WO2024040566A1 - 基于图像识别的变电站智能巡检系统及方法 - Google Patents
基于图像识别的变电站智能巡检系统及方法 Download PDFInfo
- Publication number
- WO2024040566A1 WO2024040566A1 PCT/CN2022/115106 CN2022115106W WO2024040566A1 WO 2024040566 A1 WO2024040566 A1 WO 2024040566A1 CN 2022115106 W CN2022115106 W CN 2022115106W WO 2024040566 A1 WO2024040566 A1 WO 2024040566A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- inspection
- substation
- image
- drone
- equipment
- Prior art date
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 255
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000008569 process Effects 0.000 claims description 14
- 238000010438 heat treatment Methods 0.000 claims description 13
- 239000000284 extract Substances 0.000 claims description 12
- 238000013439 planning Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 238000003708 edge detection Methods 0.000 claims description 9
- 238000004891 communication Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 238000007499 fusion processing Methods 0.000 claims description 3
- 230000008447 perception Effects 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 2
- 238000012423 maintenance Methods 0.000 description 10
- 230000005670 electromagnetic radiation Effects 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241000255925 Diptera Species 0.000 description 1
- 241001515997 Eristalis tenax Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
- 238000000825 ultraviolet detection Methods 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
Definitions
- the present invention relates to the technical field of substation operation and maintenance, and more specifically to an intelligent substation inspection system and method based on image recognition.
- the operation quality of the substation is closely related to the safety and stable operation of the entire power grid. Inspection of substation equipment is a basic task to ensure the safety of power equipment, improve the reliability of power equipment, and ensure the operation of power equipment with a low failure rate.
- inspection robots can effectively avoid the problems of manual inspection methods and complete corresponding inspection tasks.
- intelligent inspection robots cannot charge independently and may deviate from the inspection route during the inspection process; due to the surrounding Defects in the environment or the inspection robot itself may result in unclear photos, making it impossible to obtain accurate inspection results, and making it impossible to accurately judge possible faults in the substation.
- the present invention provides an intelligent substation inspection system and method based on image recognition, which can obtain accurate substation inspection results, effectively reduce the labor intensity of manual inspections, reduce substation operation and maintenance costs, and improve inspection operations. and the level of automation and intelligence of management.
- An intelligent inspection method for substations based on image recognition including the following steps:
- the determined inspection items and the optimal inspection route arrange the codes of the inspection items, paste the codes on the corresponding equipment, and store the codes of the inspection items and the corresponding location information in the database;
- the inspection equipment is used to inspect each inspection item in sequence, and the images and electronic tag codes of each inspection item are obtained;
- this solution can realize automatic inspection of substations through drones and inspection robots, obtain accurate fault type and fault location information, reduce the use of labor, and reduce the operation and maintenance costs of substations. Improve inspection efficiency and intelligence level.
- determining the optimal inspection route for the inspection equipment specifically includes the following steps:
- drone aerial photography is carried out, and the oblique photography data of the substation is collected through the tilt aerial camera mounted on the drone;
- LiDAR is used to inspect the substation to obtain three-dimensional point cloud data for each inspection item; then, the three-dimensional point cloud data is denoised to remove noise caused by environmental factors and problems with the instrument itself, and the denoised three-dimensional point cloud data is Colorize the point cloud data to obtain colored three-dimensional point cloud data;
- the three-dimensional reconstruction of the substation by the drone is completed;
- the flight route of the drone is planned to control the drone to conduct automatic inspection of the substation.
- planning the flight route of the UAV specifically includes the following steps:
- the UAV conducts inspections according to the initial flight route, and uses the depth sensing camera mounted on the UAV to extract information including the three-dimensional position for depth perception to verify whether the UAV's flight route is optimal, so as to automatically Plan the optimal flight route of the drone during the inspection process.
- the technical effect achieved by the above technical solution is: three-dimensional reconstruction of the substation based on the data obtained by the oblique aerial camera and lidar.
- the flight route of the UAV is planned and continuously calibrated and adjusted, which can be used during the automatic inspection process.
- the optimal path is continuously determined to ensure that the drone can inspect the entire area of the substation and improve the inspection efficiency of the drone.
- the electromagnetic radiation data of each inspection item in the substation can also be obtained; based on the electromagnetic radiation data, the location information and corresponding location for image collection of each inspection item are determined.
- the acquisition equipment below adjusts parameters to improve the flight route planning of the UAV, which can avoid the influence of the electromagnetic environment of the substation and continuously adjust the status of the UAV to ensure the accuracy of its flight route.
- determining the fault type and fault location of the substation specifically includes the following steps:
- the electronic tag code on the corresponding device is scanned, and the scanned electronic tag code is uploaded to the computer through infrared communication;
- Substation images are obtained through drones and inspection robots and fused to generate three-channel status images
- the substation fault type is judged; wherein the substation fault type includes: physical damage, electrical faults caused by heat, and electrical faults caused by discharge.
- judging the type of substation fault includes the following steps:
- the industrial UV camera mounted on the drone collects UV images of the substation to determine whether there is a corona area in the three-channel status image. If there is a fault caused by discharge, eliminate background interference and determine the location and intensity of the discharge point.
- the method further includes:
- the canny operator edge detection method is used to perform edge detection on the grayscale image to determine the location of equipment with electrical faults caused by heating.
- the method further includes:
- the invention also discloses an intelligent substation inspection system based on image recognition, which includes: an inspection drone subsystem, an inspection robot subsystem, and a computer; the inspection drone subsystem and the inspection robot subsystem all pass The wireless communication module is wirelessly connected to the computer;
- the inspection drone subsystem is equipped with inspection equipment, which is used to inspect various inspection items determined according to the optimal inspection route of the drone, collect substation images and transmit them to the computer;
- the inspection robot subsystem is equipped with a handheld collector, which is used to scan the electronic label codes of the corresponding equipment of each inspection item and transmit them to the computer;
- the electronic tag codes and location information of the corresponding equipment for each inspection item are pre-stored in the computer, which is used to obtain the location information of the corresponding equipment based on the electronic tag codes obtained by scanning; it is also used to process the collected substation images and determine the location of the substation. Fault type and fault location.
- the inspection drone subsystem includes a drone and inspection equipment;
- the inspection equipment includes a tilt aerial camera, an industrial infrared camera, and an industrial ultraviolet camera;
- the UAV is equipped with a rotary arm around it, and a motor holder is provided at the end of the UAV.
- a motor is installed in the motor holder, and a rotor is provided on the top power output shaft of the motor.
- a control unit is provided below the UAV. box, and landing gears are provided on both sides of the control box;
- the oblique aerial photography instrument is used to collect oblique photography data of the substation when performing drone aerial photography;
- the industrial infrared camera is used to collect infrared images of substations to determine whether there are electrical faults caused by heat in the corresponding equipment of each inspection item;
- the industrial UV camera is used to collect UV images of the substation to determine whether there are electrical faults caused by discharge in the corresponding equipment of each inspection item.
- the inspection robot subsystem includes: inspection robot, handheld collector, and lidar;
- the inspection robot is a four-wheeled mobile body, and the upper end of the mobile body is provided with a pan/tilt;
- the handheld collector is installed on the mobile vehicle body and is used to scan the electronic tag codes on the corresponding equipment of each inspection item to determine the location information of the equipment;
- the lidar is installed on the mobile vehicle body and is used to obtain three-dimensional point cloud data of each inspection item for three-dimensional reconstruction of the substation and planning of the drone flight route.
- the present invention provides an intelligent substation inspection system and method based on image recognition, which has the following beneficial effects:
- the present invention performs automatic inspections of substations based on inspection drones and inspection robots, which can obtain accurate substation inspection results, reduce labor use for manual inspections, reduce substation operation and maintenance costs, and improve inspection operations. and the level of automation and intelligence of management, providing more convenient and intelligent inspection methods for smart substations and unattended substations, making up for the shortcomings and deficiencies of manual inspections;
- the present invention performs three-dimensional reconstruction of the substation based on oblique photography data and three-dimensional point cloud data. On this basis, it plans the flight route of the drone and continuously checks and adjusts it to continuously find the optimal flight route during the automatic inspection process. , which can improve the efficiency of drone inspection work on the basis of ensuring that drones can inspect the entire area of the substation;
- the present invention also considers the impact of electromagnetic radiation in the substation on the flight of the drone, adjusts the status of the drone based on electromagnetic radiation data, and continuously improves the flight route planning of the drone to ensure that it flies in the correct direction. on the route;
- the present invention uses the industrial infrared camera and industrial ultraviolet camera mounted on the drone to determine whether there is a fault and the type of fault in the substation based on the three-channel image; the corresponding equipment of each inspection item is scanned through the handheld collector installed on the inspection robot.
- the electronic tag code can be used to obtain the location information of the device, and the inspection results of the accurate fault type and fault location can be obtained.
- Figure 1 is a flow chart of the intelligent substation inspection method based on image recognition
- Figure 2 is a schematic structural diagram of the intelligent substation inspection system based on image recognition.
- an embodiment of the present invention discloses an intelligent substation inspection method based on image recognition, as shown in Figure 1, including the following steps:
- the determined inspection items and the optimal inspection route arrange the codes of all inspection items, paste the codes on the corresponding equipment, and store the codes of the inspection items and the corresponding location information in the database;
- the inspection equipment is used to inspect each inspection item in sequence, and the images and electronic tag codes of each inspection item are obtained;
- automatic inspection of the substation based on inspection drones and inspection robots can obtain accurate inspection results, reduce the use of manual labor, reduce the operation and maintenance costs of the substation, and improve the intelligence of substation inspections. level of automation and automation. Compared with traditional manual inspections and robot inspections, drones can complete inspections in difficult environments and at night. They are less restricted by natural climate and terrain conditions, avoiding loopholes in inspections and saving money. Improve inspection efficiency while improving inspection time.
- the drone can fly quickly and stably along the planned route and hover at the designated location, it can also use the high-definition camera to take multi-angle shots and send back photos. After completing the flight and shooting tasks, it can return safely along the designated route.
- the drone needs to keep a safe distance from the equipment at all times during flight, and the route design needs to ensure the safe flight of the drone.
- drone aerial photography is carried out, and multi-angle oblique photography data of the substation is collected through the tilt aerial camera mounted on the drone;
- lidar Used lidar to inspect substations and obtain three-dimensional point cloud data for each inspection item
- the three-dimensional reconstruction of the substation by the drone is completed;
- the flight route of the drone is planned to control the drone to conduct automatic inspection of the substation.
- information such as the distance and angle of the photo points needs to be planned and set for different equipment in the substation, and on this basis, a smooth flight path connecting the photo points can be formed.
- the method also includes:
- Filtering divides dense laser points into ground points and non-ground points.
- the digital terrain model (DEM) of the transmission channel can be obtained by processing ground points through interpolation or mesh construction; classification
- the function of the method is to distinguish and extract data such as vegetation points and artificial feature points in non-ground points.
- planning the flight route of the UAV specifically includes the following steps:
- the UAV patrols according to the initial flight route Inspection is carried out, and the depth sensing camera mounted on the drone is used to extract information including three-dimensional position for depth sensing, to verify whether the flight route of the drone is optimal, so as to plan the optimal flight path of the drone during the automatic inspection process. Flight route.
- the drone After obtaining the initial flight route of the drone, the drone turns on the visual sense mode during flight and performs visual perception in real time.
- the UAV is equipped with two depth-sensing cameras.
- the two cameras extract information including three-dimensional positions for depth perception.
- the imaging points of the two cameras are fixed, and the visual difference of the two-channel videos imaged by the two cameras can be used. , accurately sense the distance between the current position and the target position, and accurately control the flight trajectory and shooting point location of the UAV, thereby ensuring the correctness of the UAV flight route, making the inspection work more reliable, and improving the quality of the inspection operation. and efficiency.
- the user manual of the drone basically has similar instructions: "The drone must be flown away from strong electromagnetic environments, otherwise it will interfere with the drone's compass or radio signal, causing the drone to enter attitude mode, causing the drone to Unable to hover or fly stably.” Facts have proved that interference definitely exists. Therefore, when planning the flight route of the drone, the following steps are also included:
- the corresponding position information when collecting images for each inspection item and the adjustment parameters of the collection equipment at the corresponding position are determined to improve the flight route planning of the UAV.
- Transformation equipment is in operation for a long time and is affected by environmental factors. Various faults often occur, sometimes manifested as overall or local abnormal heating. By studying the infrared image of the equipment, it can be determined whether the equipment is faulty. Based on this understanding, the detailed process of substation fault judgment in this embodiment is studied, which specifically includes the following steps:
- the electronic tag code on the corresponding device is scanned, and the scanned electronic tag code is uploaded to the computer through infrared communication;
- Substation images are acquired through drones and inspection robots and fused to generate three-channel status images; among them, visible light images, infrared images and ultraviolet images are acquired and then the channels are separated to obtain images of R channel, G channel and B channel respectively.
- information perform filtering enhancement processing on the image information in the three channels of R, G, and B, and fuse the image information of the three channels after filtering to obtain the fused image (the shooting angles of the three fused images should be consistent, so as to obtain more accurate information. Good confirmation of equipment failure);
- the substation fault type is judged; wherein the substation fault type includes: physical damage, electrical faults caused by heat, and electrical faults caused by discharge.
- substation fault type is judged, which specifically includes the following steps:
- the industrial UV camera mounted on the drone collects UV images of the substation to determine whether there is a corona area in the three-channel status image. If there is a fault caused by discharge, eliminate background interference and determine the location and intensity of the discharge point.
- the method further includes:
- the infrared image is preprocessed for filtering, noise reduction and de-bordering, using template matching technology (squared difference matching method, normalized squared difference matching method, correlation matching method, normalized correlation matching method, correlation coefficient matching method, Normalized correlation coefficient matching method) extracts the image of the corresponding equipment of each inspection item in the infrared image as the first image;
- the canny operator edge detection method is used to perform edge detection on the grayscale image to determine the location of equipment with electrical faults caused by heating.
- the method further includes:
- infrared detection is susceptible to the influence of infrared radiation in the environment, especially solar radiation. It generally cannot be performed during the day, and it usually detects late-stage faults and cannot detect early-stage failures. Issue an early warning.
- UV detection is not interfered by solar radiation and is integrated with visible light images and infrared images. Not only can the equipment be detected at any time throughout the day, but the advantages of UV, infrared and visible light detection can also be combined to improve fault detection. The detection efficiency can accurately determine the fault type and fault location, and obtain more accurate substation inspection results.
- an alarm device can also be set up.
- a sound or light alarm is used to remind the substation, and the obtained fault location information is sent to the maintenance personnel through wireless communication for the convenience of maintenance personnel. You can quickly determine the fault type and obtain the fault location to avoid wasting time and causing more serious consequences.
- the embodiment of the present invention discloses an intelligent substation inspection system based on image recognition, as shown in Figure 2, including: an inspection drone subsystem, an inspection robot subsystem, and a computer; the inspection drone subsystem, The inspection robot subsystems are wirelessly connected to the computer through wireless communication modules;
- the inspection drone subsystem is equipped with inspection equipment, which is used to inspect various inspection items determined according to the optimal inspection route of the drone, collect substation images and transmit them to the computer;
- the inspection robot subsystem is equipped with a handheld collector, which is used to scan the electronic label codes of the corresponding equipment of each inspection item and transmit them to the computer;
- the electronic tag codes and location information of the corresponding equipment for each inspection item are pre-stored in the computer, which is used to obtain the location information of the corresponding equipment based on the electronic tag code obtained by scanning; it is also used to process the collected substation images to determine the fault type of the substation. and fault location.
- the inspection drone subsystem includes a drone and inspection equipment;
- the inspection equipment includes a tilt aerial camera, an industrial infrared camera, and an industrial ultraviolet camera;
- the UAV is surrounded by a rotary arm, and the end of the rotary arm is provided with a motor holder.
- a motor is installed in the motor holder, and a rotor is provided on the top power output shaft of the motor.
- a control box is provided below the UAV. There are landing gears on both sides of the control box;
- the tilt aerial camera is used to collect oblique photography data of the substation during drone aerial photography
- Industrial infrared cameras are used to collect infrared images of substations to determine whether there are electrical faults caused by heat in the corresponding equipment of each inspection project;
- Industrial UV cameras are used to collect UV images of substations to determine whether there are electrical faults caused by discharge in the corresponding equipment of each inspection project.
- the inspection robot subsystem includes: an inspection robot, a handheld collector, and a laser radar;
- the inspection robot is a four-wheeled mobile body, and the upper end of the mobile body is equipped with a pan/tilt;
- the handheld collector is installed on the mobile vehicle body and is used to scan the electronic tag codes on the corresponding equipment of each inspection item to determine the location information of the equipment;
- the lidar is installed on the moving vehicle body and is used to obtain three-dimensional point cloud data of each inspection item for three-dimensional reconstruction of the substation and planning of the drone flight route.
- the intelligent substation inspection system is also equipped with an alarm device.
- the alarm device is connected to a computer.
- a sound or light alarm is used to remind the computer.
- the computer uses wireless communication to The obtained fault location information is sent to the maintenance personnel so that the maintenance personnel can quickly determine the fault type and obtain the fault location to avoid wasting time and causing more serious consequences.
- the present invention performs automatic inspection of substations based on inspection drones and inspection robots, which can obtain accurate substation inspection results, reduce the labor use of manual inspections, reduce the operation and maintenance costs of the substation, and improve the efficiency of inspection operations and management.
- the level of automation and intelligence provides a more convenient and intelligent inspection method for smart substations and unattended substations, making up for the defects and shortcomings of manual inspections; through the industrial infrared camera and industrial ultraviolet camera mounted on the drone, according to the three-channel
- the image determines whether there is a fault and the type of fault in the substation; through the handheld collector installed on the inspection robot, the electronic tag code of the corresponding equipment of each inspection item is scanned to obtain the location information of the equipment, and the accurate fault type and fault location can be obtained. Inspection results.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
一种基于图像识别的变电站智能巡检系统及方法。该方法包括：根据变电站的具体情况及客户的巡检需求，确定巡检项目以及巡检设备的最优巡检线路，编排巡检项目的编码，将编码粘贴在相应的设备上，并将巡检项目的编码及对应的位置信息存储到数据库中；基于确定的最优巡检线路，利用巡检设备依次对各个巡检项目进行巡检，获取各个巡检项目的图像及电子标签编码；从各个巡检项目的图像中提取信息，并与数据库中存储的信息进行对比，判断变电站的故障类型及故障位置。
Description
本发明涉及变电站运行维护技术领域，更具体的说是涉及一种基于图像识别的变电站智能巡检系统及方法。
变电站作为电网枢纽，它的运行质量与整个电网的安全、稳定运行息息相关，变电站设备的巡检工作是保证电力设备安全、提高电力设备可靠性、确保电力设备低故障率运行的一项基础工作。
然而，随着社会经济的发展，变电站的巡检工作越来越繁重，通常变电站地处偏远、地理条件恶劣，目前的变电站巡检主要依靠巡检人员定期定时进行人工巡检，由于受气候条件、环境因素、人员素质和责任心等多方面因素的制约，巡检质量和到位率无法保证。此外还存在以下问题：变电站设备型号繁多，巡检标准不齐全；缺陷分类标准不齐全、定义不准确；巡检结果数据整理和流转缺乏有效手段，设备隐患不能及时发现，引发设备故障；利用传统的巡检管理方法难以有效监督巡检人员，巡检不到位而引发的设备事故屡见不鲜。
巡检机器人的出现可以很好地避免人工巡检方式存在的问题，完成相应的巡检任务，但是目前，智能巡检机器人无法自主充电，在巡检过程中可能会偏离巡检路线；由于周围环境或者巡检机器人自身存在的不足，可能会导致拍摄的照片不清晰，无法获取准确的巡检结果，也就无法对变电站可能存在的故障进行准确判断。
因此，如何获取准确的变电站巡检结果，降低人工巡检的劳动强度，保证巡检质量，提高巡检作业和管理的自动化和智能化水平是本领域技术人员 亟需解决的问题。
发明内容
有鉴于此，本发明提供了一种基于图像识别的变电站智能巡检系统及方法，能够获取准确的变电站巡检结果，有效降低人工巡检的劳动强度，降低变电站运维成本，提高巡检作业和管理的自动化和智能化水平。
为了实现上述目的，本发明提供如下技术方案：
一种基于图像识别的变电站智能巡检方法，包括以下步骤：
根据变电站的具体情况及客户的巡检需求，确定巡检项目以及巡检设备的最优巡检线路；
根据确定的巡检项目及最优巡检线路，编排所述巡检项目的编码，将编码粘贴在相应的设备上，并将巡检项目的编码及对应的位置信息存储到数据库中；
基于确定的最优巡检线路，利用巡检设备依次对各个巡检项目进行巡检，获取各个巡检项目的图像及电子标签编码；
从各个巡检项目的图像中提取信息，并与数据库中存储的信息进行对比，判断变电站的故障类型及故障位置。
上述技术方案达到的技术效果为：该方案可以通过无人机及巡检机器人实现对变电站的自动巡检，获取准确的故障类型及故障位置信息，减少劳动力的使用，降低变电站的运维成本，提高巡检效率及智能化水平。
可选的，所述确定巡检设备的最优巡检线路，具体包括以下步骤：
针对变电站的地理环境，结合巡检需求进行无人机航摄，通过无人机搭载的倾斜航摄仪采集变电站的倾斜摄影数据；
利用激光雷达对变电站进行巡检，获取各个巡检项目的三维点云数据；之后，对三维点云数据进行降噪，去除因环境因素和仪器本身问题而导致的噪点，对降噪后的三维点云数据进行赋色，得到彩色的三维点云数据；
基于所述变电站的倾斜摄影数据和各个巡检项目的三维点云数据，以及无人机实时位置信息，完成无人机对变电站的三维重建；
基于三维重建的结果，规划无人机的飞行航线，以控制无人机对变电站进行自动巡检。
可选的，所述规划无人机的飞行航线，具体包括以下步骤：
基于三维重建的结果，选择无人机飞行的起点、终点及拍摄视点，将起点、所有拍摄视点及终点连接起来即获得无人机的初始飞行航线；
无人机按照所述初始飞行航线进行巡检，并利用无人机搭载的深度感知摄像头提取包括三维位置在内的信息进行深度感知，校验无人机的飞行航线是否最佳，以在自动巡检过程中规划无人机的最优飞行航线。
上述技术方案达到的技术效果为：基于倾斜航摄仪及激光雷达获取的数据进行变电站的三维重建，在此基础上规划无人机的飞行航线并不断进行校验调整，可以在自动巡检过程中不断确定最优路径，保证无人机能够巡检变电站的整个区域，提高无人机的巡检效率。
此外，在规划无人机的飞行航线时，还可以获取变电站内各个巡检项目的电磁辐射数据；根据电磁辐射数据，确定对每个巡检项目进行图像采集时所对应的位置信息和相应位置下的采集设备调整参数，以完善无人机的飞行航线规划，可以避免变电站的电磁环境影响，不断对无人机状态进行调整，保证其飞行航线的准确性。
可选的，所述判断变电站的故障类型及故障位置，具体包括以下步骤：
通过巡检机器人搭载的掌上式采集器对各个巡检项目进行巡检时，扫描相应设备上的电子标签编码，并通过红外线通信方式将扫描的电子标签编码上传至计算机中；
根据扫描的电子标签编码在数据库中进行查找确认，获取与所述电子标签编码对应的巡检项目的位置信息；
通过无人机及巡检机器人获取变电站图像并进行融合处理，生成三通道状态图像；
根据所述三通道状态图像，进行变电站故障类型的判断；其中，所述变电站故障类型包括：物理损坏、发热引起的电气故障、放电引起的电气故障。
可选的，所述进行变电站故障类型的判断，具体包括以下步骤：
根据所述三通道状态图像判断变电站中各个巡检项目的相关设备是否存在物理损坏，当存在物理损坏时，确定该设备的物理损坏位置；
通过无人机搭载的工业红外相机获取变电站的红外图像，判断所述三通道状态图像中是否存在异常发热点，若存在发热引起的电气故障，确定发热点位置及温度值；
通过无人机搭载的工业紫外相机采集变电站的紫外图像，判断所述三通道状态图像中是否存在电晕区域，若存在放电引起的故障，排除背景干扰，确定放电点位置及放电强度。
可选的，获取变电站的红外图像之后，所述方法还包括：
对所述红外图像进行滤波、降噪和去边框预处理，采用模板匹配技术提取出红外图像中各个巡检项目相应设备的图像，作为第一图像；
将所述第一图像转换为灰度图像，判断是否存在异常灰度值；若存在，判断所述异常灰度值对应的范围是否达到了相应设备尺寸程度，如果达到，则该设备存在发热引起的电气故障；
采用canny算子边缘检测法对所述灰度图像进行边缘检测，确定存在发热引起的电气故障的设备位置。
可选的，采集变电站的紫外图像之后，所述方法还包括：
对所述紫外图像进行增强处理，生成相应的增强图像；
从所述增强图像中提取带有紫外光斑的第二图像，并对所述第二图像进行处理，获得二值图像；
基于设备特征，从所述三通道状态图像中提取设备图像，并对所述设备图像和第二图像进行融合处理；
对所述二值图像进行边缘检测，根据边缘点坐标在融合后的图像中标注出放电区域，计算放电面积，确定放电位置及放电强度。
本发明还公开了一种基于图像识别的变电站智能巡检系统，包括：巡检无人机子系统、巡检机器人子系统、计算机；所述巡检无人机子系统、巡检机器人子系统均通过无线通信模块与计算机无线连接；
所述巡检无人机子系统搭载巡检设备，用于根据无人机的最优巡检线路对确定的各个巡检项目进行巡检，采集变电站图像并传输至计算机中；
所述巡检机器人子系统搭载掌上式采集器，用于扫描各个巡检项目相应设备的电子标签编码并传输至计算机中；
所述计算机中预先存储有各个巡检项目相应设备的电子标签编码及位置信息，用于根据扫描获得电子标签编码获取相应设备的位置信息；还用于对采集的变电站图像进行处理，判断变电站的故障类型及故障位置。
可选的，所述巡检无人机子系统包括无人机、巡检设备；所述巡检设备包括倾斜航摄仪、工业红外相机、工业紫外相机；
所述无人机四周设置有旋臂，旋臂的端部设置有电机固定座，电机固定座内安装有电动机，电动机的顶部动力输出轴上设置有旋翼；所述无人机下方设置有控制箱，控制箱两侧设置有起落架；
所述倾斜航摄仪，在进行无人机航摄时，用于采集变电站的倾斜摄影数据；
所述工业红外相机，用于采集变电站的红外图像，以判断各个巡检项目相应设备是否存在发热引起的电气故障；
所述工业紫外相机，用于采集变电站的紫外图像，以判断各个巡检项目相应设备是否存在放电引起的电气故障。
可选的，所述巡检机器人子系统包括：巡检机器人、掌上式采集器、激光雷达；
所述巡检机器人为四轮的移动车体，所述移动车体上端设置有云台；
所述掌上式采集器设置在移动车体上，用于扫描各个巡检项目相应设备上的电子标签编码，以确定该设备的位置信息；
所述激光雷达设置在移动车体上，用于获取各个巡检项目的三维点云数据，以进行变电站的三维重建及无人机飞行航线的规划。
经由上述的技术方案可知，与现有技术相比，本发明公开提供了一种基于图像识别的变电站智能巡检系统及方法，具有以下有益效果：
(1)本发明基于巡检无人机和巡检机器人对变电站进行自动巡检，能够获取准确的变电站巡检结果，减少人工巡检的劳动力使用，降低变电站的运维成本，提高巡检作业和管理的自动化和智能化水平，为智能变电站和无人值守变电站提供更加方便智能的巡检手段，弥补人工巡检存在的缺陷和不足；
(2)本发明基于倾斜摄影数据和三维点云数据对变电站进行三维重建，在此基础上规划无人机的飞行航线并不断校验调整，以在自动巡检过程中持续寻找最优飞行航线，可以在保证无人机能够巡检变电站全区域的基础上，提高无人机巡检工作的效率；
(3)本发明还考虑了变电站内的电磁辐射对无人机飞行的影响，基于电磁辐射数据对无人机的状态进行调整，不断完善无人机的飞行航线规划，以保证其飞行在正确的航线上；
(4)本发明通过无人机搭载的工业红外相机和工业紫外相机，根据三通道图像判断变电站是否存在故障及故障类型；通过巡检机器人上设置的掌上式采集器扫描各个巡检项目相应设备的电子标签编码，获取该设备的位置信息，可以获取准确的故障类型及故障位置的巡检结果。
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据提供的附图获得其他的附图。
图1为基于图像识别的变电站智能巡检方法的流程示意图；
图2为基于图像识别的变电站智能巡检系统的结构示意图。
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
实施例1
为满足经济发展的需要，作为国民经济发展的先行产业，电力行业投入了大量的人力、物力发展基础设施，变电站作为电力系统的重要节点，建设数量快速增加，其运行质量与整个电网的安全、稳定运行息息相关，定期的 巡视及日常消缺工作是维护变电站稳定运行的重要方式。巡检机器人可以代替人工巡检的大量工作，减少人为干扰因素，但因应用场景复杂、技术成熟度不足等多重因素，使巡检机器人在实际应用中也出现了一些问题，目前该技术还不够成熟，可靠性、安全性、智能化程度不够高。
为此，本发明实施例公开了一种基于图像识别的变电站智能巡检方法，如图1所示，包括以下步骤：
根据变电站的具体情况及客户的巡检需求，确定巡检项目以及巡检设备的最优巡检线路；
根据确定的巡检项目及最优巡检线路，编排所有巡检项目的编码，将编码粘贴在相应的设备上，并将巡检项目的编码及对应的位置信息存储到数据库中；
基于确定的最优巡检线路，利用巡检设备依次对各个巡检项目进行巡检，获取各个巡检项目的图像及电子标签编码；
从各个巡检项目的图像中提取信息，并与数据库中存储的信息进行对比，判断变电站的故障类型及故障位置。
在本实施例中，基于巡检无人机和巡检机器人对变电站进行自动巡检，可以获取准确的巡检结果，减少人工的劳动力使用，降低变电站的运维成本，提高变电站巡检的智能化和自动化水平。相比于传统的人工巡检及机器人巡检的方式，无人机能够完成在条件艰苦环境下及夜晚期间的巡检工作，受到自然气候、地形条件的限制小，避免巡检的漏洞，节省巡检时间的同时提高巡检效率。
接下来，对本实施例中技术方案的具体过程进行更详细的了解。
无人机不但能够快速、稳定地按照规划路线飞行并在指定的位置悬停，还能利用高清摄像头进行多角度拍摄并回传照片，完成飞行和拍摄任务后再 依照指定路线安全返航。无人机在飞行过程中需要时刻与设备保持安全距离，航线的设计需要确保实现无人机的安全飞行。
因此，在确定巡检设备的最优巡检线路时，具体包括以下步骤：
针对变电站的地理环境，结合巡检需求进行无人机航摄，通过无人机搭载的倾斜航摄仪采集变电站多角度的倾斜摄影数据；
利用激光雷达对变电站进行巡检，获取各个巡检项目的三维点云数据；
基于变电站的倾斜摄影数据和各个巡检项目的三维点云数据，以及无人机实时位置信息，完成无人机对变电站的三维重建；
基于三维重建的结果，规划无人机的飞行航线，以控制无人机对变电站进行自动巡检。其中，需要针对变电站中的不同设备规划设置拍照点的距离及角度等信息，在此基础上形成连接各拍照点平滑的飞行航迹。
进一步地，获取各个巡检项目的三维点云数据之后，所述方法还包括：
对三维点云数据进行降噪，去除因环境因素和仪器本身问题而导致的噪点；对降噪后的三维点云数据进行赋色，得到彩色的三维点云数据。目前处理点云数据最主要的方法是滤波和分类，滤波即将密集的激光点分为地面点和非地面点，通过插值或构网处理地面点可以得到输电通道的数字地形模型(DEM)；分类方法的作用是将非地面点中的植被点、人工地物点等数据进行区分和提取。
进一步地，所述规划无人机的飞行航线，具体包括以下步骤：
基于三维重建的结果，选择无人机飞行的起点、终点及拍摄视点，将起点、所有拍摄视点及终点连接起来即获得无人机的初始飞行航线；无人机按照所述初始飞行航线进行巡检，并利用无人机搭载的深度感知摄像头提取包括三维位置在内的信息进行深度感知，校验无人机的飞行航线是否最佳，以在自动巡检过程中规划无人机的最优飞行航线。
在获取无人机的初始飞行航线后，无人机在飞行中开启视觉感模式，实时进行视觉感知。在无人机上搭载两个深度感知摄像头，通过两个摄像机提取包括三维位置在内的信息进行深度感知，两个摄像头的成像点是固定的，可以根据两个摄像头成像的两路视频的视觉差，准确感知当前位置与目标位置的距离，对无人机的飞行轨迹及拍摄点位置进行精确控制，进而保证无人机飞行航线的正确性，使巡检工作更加可靠，提高巡检作业的质量和效率。
此外，无人机的使用说明书基本都有类似的说明“无人机一定要远离强电磁环境飞行，否则会干扰无人机的指南针或者无线电信号，造成无人机进入姿态模式，使得无人机无法稳定悬停或者飞行”，事实证明，干扰肯定是存在的，因此，在规划无人机的飞行航线时，还包括以下步骤：
获取变电站内各个巡检项目的电磁辐射数据；
根据所述电磁辐射数据，确定对每个巡检项目进行图像采集时所对应的位置信息和相应位置下的采集设备调整参数，以完善无人机的飞行航线规划。
变电设备长期处于运行状态且受到环境因素的影响，往往会出现各种各样的故障，有时表现为整体或局部的异常发热，通过研究设备的红外图像能够判断设备是否故障。在此点认知的基础上，研究本实施例中变电站故障判断的详细过程，具体包括以下步骤：
通过巡检机器人搭载的掌上式采集器对各个巡检项目进行巡检时，扫描相应设备上的电子标签编码，并通过红外线通信方式将扫描的电子标签编码上传至计算机中；
根据扫描的电子标签编码在数据库中进行查找确认，获取与所述电子标签编码对应的巡检项目的位置信息；
通过无人机及巡检机器人获取变电站图像并进行融合处理，生成三通道状态图像；其中，获取可见光图像、红外图像和紫外图像后进行通道分离，分别得到R通道、G通道、B通道的图像信息，对R、G、B三种通道下的图 像信息作滤波增强处理，并在滤波后将三通道的图像信息融合，得到融合图像(进行融合的三种图像的拍摄角度应该一致，以便更好的确认设备故障)；
根据所述三通道状态图像，进行变电站故障类型的判断；其中，所述变电站故障类型包括：物理损坏、发热引起的电气故障、放电引起的电气故障。
进一步地，进行变电站故障类型的判断，具体包括以下步骤：
根据三通道状态图像判断变电站中各个巡检项目的相关设备是否存在物理损坏，当存在物理损坏时，确定该设备的物理损坏位置；
通过无人机搭载的工业红外相机获取变电站的红外图像，判断三通道状态图像中是否存在异常发热点，若存在发热引起的电气故障，确定发热点位置及温度值；
通过无人机搭载的工业紫外相机采集变电站的紫外图像，判断三通道状态图像中是否存在电晕区域，若存在放电引起的故障，排除背景干扰，确定放电点位置及放电强度。
进一步地，获取变电站的红外图像之后，所述方法还包括：
对所述红外图像进行滤波、降噪和去边框预处理，采用模板匹配技术(平方差匹配法、归一化平方差匹配法、相关匹配法、归一化相关匹配法、相关系数匹配法、归一化相关系数匹配法)提取出红外图像中各个巡检项目相应设备的图像，作为第一图像；
将所述第一图像转换为灰度图像，判断是否存在异常灰度值；若存在，判断所述异常灰度值对应的范围是否达到了相应设备尺寸程度，如果达到，则该设备存在发热引起的电气故障；
采用canny算子边缘检测法对所述灰度图像进行边缘检测，确定存在发热引起的电气故障的设备位置。
进一步地，采集变电站的紫外图像之后，所述方法还包括：
对所述紫外图像进行增强处理，生成相应的增强图像；
从所述增强图像中提取带有紫外光斑的第二图像，并对所述第二图像进行处理，获得二值图像；
基于设备特征，从所述三通道状态图像中提取设备图像，并对所述设备图像和第二图像进行融合处理；
对所述二值图像进行边缘检测，根据边缘点坐标在融合后的图像中标注出放电区域，计算放电面积，确定放电位置及放电强度。
此外，红外图像与可见光图像有一定的区别，具有自身的特点，红外检测易受环境中红外辐射尤其是太阳辐射的影响，一般不能在白天进行，且其通常检测后期故障，无法对前期故障做出预警。而在本实施例中，紫外检测不受太阳辐射干扰，与可见光图像、红外图像融合，不仅可以在全天任何时间段对设备进行检测，还能结合紫外、红外、可见光检测的优势，提高故障检测的效率，对故障类型及故障位置进行准确判断，获取更加精确的变电站巡检结果。
在更进一步的实施例中，还可以设置报警装置，当变电站存在故障时，通过声音或灯光报警的方式进行提醒，并通过无线通信的方式将获取的故障位置信息发送给检修人员，以便检修人员可以快速判断故障类型以及获取故障位置，避免浪费时间，造成更严重的后果。
实施例2
本发明实施例公开了一种基于图像识别的变电站智能巡检系统，如图2所示，包括：巡检无人机子系统、巡检机器人子系统、计算机；所述巡检无人机子系统、巡检机器人子系统均通过无线通信模块与计算机无线连接；
巡检无人机子系统搭载巡检设备，用于根据无人机的最优巡检线路对确定的各个巡检项目进行巡检，采集变电站图像并传输至计算机中；
巡检机器人子系统搭载掌上式采集器，用于扫描各个巡检项目相应设备的电子标签编码并传输至计算机中；
计算机中预先存储有各个巡检项目相应设备的电子标签编码及位置信息，用于根据扫描获得电子标签编码获取相应设备的位置信息；还用于对采集的变电站图像进行处理，判断变电站的故障类型及故障位置。
进一步地，所述巡检无人机子系统包括无人机、巡检设备；所述巡检设备包括倾斜航摄仪、工业红外相机、工业紫外相机；
无人机四周设置有旋臂，旋臂的端部设置有电机固定座，电机固定座内安装有电动机，电动机的顶部动力输出轴上设置有旋翼；所述无人机下方设置有控制箱，控制箱两侧设置有起落架；
倾斜航摄仪，在进行无人机航摄时，用于采集变电站的倾斜摄影数据；
工业红外相机，用于采集变电站的红外图像，以判断各个巡检项目相应设备是否存在发热引起的电气故障；
工业紫外相机，用于采集变电站的紫外图像，以判断各个巡检项目相应设备是否存在放电引起的电气故障。
进一步地，所述巡检机器人子系统包括：巡检机器人、掌上式采集器、激光雷达；
巡检机器人为四轮的移动车体，所述移动车体上端设置有云台；
掌上式采集器设置在移动车体上，用于扫描各个巡检项目相应设备上的电子标签编码，以确定该设备的位置信息；
激光雷达设置在移动车体上，用于获取各个巡检项目的三维点云数据，以进行变电站的三维重建及无人机飞行航线的规划。
在更进一步的实施例中，所述变电站智能巡检系统还设置有报警装置，所述报警装置与计算机相连，当变电站存在故障时，通过声音或灯光报警的方式进行提醒，计算机通过无线通信的方式将获取的故障位置信息发送给检修人员，以便检修人员可以快速判断故障类型以及获取故障位置，避免浪费时间，造成更严重的后果。
本发明基于巡检无人机和巡检机器人对变电站进行自动巡检，能够获取准确的变电站巡检结果，减少人工巡检的劳动力使用，降低变电站的运维成本，提高巡检作业和管理的自动化和智能化水平，为智能变电站和无人值守变电站提供更加方便智能的巡检手段，弥补人工巡检存在的缺陷和不足；通过无人机搭载的工业红外相机和工业紫外相机，根据三通道图像判断变电站是否存在故障及故障类型；通过巡检机器人上设置的掌上式采集器扫描各个巡检项目相应设备的电子标签编码，获取该设备的位置信息，可以获取准确的故障类型及故障位置的巡检结果。
本说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似部分互相参见即可。对于实施例公开的系统而言，由于其与实施例公开的方法相对应，所以描述的比较简单，相关之处参见方法部分说明即可。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本发明。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况下，在其它实施例中实现。因此，本发明将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。
Claims (10)
- 一种基于图像识别的变电站智能巡检方法，其特征在于，包括以下步骤：根据变电站的具体情况及客户的巡检需求，确定巡检项目以及巡检设备的最优巡检线路；根据确定的巡检项目及最优巡检线路，编排所述巡检项目的编码，将编码粘贴在相应的设备上，并将巡检项目的编码及对应的位置信息存储到数据库中；基于确定的最优巡检线路，利用巡检设备依次对各个巡检项目进行巡检，获取各个巡检项目的图像及电子标签编码；从各个巡检项目的图像中提取信息，并与数据库中存储的信息进行对比，判断变电站的故障类型及故障位置。
- 根据权利要求1所述的一种基于图像识别的变电站智能巡检方法，其特征在于，所述确定巡检设备的最优巡检线路，具体包括以下步骤：针对变电站的地理环境，结合巡检需求进行无人机航摄，通过无人机搭载的倾斜航摄仪采集变电站的倾斜摄影数据；利用激光雷达对变电站进行巡检，获取各个巡检项目的三维点云数据；基于所述变电站的倾斜摄影数据和各个巡检项目的三维点云数据，以及无人机实时位置信息，完成无人机对变电站的三维重建；基于三维重建的结果，规划无人机的飞行航线，以控制无人机对变电站进行自动巡检。
- 根据权利要求2所述的一种基于图像识别的变电站智能巡检方法，其特征在于，所述规划无人机的飞行航线，具体包括以下步骤：基于三维重建的结果，选择无人机飞行的起点、终点及拍摄视点，将起点、所有拍摄视点及终点连接起来即获得无人机的初始飞行航线；无人机按照所述初始飞行航线进行巡检，并利用无人机搭载的深度感知摄像头提取包括三维位置在内的信息进行深度感知，校验无人机的飞行航线是否最佳，以在自动巡检过程中规划无人机的最优飞行航线。
- 根据权利要求1所述的一种基于图像识别的变电站智能巡检方法，其特征在于，所述判断变电站的故障类型及故障位置，具体包括以下步骤：通过巡检机器人搭载的掌上式采集器对各个巡检项目进行巡检时，扫描相应设备上的电子标签编码，并通过红外线通信方式将扫描的电子标签编码上传至计算机中；根据扫描的电子标签编码在数据库中进行查找确认，获取与所述电子标签编码对应的巡检项目的位置信息；通过无人机及巡检机器人获取变电站图像并进行融合处理，生成三通道状态图像；根据所述三通道状态图像，进行变电站故障类型的判断；其中，所述变电站故障类型包括：物理损坏、发热引起的电气故障、放电引起的电气故障。
- 根据权利要求4所述的一种基于图像识别的变电站智能巡检方法，其特征在于，所述进行变电站故障类型的判断，具体包括以下步骤：根据所述三通道状态图像判断变电站中各个巡检项目的相关设备是否存在物理损坏，当存在物理损坏时，确定该设备的物理损坏位置；通过无人机搭载的工业红外相机获取变电站的红外图像，判断所述三通道状态图像中是否存在异常发热点，若存在发热引起的电气故障，确定发热点位置及温度值；通过无人机搭载的工业紫外相机采集变电站的紫外图像，判断所述三通道状态图像中是否存在电晕区域，若存在放电引起的故障，排除背景干扰，确定放电点位置及放电强度。
- 根据权利要求5所述的一种基于图像识别的变电站智能巡检方法，其特征在于，获取变电站的红外图像之后，所述方法还包括：对所述红外图像进行滤波、降噪和去边框预处理，采用模板匹配技术提取出红外图像中各个巡检项目相应设备的图像，作为第一图像；将所述第一图像转换为灰度图像，判断是否存在异常灰度值；若存在，判断所述异常灰度值对应的范围是否达到了相应设备尺寸程度，如果达到，则该设备存在发热引起的电气故障；采用canny算子边缘检测法对所述灰度图像进行边缘检测，确定存在发热引起的电气故障的设备位置。
- 根据权利要求5所述的一种基于图像识别的变电站智能巡检方法，其特征在于，采集变电站的紫外图像之后，所述方法还包括：对所述紫外图像进行增强处理，生成相应的增强图像；从所述增强图像中提取带有紫外光斑的第二图像，并对所述第二图像进行处理，获得二值图像；基于设备特征，从所述三通道状态图像中提取设备图像，并对所述设备图像和第二图像进行融合处理；对所述二值图像进行边缘检测，根据边缘点坐标在融合后的图像中标注出放电区域，计算放电面积，确定放电位置及放电强度。
- 一种基于图像识别的变电站智能巡检系统，其特征在于，包括：巡检无人机子系统、巡检机器人子系统、计算机；所述巡检无人机子系统、巡检机器人子系统均通过无线通信模块与计算机无线连接；所述巡检无人机子系统搭载巡检设备，用于根据无人机的最优巡检线路对确定的各个巡检项目进行巡检，采集变电站图像并传输至计算机中；所述巡检机器人子系统搭载掌上式采集器，用于扫描各个巡检项目相应设备的电子标签编码并传输至计算机中；所述计算机中预先存储有各个巡检项目相应设备的电子标签编码及位置信息，用于根据扫描获得电子标签编码获取相应设备的位置信息；还用于对采集的变电站图像进行处理，判断变电站的故障类型及故障位置。
- 根据权利要求8所述的一种基于图像识别的变电站智能巡检系统，其特征在于，所述巡检无人机子系统包括无人机、巡检设备；所述巡检设备包括倾斜航摄仪、工业红外相机、工业紫外相机；所述无人机四周设置有旋臂，旋臂的端部设置有电机固定座，电机固定座内安装有电动机，电动机的顶部动力输出轴上设置有旋翼；所述无人机下方设置有控制箱，控制箱两侧设置有起落架；所述倾斜航摄仪，在进行无人机航摄时，用于采集变电站的倾斜摄影数据；所述工业红外相机，用于采集变电站的红外图像，以判断各个巡检项目相应设备是否存在发热引起的电气故障；所述工业紫外相机，用于采集变电站的紫外图像，以判断各个巡检项目相应设备是否存在放电引起的电气故障。
- 根据权利要求8所述的一种基于图像识别的变电站智能巡检系统，其特征在于，所述巡检机器人子系统包括：巡检机器人、掌上式采集器、激光雷达；所述巡检机器人为四轮的移动车体，所述移动车体上端设置有云台；所述掌上式采集器设置在移动车体上，用于扫描各个巡检项目相应设备上的电子标签编码，以确定该设备的位置信息；所述激光雷达设置在移动车体上，用于获取各个巡检项目的三维点云数据，以进行变电站的三维重建及无人机飞行航线的规划。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2022/115106 WO2024040566A1 (zh) | 2022-08-26 | 2022-08-26 | 基于图像识别的变电站智能巡检系统及方法 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2022/115106 WO2024040566A1 (zh) | 2022-08-26 | 2022-08-26 | 基于图像识别的变电站智能巡检系统及方法 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2024040566A1 true WO2024040566A1 (zh) | 2024-02-29 |
Family
ID=90012158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2022/115106 WO2024040566A1 (zh) | 2022-08-26 | 2022-08-26 | 基于图像识别的变电站智能巡检系统及方法 |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2024040566A1 (zh) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117875946A (zh) * | 2024-03-11 | 2024-04-12 | 国网安徽省电力有限公司合肥供电公司 | 一种用于变电站设备运维的人机协同自主红外巡检方法 |
| CN118075314A (zh) * | 2024-04-18 | 2024-05-24 | 五凌电力有限公司 | 发电企业智能治安安全监控方法及系统 |
| CN118171470A (zh) * | 2024-03-18 | 2024-06-11 | 广西电网有限责任公司 | 一种用于电力调度的多源数据融合智能巡检系统及方法 |
| CN118259690A (zh) * | 2024-05-30 | 2024-06-28 | 江西科晨洪兴信息技术有限公司 | 一种变电站无人机自主巡视航线优化方法及系统 |
| CN118330383A (zh) * | 2024-04-10 | 2024-07-12 | 国网上海市电力公司 | 综合能源系统多智能体协同的设备故障检测系统及方法 |
| CN118623893A (zh) * | 2024-08-09 | 2024-09-10 | 国网江西省电力有限公司南昌供电分公司 | 一种变电站无人机巡视路径规划方法及系统 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6834259B1 (en) * | 1999-10-15 | 2004-12-21 | Timekeeping Systems, Inc. | Guard tour system |
| JP2010146398A (ja) * | 2008-12-19 | 2010-07-01 | Chugoku Electric Power Co Inc:The | 巡視支援システム、巡視支援方法およびプログラム |
| KR101400169B1 (ko) * | 2014-02-06 | 2014-05-28 | (주)라이드소프트 | 방범 관제를 위한 가상현실 기법을 이용한 시각적 순찰 시스템 및 그 방법 |
| CN112884931A (zh) * | 2019-11-29 | 2021-06-01 | 众芯汉创（北京）科技有限公司 | 一种用于变电站的无人机巡检方法及系统 |
| CN113569714A (zh) * | 2021-07-24 | 2021-10-29 | 国网河南省电力公司检修公司 | 基于无人机巡检的变电站智能图像识别系统 |
| CN113625748A (zh) * | 2021-07-27 | 2021-11-09 | 国家电网有限公司 | 一种变电站无人机巡检航线规划方法 |
| CN114636424A (zh) * | 2019-02-21 | 2022-06-17 | 国网浙江平湖市供电有限公司 | 基于可穿戴设备的变电站巡检路径规划方法 |
-
2022
- 2022-08-26 WO PCT/CN2022/115106 patent/WO2024040566A1/zh unknown
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6834259B1 (en) * | 1999-10-15 | 2004-12-21 | Timekeeping Systems, Inc. | Guard tour system |
| JP2010146398A (ja) * | 2008-12-19 | 2010-07-01 | Chugoku Electric Power Co Inc:The | 巡視支援システム、巡視支援方法およびプログラム |
| KR101400169B1 (ko) * | 2014-02-06 | 2014-05-28 | (주)라이드소프트 | 방범 관제를 위한 가상현실 기법을 이용한 시각적 순찰 시스템 및 그 방법 |
| CN114636424A (zh) * | 2019-02-21 | 2022-06-17 | 国网浙江平湖市供电有限公司 | 基于可穿戴设备的变电站巡检路径规划方法 |
| CN112884931A (zh) * | 2019-11-29 | 2021-06-01 | 众芯汉创（北京）科技有限公司 | 一种用于变电站的无人机巡检方法及系统 |
| CN113569714A (zh) * | 2021-07-24 | 2021-10-29 | 国网河南省电力公司检修公司 | 基于无人机巡检的变电站智能图像识别系统 |
| CN113625748A (zh) * | 2021-07-27 | 2021-11-09 | 国家电网有限公司 | 一种变电站无人机巡检航线规划方法 |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117875946A (zh) * | 2024-03-11 | 2024-04-12 | 国网安徽省电力有限公司合肥供电公司 | 一种用于变电站设备运维的人机协同自主红外巡检方法 |
| CN117875946B (zh) * | 2024-03-11 | 2024-06-04 | 国网安徽省电力有限公司合肥供电公司 | 一种用于变电站设备运维的人机协同自主红外巡检方法 |
| CN118171470A (zh) * | 2024-03-18 | 2024-06-11 | 广西电网有限责任公司 | 一种用于电力调度的多源数据融合智能巡检系统及方法 |
| CN118330383A (zh) * | 2024-04-10 | 2024-07-12 | 国网上海市电力公司 | 综合能源系统多智能体协同的设备故障检测系统及方法 |
| CN118075314A (zh) * | 2024-04-18 | 2024-05-24 | 五凌电力有限公司 | 发电企业智能治安安全监控方法及系统 |
| CN118259690A (zh) * | 2024-05-30 | 2024-06-28 | 江西科晨洪兴信息技术有限公司 | 一种变电站无人机自主巡视航线优化方法及系统 |
| CN118623893A (zh) * | 2024-08-09 | 2024-09-10 | 国网江西省电力有限公司南昌供电分公司 | 一种变电站无人机巡视路径规划方法及系统 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2024040566A1 (zh) | 基于图像识别的变电站智能巡检系统及方法 | |
| CN112633535B (zh) | 一种基于无人机图像的光伏电站智能巡检方法及系统 | |
| CN111583337B (zh) | 一种基于多传感器融合的全方位障碍物检测方法 | |
| US11731762B2 (en) | Crisscross boustrophedonic flight patterns for UAV scanning and imaging | |
| CN110297498B (zh) | 一种基于无线充电无人机的轨道巡检方法及系统 | |
| CN111611855B (zh) | 一种变电站三维可视化机器人智能巡检系统 | |
| CN110850723B (zh) | 一种基于变电站巡检机器人系统的故障诊断及定位方法 | |
| CN104091168B (zh) | 基于无人机影像的电力线自动提取定位方法 | |
| CN108255189A (zh) | 一种电力巡检无人机系统 | |
| CN111198004A (zh) | 一种基于无人机的电力巡查信息采集系统 | |
| CN110888457A (zh) | 利用无人机和机器人对变电设备开展立体巡视系统及方法 | |
| CN112528979B (zh) | 变电站巡检机器人障碍物判别方法及系统 | |
| CN114240868A (zh) | 一种基于无人机的巡检分析系统及方法 | |
| CN207691309U (zh) | 一种用于输电线路大型无人机进行巡检工作的装置 | |
| CN107564111A (zh) | 基于计算机视觉的电力线空间安全分析方法 | |
| CN111244822B (zh) | 一种复杂地理环境的固定翼无人机巡线方法、系统和装置 | |
| CN110186375A (zh) | 智能化高铁白车身焊接装配特征检测装置及检测方法 | |
| CN115995058A (zh) | 基于人工智能的输电通道安全在线监测方法 | |
| CN116245844A (zh) | 基于视觉多模态融合的配电线路缺陷智能识别方法及系统 | |
| Fang et al. | A framework of power pylon detection for UAV-based power line inspection | |
| CN112801432A (zh) | 一种风机组叶片智能巡检系统及风机组叶片巡检方法 | |
| CN115355952B (zh) | 一种原油储罐智能巡检系统 | |
| Sha et al. | Design of patrol monitoring and control system for hot spot of solar photovoltaic module | |
| Zhou et al. | Three dimensional fully autonomous inspection method for wind power employing unmanned aerial vehicle based on 5G wireless communication and artificial intelligence | |
| CN115755946A (zh) | 一种基于数字孪生的电力巡检方法、装置及系统 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22956112 Country of ref document: EP Kind code of ref document: A1 |