth.

[0007] The weather acquisition module acquires meteorological information, which is defined as meteorological data.

[0008] The fault diagnosis module acquires the detected image information and outputs the inspection results, which include normal operation data and abnormal data.

[0009] The maintenance record module has a pre-set maintenance database. The abnormal data is recorded in the maintenance database and historical fault data is generated. Each historical fault data includes the fault time, fault location, faulty part, and number of parts replaced.

[0010] The statistics module is configured with statistical strategies to calculate the predicted failure probability for each region. The predicted failure probability is related to the terrain, vegetation, meteorological data, and historical failure data.

[0011] The planning module formulates an inspection plan within a preset planning period based on the predicted failure probability. The inspection plan includes inspection routes and key inspection areas.

[0012] The inventory configuration module divides maintenance areas according to the location of each power station. Each maintenance area covers the inspection route. The inventory configuration module configures the inventory quantity according to the predicted failure probability of each part in each maintenance area and allocates the inventory among multiple maintenance areas.

[0013] In this invention, preferably, the geographic acquisition module is configured with a first processing strategy, the first processing strategy specifically being...

[0014] Step 1: Obtain several linear features within the detected image information to define them as predicted lines.

[0015] Step 2: Determine if there are two parallel prediction lines. If so, define these two prediction lines as the reference set.

[0016] Step 3: Select the group with the smallest actual spacing within the reference group as the prediction group. The actual spacing represents the distance between two predicted straight lines within the same reference group. The two parallel straight lines within the prediction group are defined as the line edge lines. Measure the length of the line edge lines to define the exposure length.

[0017] Step 4: Extend each of the line edges to intersect with the edge of the detected image information to obtain extended line segments, obtain the length of the extended line segments as the predicted length, and calculate the occlusion data, which is expressed as the ratio of the difference between the predicted length and the exposed length to the predicted length.

[0018] In this invention, preferably, the geographic acquisition module is configured with a plant database, the plant database being configured with texture data and corresponding vegetation information, and the geographic acquisition module is configured with a second processing strategy, the second processing strategy specifically being...

[0019] Step 1: Perform binarization processing on the detected image information to segment the detected image information into multiple detection regions;

[0020] Step 2: Based on the RGB image of the detected image information, obtain the color features and texture features of each detection area, and use the texture features as an index to find the corresponding vegetation information. The ratio of the size of the detection area corresponding to the vegetation information to the area size of the detected image information is the plant coverage data.

[0021] Step 3: Segment the detection area based on edge detection to obtain each leaf region, and calculate the average area of ​​each leaf region. The average area reflects the average size of each leaf within the detection image information.

[0022] Step four: Obtain the terrain conditions and use them as terrain data; obtain the flight altitude of the drone and define it as altitude data; and combine it with the average area to obtain the vegetation growth.

[0023] In this invention, preferably, the maintenance scope of each power station can be extracted based on the terrain and vegetation conditions to identify multiple disaster-prone locations. These disaster-prone locations specifically include lightning strike-prone locations, impact-prone locations, flood-prone locations, landslide-prone locations, and vegetation conflict points. The key inspection area includes these disaster-prone locations.

[0024] In this invention, preferably, the inventory configuration module calculates the predicted inventory quantity based on the predicted failure probability. The inventory configuration module presets a minimum inventory quantity. If the minimum value of the predicted inventory quantity is less than the minimum inventory quantity, the minimum inventory quantity is used as the lower limit threshold of the inventory. Otherwise, the minimum predicted inventory quantity is used as the lower limit threshold of the inventory, and the maximum value of the predicted inventory quantity is used as the upper limit threshold of the inventory.

[0025] In this invention, preferably, the actual inventory quantity is compared with the lower inventory threshold and the upper inventory threshold. The actual inventory quantity reflects the actual quantity of each part. If the actual inventory quantity is greater than the upper inventory threshold, the corresponding part is transferred to other repair areas. If the actual inventory quantity is less than the lower inventory threshold, the corresponding part is transferred from other areas.

[0026] In this invention, preferably, the inventory configuration module is configured with an inventory scheduling strategy, selecting several allocation strategies so that the actual inventory quantity of each maintenance area meets the upper and lower inventory thresholds. The allocation strategy includes the number of parts to be scheduled and the scheduling route. The allocation strategy with the shortest scheduling route length is calculated and selected as the inventory scheduling strategy.

[0027] In this invention, preferably, the predicted inventory value is set to a minimum positive integer not less than the calculated value.

[0028] In this invention, preferably, the statistics module is equipped with a time correction unit, which corrects the weight parameters of historical fault data based on the time span since the last maintenance. The larger the time span, the higher the weight parameters of the historical fault data.

[0029] In this invention, preferably, the meteorological information is queried by the meteorological bureau.

[0030] The beneficial effects of this invention are:

[0031] This invention obtains terrain and vegetation conditions through a geographic data acquisition module, acquires meteorological information through a weather acquisition module, and combines it with historical fault data. It uses a statistical module to calculate the predicted fault probability of each region to identify key inspection areas and configure inventory quantities. By using weather, actual terrain, and vegetation conditions, it predicts and marks potential disasters in the region. This allows UAVs to focus on key inspection areas during inspections and fault diagnosis, predict the inventory of each part, and schedule inventory quantities in advance, thereby improving inventory utilization and reducing the occurrence of insufficient parts when faults occur. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the system architecture of the present invention.

[0033] Figure label:

[0034] 1. Geographic data acquisition module; 2. Weather acquisition module; 3. Fault diagnosis module; 4. Maintenance record module; 5. Statistics module; 6. Planning module; 7. Inventory configuration module. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] It should be noted that when a component is described as "fixed to" another component, it can be directly on the other component or may have a component in between. When a component is considered "connected to" another component, it can be directly connected to the other component or may have a component in between. When a component is considered "set on" another component, it can be directly set on the other component or may have a component in between. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0038] Reference Figure 1 As shown, this embodiment provides a power distribution line UAV intelligent inspection and dynamic inventory management system, including a geographic acquisition module 1, a weather acquisition module 2, a fault diagnosis module 3, a maintenance record module 4, a statistics module 5, a planning module 6, and an inventory configuration module 7. By utilizing weather, actual terrain, and vegetation conditions, the system predicts and marks potential disasters in the area. This allows the UAV to focus on key inspection areas during inspections and fault diagnosis, predict the inventory levels of various parts, and pre-allocate inventory quantities, thereby improving inventory utilization and reducing the occurrence of insufficient parts in the event of a fault.

[0039] The geographic acquisition module 1 is pre-set with terrain information, including elevation, elevation difference, and landform. The geographic acquisition module 1 acquires and identifies the detection image information returned by the UAV and outputs vegetation information, including vegetation coverage data, occlusion data, and vegetation growth. The vegetation coverage data reflects the types and areas of plant coverage in the detection image, the occlusion data reflects the degree of occlusion by the power distribution line, and the vegetation growth reflects the lushness of plant growth. The geographic acquisition module 1 is configured with a first processing strategy, which specifically includes the following steps: Step 1: Acquire several straight line features within the detected image information to define them as predicted straight lines; Step 2: Determine whether there are two parallel predicted straight lines. If so, define these two predicted straight lines as a reference group; Step 3: Select the group with the smallest actual distance within the reference group as the prediction group. The actual distance represents the distance between the two predicted straight lines within the same reference group. The two parallel straight lines within the prediction group are defined as line edges, and the length of the line edges is measured to define the exposure length; Step 4: Extend each line edge to intersect with the edge of the detected image information to obtain extended line segments. Obtain the length of the extended line segments to define the predicted length, and calculate occlusion data. The occlusion data is expressed as the ratio of the difference between the predicted length and the exposure length to the predicted length. Occlusion data can reflect the impact of vegetation height on distribution network lines. For example, when the degree of vegetation occlusion is high, the falling of vegetation and the breakage of branches can damage the distribution network lines. The impact of weather events such as heavy rain and strong winds is even greater.

[0040] The geographic acquisition module 1 is configured with a plant database, which contains texture data and corresponding vegetation information. The geographic acquisition module 1 is also configured with a second processing strategy, which specifically includes the following steps: Step 1: Binarize the detected image information to segment it into multiple detection regions; Step 2: Obtain the color and texture features of each detection region based on the RGB image of the detected image information, and use the texture features as an index to find the corresponding vegetation information. Calculate the ratio of the size of the detection region corresponding to the vegetation information to the area size of the detected image information, which is the plant cover data; Step 3: Segment the detection region based on edge detection to obtain each leaf region, and calculate the average area of ​​each leaf region. The average area reflects the average size of each leaf within the detected image information; Step 4: Obtain the terrain conditions as terrain data, obtain the drone's flight altitude as altitude data, and combine it with the average area to obtain the vegetation growth status. The maintenance scope of each power station can be determined by extracting multiple disaster-prone locations based on terrain and vegetation conditions. These locations specifically include areas prone to lightning strikes, impacts, floods, landslides, and vegetation conflicts. Key inspection areas include these disaster-prone locations. Vegetation cover data can be used to predict the types of natural disasters that may occur, and determining the inventory of spare parts based on the types of accidents that are likely to occur is more intuitive and accurate.

[0041] The weather acquisition module 2 acquires meteorological information, which has been defined as meteorological data. The meteorological information is queried by the meteorological bureau.

[0042] The fault diagnosis module 3 acquires the detected image information and outputs the inspection results, which include normal operation data and abnormal data. The fault diagnosis module 3 utilizes existing judgment logic, and therefore will not be elaborated upon in this application. The image information captured by the drone is transmitted back to the inspection system in real time. The inspection system uses AI image recognition to intelligently analyze potential risks and hazards. If a risk or hazard is detected, a pop-up notification is displayed on the app, and relevant responsible persons are also notified via SMS, WeChat mini-program, etc.

[0043] The maintenance record module 4 has a pre-set maintenance database. Abnormal data is recorded in the maintenance database and historical fault data is generated. Each historical fault data includes the fault time, fault location, faulty part, and number of replaced parts. Historical fault data is updated based on abnormal signals and after manual confirmation.

[0044] The statistics module 5 is configured with a statistical strategy to calculate the predicted failure probability for each region. The predicted failure probability is related to terrain conditions, vegetation conditions, meteorological data, and historical failure data. The statistics module 5 is also configured with a time correction unit, which adjusts the weight parameters of historical failure data based on the time span since the last maintenance. The larger the time span, the higher the weight parameters of the historical failure data.

[0045] The planning module 6 formulates an inspection plan within a preset planning period based on the predicted failure probability. The inspection plan includes the inspection route and key inspection areas. The key inspection areas are related to the terrain and vegetation conditions. In addition to the disaster-prone points mentioned above, the locations of easily damaged parts are also considered key inspection areas. The power supply station conducts daily drone inspections using a drone intelligent inspection APP. The inspection focus can be dynamically adjusted before drone takeoff and during the inspection process, and the adjusted inspection focus is synchronized to the drone in real time.

[0046] The inventory configuration module 7 divides maintenance areas according to the location of each power station. Each maintenance area covers the inspection route. The inventory configuration module 7 configures the inventory quantity based on the predicted failure probability of each part in each maintenance area and allocates the inventory among multiple maintenance areas. The inventory configuration module 7 calculates the predicted inventory quantity based on the predicted failure probability. The inventory configuration module 7 presets a minimum inventory quantity. If the minimum predicted inventory quantity is less than the minimum inventory quantity, the minimum inventory quantity is used as the lower inventory limit threshold; otherwise, the minimum predicted inventory quantity is used as the lower inventory limit threshold, and the maximum predicted inventory quantity is used as the upper inventory limit threshold. The actual inventory quantity is compared with the lower and upper inventory limit thresholds. The actual inventory quantity reflects the actual quantity of each part. If the actual inventory quantity is greater than the upper inventory limit threshold, the corresponding part is allocated to other maintenance areas; if the actual inventory quantity is less than the lower inventory limit threshold, the corresponding part is allocated from other areas. The inventory configuration module 7 is configured with inventory scheduling strategies. Several allocation strategies are selected to ensure that the actual inventory levels in each maintenance area meet the upper and lower inventory thresholds. The allocation strategies include the number of parts to be allocated and the allocation routes. The allocation strategy with the shortest allocation route length is calculated and selected as the inventory scheduling strategy. The predicted inventory level is set to the smallest positive integer not less than the calculated value.

[0047] The drone-based intelligent inspection system, based on big data analysis and considering geographical location and climate characteristics, dynamically adjusts the upper and lower limits of spare parts inventory at power supply stations. Through inspection logs and maintenance records, and based on statistical information of all equipment on the distribution network lines under the jurisdiction of the power supply station, combined with fault rates, climate, and geographical location, the system performs intelligent big data analysis to adjust the lower limit of spare parts inventory in real time. Spare parts with inventory below the lower limit are promptly allocated to the warehouse to ensure sufficient spare parts for repair should a fault occur in the distribution network. Spare parts with inventory exceeding the upper limit are promptly transferred to the central warehouse to ensure the most economical inventory allocation.

[0048] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A smart unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines, characterized in that: The system includes a geographic acquisition module (1), which is pre-set with terrain information, including elevation, elevation difference, and landform. The geographic acquisition module (1) acquires and identifies the detection image information transmitted back by the UAV and outputs vegetation information, including vegetation cover data, occlusion data, and vegetation growth. The vegetation cover data reflects the types and areas of plant cover in the detection image, the occlusion data reflects the degree of occlusion by the power distribution line, and the vegetation growth reflects the lushness of plant growth. Weather acquisition module (2), wherein the weather acquisition module (2) acquires meteorological information that has been defined as meteorological data, The fault judgment module (3) acquires the detection image information and outputs the inspection results, which include normal operation data and abnormal data. The maintenance record module (4) has a pre-set maintenance database. The abnormal data is recorded in the maintenance database and historical fault data is generated. Each piece of historical fault data includes the fault time, fault location, faulty part, and number of parts replaced. The statistics module (5) is configured with a statistical strategy to calculate the predicted failure probability of each region. The predicted failure probability is related to the terrain, vegetation, meteorological data, and historical failure data. The planning module (6) formulates an inspection plan within a preset planning period based on the predicted failure probability. The inspection plan includes an inspection route and key inspection areas. The inventory configuration module (7) divides the maintenance range according to the location of each power supply station. Each maintenance range covers the inspection route. The inventory configuration module (7) configures the inventory quantity according to the predicted failure probability of each part in each maintenance range and allocates the inventory among multiple maintenance ranges. The geographic acquisition module (1) is configured with a first processing strategy, which is specifically as follows: Step 1: Obtain several linear features within the detected image information to define them as predicted lines. Step 2: Determine if there are two parallel prediction lines. If so, define these two prediction lines as the reference set. Step 3: Select the group with the smallest actual spacing within the reference group as the prediction group. The actual spacing represents the distance between two predicted straight lines within the same reference group. The two parallel straight lines within the prediction group are defined as the line edge lines. Measure the length of the line edge lines to define the exposure length. Step 4: Extend each of the line edges to intersect with the edge of the detected image information to obtain extended line segments, obtain the length of the extended line segments as the predicted length, and calculate the occlusion data, which is expressed as the ratio of the difference between the predicted length and the exposed length to the predicted length.

2. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 1, characterized in that: The geographic acquisition module (1) is configured with a plant database, which contains texture data and corresponding vegetation information. The geographic acquisition module (1) is also configured with a second processing strategy, which specifically includes... Step 1: Perform binarization processing on the detected image information to segment the detected image information into multiple detection regions; Step 2: Based on the RGB image of the detected image information, obtain the color features and texture features of each detection area, and use the texture features as an index to find the corresponding vegetation information. The ratio of the size of the detection area corresponding to the vegetation information to the area size of the detected image information is the plant cover data. Step 3: Segment the detection area based on edge detection to obtain each leaf region, and calculate the average area of ​​each leaf region. The average area reflects the average size of each leaf within the detection image information. Step four: Obtain the terrain conditions and use them as terrain data; obtain the flight altitude of the drone and define it as altitude data; and combine it with the average area to obtain the vegetation growth.

3. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 1, characterized in that: The maintenance scope of each power supply station can be determined by extracting multiple disaster-prone locations based on the terrain and vegetation conditions. These disaster-prone locations specifically include lightning strike-prone locations, impact-prone locations, flood-prone locations, landslide-prone locations, and vegetation conflict points. The key inspection areas include these disaster-prone locations.

4. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 1, characterized in that: The inventory configuration module (7) calculates the predicted inventory quantity based on the predicted failure probability. The inventory configuration module (7) has a preset minimum inventory quantity. If the minimum value of the predicted inventory quantity is less than the minimum inventory quantity, the minimum inventory quantity is used as the lower limit threshold of the inventory. Otherwise, the minimum predicted inventory quantity is used as the lower limit threshold of the inventory, and the maximum value of the predicted inventory quantity is used as the upper limit threshold of the inventory.

5. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 4, characterized in that: The actual inventory quantity is compared with the lower inventory threshold and the upper inventory threshold. The actual inventory quantity reflects the actual quantity of each part. If the actual inventory quantity is greater than the upper inventory threshold, the corresponding part is transferred to other repair areas. If the actual inventory quantity is less than the lower inventory threshold, the corresponding part is transferred from other areas.

6. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 4, characterized in that: The inventory configuration module (7) is configured with an inventory scheduling strategy. Several allocation strategies are selected so that the actual inventory of each maintenance area meets the upper limit threshold and the lower limit threshold of the inventory. The allocation strategy includes the number of parts to be scheduled and the scheduling route. The allocation strategy with the shortest scheduling route length is calculated and selected and the inventory scheduling strategy is implemented.

7. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 4, characterized in that: The predicted inventory value is set to the smallest positive integer not less than the calculated value.

8. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 1, characterized in that: The statistics module (5) is equipped with a time correction unit. The time correction unit corrects the weight parameters of historical fault data based on the time span since the last maintenance. The larger the time span, the higher the weight parameters of the historical fault data.

9. The intelligent unmanned aerial vehicle (UAV) inspection and dynamic inventory management system for power distribution lines according to claim 1, characterized in that: The meteorological information was obtained through a query by the meteorological bureau.

Images