WO2022037278A1 - 基于人工智能的变电站巡检机器人系统 - Google Patents

基于人工智能的变电站巡检机器人系统 Download PDF

Info

Publication number
WO2022037278A1
WO2022037278A1 PCT/CN2021/103956 CN2021103956W WO2022037278A1 WO 2022037278 A1 WO2022037278 A1 WO 2022037278A1 CN 2021103956 W CN2021103956 W CN 2021103956W WO 2022037278 A1 WO2022037278 A1 WO 2022037278A1
Authority
WO
WIPO (PCT)
Prior art keywords
robot
artificial intelligence
module
substation inspection
controller
Prior art date
Application number
PCT/CN2021/103956
Other languages
English (en)
French (fr)
Inventor
朱明增
孙春日
覃剑永
岑建军
刘荣洲
张炜
黎华
梁兆庭
周承秀
覃秋勤
刘小兰
周虹妤
胡凯博
黄新华
杨波
谢辉
张展声
吕鸣
胡清智
朱敦森
丘浩
林翔宇
莫梓樱
陈少暖
黄承伟
陈琴
蒙亮
李永栈
杨芳
韦妙香
陈极万
罗小波
Original Assignee
广西电网有限责任公司贺州供电局
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 广西电网有限责任公司贺州供电局 filed Critical 广西电网有限责任公司贺州供电局
Publication of WO2022037278A1 publication Critical patent/WO2022037278A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02B—BOARDS, SUBSTATIONS OR SWITCHING ARRANGEMENTS FOR THE SUPPLY OR DISTRIBUTION OF ELECTRIC POWER
    • H02B3/00—Apparatus specially adapted for the manufacture, assembly, or maintenance of boards or switchgear
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network

Definitions

  • the invention relates to the technical field of robot design, in particular to an artificial intelligence-based substation inspection robot system.
  • the traditional single inspection method of substations is manual inspection, which has problems such as high labor intensity, low inspection efficiency, inadequate inspection, inconsistent inspection standards, and difficulty in inspection in harsh environments such as rain and snow. And simply relying on the senses and experience of the inspectors, it is difficult to make an objective, comprehensive and accurate judgment, and lay a potential safety hazard for the safe operation of the equipment.
  • the existing inspection systems at home and abroad are mainly integrated through video, access control and security-related applications.
  • Video standards have become mature, but each standard is not compatible with each other, and related compatible protocol equipment is basically concentrated in video.
  • the technical problem solved by the invention is to overcome the defects of the prior art and provide an artificial intelligence-based substation inspection robot that realizes intelligent perception, intelligent real-time transmission, intelligent diagnosis and analysis, and intelligent remote real-time sharing of substation equipment status data. system.
  • a substation inspection robot system based on artificial intelligence including a robot and a service platform, the service platform issues a cruise command to the robot and receives relevant signals sent by the robot;
  • the robot includes a main controller, a laser navigator, a motion controller, an information
  • the acquisition module and the information transmission module the service platform includes a signal-connected server, an Ethernet network, a wireless bridge and a WIFI antenna, the WIFI antenna is signal-connected to the main robot controller, the information acquisition module is electrically connected to the information transmission module, and the information transmission module is used for The obtained signal is fed back to the main controller, and the main controller respectively controls the laser navigator, the motion controller, the information acquisition module and the information transmission module and receives signal feedback.
  • the laser navigator includes a mapping module and a motor-driven wheeled module.
  • the motor-driven wheeled module automatically runs on a fixed path by combining the motion control coding signals and IMU signals provided by the motion controller.
  • the site generates a picture of the environment.
  • the information collection module includes one or more of an image perception camera, a temperature diagnostic thermal imager, a gas sensor, a sound detection sensor, and a temperature and humidity sensor.
  • the robot is also equipped with a pose control system, which can intelligently perceive the device status through an intelligent mobile chassis.
  • the robot is also provided with sound and light alarms.
  • the robot also has an interactive real-time intercom platform.
  • the substation inspection robot system also includes a charging pile that provides instant charging services for the robot.
  • the charging pile includes a charging room door, a charging pile controller, a charger and WIFI or Ethernet, and the charging room door is controlled by the charging pile controller. After receiving WIFI or Ethernet related signals, it is controlled to turn on or off, and the WIFI antenna of the service platform is connected to the WIFI or Ethernet signal of the charging pile.
  • the present invention has the following beneficial effects:
  • It can carry out functions such as infrared temperature measurement, meter reading, switching actuator identification and abnormal state alarm for the substation equipment in the station, and provide background functions such as real-time upload of inspection data, data analysis, information display and automatic report generation. It has the advantages of high inspection efficiency, and can carry out inspection in harsh environments such as rain and snow, effectively improving the reliability of substation equipment operation, and at the same time realizing intelligent perception of substation equipment status data, intelligent real-time transmission, intelligent diagnosis and analysis, and intelligent remote real-time. purpose of sharing.
  • FIG. 1 is a topology diagram of the artificial intelligence-based substation inspection robot system according to Embodiment 1;
  • FIG. 2 is an electrical diagram of the robot, service platform and charging pile described in Embodiment 1.
  • FIG. 2 is an electrical diagram of the robot, service platform and charging pile described in Embodiment 1.
  • an artificial intelligence-based substation inspection robot system including a robot and a service platform.
  • the service platform issues a cruise command to the robot and receives relevant signals from the robot at the same time;
  • the robot includes a main controller. , laser navigator, motion controller, information acquisition module and information transmission module
  • the service platform includes signal connection server, Ethernet, wireless bridge and WIFI antenna, WIFI antenna and robot main controller signal connection, information acquisition module and information
  • the transmission module is electrically connected, and the information transmission module is used to feed back the obtained signal to the main controller, and the main controller respectively controls the laser navigator, the motion controller, the information acquisition module and the information transmission module and receives signal feedback.
  • the robot can replace the manual monitoring and judgment of the operating status of the equipment in a special environment, and realize the full coverage of the equipment area.
  • the laser navigator includes a mapping module and a motor-driven wheeled module.
  • the motor-driven wheeled module combines the motion provided by the motion controller. Control the coded signal and IMU signal to run automatically with a fixed path, and the mapping module is used for the robot to generate environmental pictures in unfamiliar places.
  • the robot can start the inspection and inspection work regularly according to the daily planned inspection and inspection tasks; at the same time, it can perform automatic inspections in sequence along the predetermined trajectory according to the positions of the preset inspection points.
  • the robot also supports fixed-point or fixed-task inspection. Just select the inspection point you want to inspect through the service platform, and dispatch a temporary inspection task to the robot. The robot will plan the optimal path to complete the inspection according to the selected task content. Task.
  • the laser navigator can realize trackless navigation. It does not need to pre-embed magnetic tracks or other forms of tracks to realize the autonomous walking of the inspection robot.
  • the laser navigator is mainly guided by positioning technology and navigation control technology.
  • the laser navigator also includes a laser radar positioning module, which determines the position of the robot in the global coordinate system through a laser sensor.
  • Laser positioning is a positioning method based on artificial road signs. Matching positioning method; when the robot enters the inspection environment for the first time, it will scan the surrounding environment through the lidar, generate the environment map through the synchronous map construction and positioning (SLAM) algorithm, and in the subsequent inspection process, the laser will scan the surrounding environment in real time.
  • the terrain is precisely matched to the environmental terrain to determine the precise position of the robot.
  • the robot can travel smoothly along the passage in the substation. If it needs to return to the original road in special circumstances, the robot can travel on non-hardened roads such as gravel and grass, and it will not affect the navigation, positioning and shooting when traveling on non-hardened roads.
  • the robot can also turn around on the spot.
  • the drive unit of the wheeled inspection robot is equipped with a four-wheel drive chassis and adopts differential steering technology, which can realize no-radius turning on the spot.
  • the information collection module includes one or more of image perception cameras, temperature diagnostic thermal imagers, gas sensors, sound detection sensors, and temperature and humidity sensors.
  • image perception camera is a high-definition camera, which collects on-site images in real time, and transmits them in real time through the wireless network system, and sets up network relays to ensure that the video is not lost or distorted, and can monitor equipment and the environment in the largest range.
  • the thermal imager for temperature diagnosis is an infrared thermal imaging sensor. It adopts morphological processing algorithms, intelligent trend analysis and lateral comparison diagnosis to realize intelligent identification of equipment temperature defects.
  • the temperature identification process is to first collect infrared images.
  • wavelet algorithm denoising processing and intensity normalization processing are required to preprocess the infrared image in advance.
  • the device shape filtering is performed to extract the edge of the device, and the device and the outside world are separated from the image and other morphological processing.
  • Image processing Then calculate the maximum temperature correlation, optimize the temperature image recognition, identify the data for defect diagnosis and analysis, and give an alarm in time if there is an abnormality.
  • the gas sensor includes a toxic and harmful gas detection integrated module (CH4, CO, O3, O2, SF6, etc.), and the sensing module is selected according to the gas environment; it can also be equipped with an ultraviolet corona detector for equipment partial discharge detection and other related functions expansion.
  • the equipment status data collected by the robot is intelligently transmitted through the wireless network system.
  • the service platform uses accurate algorithms to realize intelligent diagnosis and analysis of equipment perception data, and determine equipment failures.
  • the service platform can also analyze and output defect reports for equipment defects, for decision-making.
  • the layer provides the decision-making basis for the health status of the equipment, and the service platform can also cooperate with subsystem modules such as access control, security, ventilation, etc. to realize all-round local automatic monitoring and remote monitoring of the substation.
  • the service platform can realize functions such as video storage, distribution, system configuration, real-time monitoring, information release, video query, video playback, real-time control, image capture, and recording, and truly realize the intelligent management of video surveillance of intelligent inspection robots.
  • the service platform conducts targeted training for the collected image samples. Due to the influence of external factors such as the orientation, location, and illumination of the field equipment, the processing system performs preprocessing such as grayscale, binarization, and correction, so that the sample images are in the best state of processing effect.
  • preprocessing different algorithms are used to extract classification features for different equipment image objects (pointer class, liquid level stroke class, splitting class, text class, etc.).
  • feature dimension reduction processing is required. , reduce the feature dimension and improve the feature processing speed. After facing the more difficult feature extraction, perform classification and identification, identify the data results, and output in real time.
  • the robot system of this embodiment relies on an intelligent mobile platform to realize intelligent control of robot movement, communication, power supply, integration of various sensors, etc.; combined with laser navigation algorithms, realizes intelligent robot mobile navigation, precise positioning, and performs various inspection tasks.
  • Some of the relevant parameters of the robot are selected as follows:
  • Robot Master controller: Intel Core i5-6500; Can bus: Can 2.01Mbit/s; 6-DOF inertial unit: MPU6500A; Vehicle switch: 10/100BaseT(x) automatic detection, full/half duplex, MDI/ MDI-X self-adaptive; air switch: Schneider A9N22408 (60V DC32A); maximum driving speed: 1m/s; climbing ability: 20°; wading depth: 100mm; obstacle crossing ability: 50mm; IP protection level: IP65; repeat Navigation positioning error: ⁇ 10mm; turning radius: turn in place.
  • Thermal imager for temperature diagnosis Detector pixel: 640*512; Spectral range: 7.5 ⁇ 13.5 ⁇ m; Thermal sensitivity: ⁇ 40mK; Temperature measurement range: -40°C ⁇ +550°C; Low temperature mode: -40°C ⁇ 160°C; High temperature mode: -40°C to 550°C; temperature measurement accuracy: Max ( ⁇ 2°C, ⁇ 2%); protection grade: IP66; working environment: -40°C to +70°C.
  • the robot is also equipped with a pose control system, which can intelligently perceive the status of the equipment through an intelligent mobile chassis, and realize accurate perception of status data such as equipment meters, infrared, and appearance.
  • the robot is also equipped with sound and light alarms for alarming emergency accidents.
  • Emergency accidents in substations mainly refer to production accidents that seriously threaten the safety of substation equipment and property and the personal safety of inspection personnel due to large-scale power outages, fires, and explosions.
  • the control center finds the fault signal in the substation, it will quickly notify the operation and maintenance personnel to check it out.
  • the main station can use the robot as a striker after receiving the notification of the accident information, go deep into the scene at the first time, and use the mobile high-definition video to assist the operation and maintenance personnel to observe the actual condition of the equipment.
  • the robot infrared thermal imager can quickly find the fault point through the smoke, and timely feedback the accident information.
  • the use of robots to transmit on-site information in real time is conducive to the formation of a joint analysis team of remote network experts, strengthening the real-time analysis of equipment difficulties and defects, and facilitating operation and maintenance personnel to quickly understand the on-site situation and grasp on-site dynamics from a distance.
  • the robot also has equipment such as visible light cameras, and can upload the collected video to the monitoring background.
  • the robot itself has the function of visible light image acquisition.
  • the staff can operate the monitoring system to move to the designated position through the software of the host computer, and control the free rotation of the PTZ, which can realize the close observation of the target object and cover the monitoring range to the blind spot.
  • the robot can collect video images of the surrounding environment, and transmit the image information back to the main control room through the wireless network according to the needs of the staff, and automatically record the shooting location and equipment name, so that the staff can query the complete information in the future.
  • the robot is also equipped with ultrasonic sensors and has the function of obstacle detection. If an obstacle is encountered during the walking process, it should stop in time and give an alarm. After the obstacle is removed within the specified time, it can resume walking. If the obstacle is not removed within the specified time, reasonable measures can be taken. avoidance strategy.
  • the robot also has an interactive real-time intercom platform for the operator to interact and feedback information with the robot in real time.
  • the substation inspection robot system also includes a charging pile that provides instant charging services for the robot.
  • the charging pile includes a charging room door, a charging pile controller, a charger and WIFI or Ethernet.
  • the charging room door receives the WIFI or Ethernet from the charging pile controller.
  • the WIFI antenna of the service platform is connected to the WIFI or Ethernet signal of the charging pile;
  • WIFI or Ethernet receives the door opening signal sent by the service platform, and the charging pile controller opens the charging room door, and the robot is charging After the door is opened, navigate to the charging pile.
  • the charging pile judges that the robot is docked through the detection signal, turn on the charger to complete the automatic charging.
  • the robot uses 220V AC power supply, the battery of the robot takes 2 hours to fully charge, and the number of charge and discharge is 1500 times. It is convenient to select the relevant location on the spot and arrange the charging pile.
  • the robot monitors the power status in real time. If the power is too low, it will automatically trigger the return-to-home charging task and give a low-power alarm; when it returns to the charging position, Execute the autonomous charging task and confirm whether the charging is successful. If the charging is not successful, repeat the charging task, and continue to perform the inspection task after the charging is completed; if the battery is sufficient during the execution of the inspection task, continue to perform the inspection task until the inspection task is completed. Finish.

Abstract

一种基于人工智能的变电站巡检机器人系统，包括机器人和服务平台，服务平台下发巡航指令给机器人同时接收机器人传送来的相关信号；机器人包括主控制器、激光导航器、运动控制器、信息采集模块和信息传输模块，服务平台包括信号连接的服务器、以太网、无线网桥和WIFI天线，WIFI天线与机器人主控制器信号连接，信息采集模块与信息传输模块电连接，信息传输模块用于将得到的信号反馈给主控制器，主控制器分别对激光导航器、运动控制器、信息采集模块、信息传输模块进行指令控制和接收信号反馈。提升变电站设备运行可靠性，实现变电站设备状态数据智能感知、智能实时传输、智能诊断分析、智能远程实时共享。

Description

基于人工智能的变电站巡检机器人系统 技术领域
本发明涉及机器人设计技术领域，具体地，涉及一种基于人工智能的变电站巡检机器人系统。
背景技术
变电站传统的单一巡检方式为人工巡检，存在劳动强度大、巡检效率低、巡检不到位、巡检标准不统一、雨雪恶劣环境下巡视困难等问题。并且简单依靠巡检人员的感官和经验，很难做到客观、全面、准确的评判，给设备的安全运行埋下安全隐患。目前，国内外现有的巡检系统主要通过视频、门禁及安防相关应用集成较多，视频标准已经趋于成熟化，但是各个标准之间无法互相兼容，相关的兼容协议设备基本也集中于视频、门禁及安防相关应用，无法应用于电力系统中视频设备与物联网设备应用的集成；现有的智能分析及运维管理主要针对平安城市应用较多，其他行业应用多在平安城市应用基础上。因此，针对电力系统机器视觉和听觉能力的电网智能音视频管控关键技术研究也基本是空白，相关研究较少。
近年来，随着科技进步，以“信息化、数字化、自动化、互动化”为特征的智能电网和特高压建设的逐渐深入，传统的运检模式已不满足现代电网的发展需求，如何采用新形势下的新技术、新手段实现运维模式的转变，减轻变电站运维人员的运维负担，有效保障设备安全可靠运行，是今后所面临的主要问题之一。
发明内容
本发明解决的技术问题在于克服现有技术的缺陷，提供一种实现变电站设备状态数据智能化感知、智能化实时传输、智能化诊断分析、智能化远程实时共享的基于人工智能的变电站巡检机器人系统。
本发明的目的通过以下技术方案实现：
一种基于人工智能的变电站巡检机器人系统，包括机器人和服务平台，服务平台下发巡航指令给机器人同时接收机器人传送来的相关信号；机器人包括主控制器、激光导航器、运动控制器、信息采集模块和信息传输模块，服务平台包括信号连接的服务器、以太网、无线网桥和WIFI天线，WIFI天线与机器人主控制器信号连接，信息采集模块与信息传输模块电连接，信息传输模块用于将得到的信号反馈给主控制器， 主控制器分别对激光导航器、运动控制器、信息采集模块、信息传输模块进行指令控制和接收信号反馈。
进一步地，激光导航器包括建图模块和电机驱动轮式模块，电机驱动轮式模块通过结合运动控制器提供的运动控制编码信号、IMU信号进行固定路径自动运行，建图模块用于机器人在陌生场所生成环境图片。
更进一步地，信息采集模块包括图像感知摄像机、温度诊断热像仪、气体传感器、声音检测传感器、温湿度传感器中的一种或多种。
再进一步地，机器人还搭载有位姿控制系统，通过智能移动底盘进行设备状态智能感知。
进一步地，机器人还设有声光报警器。
更进一步地，机器人还设有交互式实时对讲平台。
再进一步地，变电站巡检机器人系统还包括为机器人提供即时充电服务的充电桩，充电桩包括充电房门、充电桩控制器、充电器和WIFI或以太网，充电房门由充电桩控制器在接收WIFI或以太网相关信号后控制打开或关闭，服务平台WIFI天线与充电桩的WIFI或以太网信号连接。
与现有技术相比，本发明具有以下有益效果：
可对站内变电设备开展红外测温、表计读数、分合执行机构识别及异常状态报警等功能，并提供巡检数据的实时上传和数据分析、信息显示和报表自动生成等后台功能，具有巡检效率高、雨雪等恶劣环境下可开展巡检等优点，有效提升变电站设备运行可靠性，同时实现变电站设备状态数据智能化感知、智能化实时传输、智能化诊断分析、智能化远程实时共享的目的。
附图说明
图1为实施例1所述的基于人工智能的变电站巡检机器人系统的拓扑图；
图2为实施例1所述的机器人、服务平台和充电桩的电气图。
具体实施方式
下面结合具体实施方式对本发明作进一步的说明，其中，附图仅用于示例性说明，表示的仅是示意图，而非实物图，不能理解为对本专利的限制；为了更好地说明本发明的实施例，附图某些部件会有省略、放大或缩小，并不代表实际产品的尺寸；对本领域技术人员来说，附图中某些公知结构及其说明可能省略是可以理解的。
实施例1
如图1和图2所示，提供一种基于人工智能的变电站巡检机器人系统，包括机器人和服务平台，服务平台下发巡航指令给机器人同时接收机器人传送来的相关信号；机器人包括主控制器、激光导航器、运动控制器、信息采集模块和信息传输模块，服务平台包括信号连接的服务器、以太网、无线网桥和WIFI天线，WIFI天线与机器人主控制器信号连接，信息采集模块与信息传输模块电连接，信息传输模块用于将得到的信号反馈给主控制器，主控制器分别对激光导航器、运动控制器、信息采集模块、信息传输模块进行指令控制和接收信号反馈。
机器人可代替人工进行特殊环境下设备运行状态的监测判断，实现设备区域全覆盖巡视，其中激光导航器包括建图模块和电机驱动轮式模块，电机驱动轮式模块通过结合运动控制器提供的运动控制编码信号、IMU信号进行固定路径自动运行，建图模块用于机器人在陌生场所生成环境图片。机器人可按照每日规划的巡视检测任务，定时开始巡视检测工作；同时可根据预先设定的巡检点的位置，沿着预定轨迹依次进行自动巡检。
机器人也支持定点或定任务巡检，只需通过服务平台选择想要进行巡视的巡检点，向机器人派发临时巡检任务，机器人就会按照选择的任务内容，规划出最优路径完成巡检任务。
激光导航器可实现导航无轨化，不需要事先预埋磁轨或其他形式的轨道来实现巡检机器人自主行走，激光导航器主要由定位技术和导航控制技术进行指导。激光导航器还包括激光雷达定位模块，其是通过激光传感器确定机器人在全局坐标系下的位置，激光定位是一种基于人工路标的定位方式，巡检车激光定位是一种基于激光数据与地图匹配的定位方式；机器人首次进入巡检环境时，将通过激光雷达对周边环境进行扫描，通过同步地图构建与定位(SLAM)算法生成环境地图，并在后续巡检过程中，将激光实时扫描的地形与环境地形进行精确匹配，从而确定机器人精确位置。
机器人在变电站内可沿通道畅行，如果碰到特殊情况需要原路返回，机器人可在砂石、草地等非硬化路面行进，在非硬化路面行进时不影响导航定位和拍摄。机器人也可原地掉头，轮式巡检机器人的驱动单元配备了四轮驱动底盘，采用差速转向技术，可实现原地无半径转弯。
信息采集模块包括图像感知摄像机、温度诊断热像仪、气体传感器、声音检测传 感器、温湿度传感器中的一种或多种，通过移动监测的方式，实现变电站信息监控的全覆盖、全检测。图像感知摄像机为高清摄像机，实时采集现场图像，并经过无线网络系统进行实时传输，并设置网络中继保证视频不丢包，不失真，最大范围进行设备及环境监控。温度诊断热像仪为红外热成像传感器，采用形态处理算法以及智能趋势分析和横向对别诊断，实现设备温度缺陷智能识别，温度识别过程为首先采集红外图像，由于设备外界温度、风速、光照等噪声环境需要进行小波算法去噪处理和强度归一处理等方式进行红外图像提前预处理，预处理后再进行设备形态滤波，将设备边缘进行提取，设备与外界进行图像分离等形态处理，图像处理后进行温度最大相关量计算，最优化进行温度图像识别，识别数据进行缺陷诊断分析，如出现异常及时进行告警。气体传感器包括有毒有害气体检测集成模块(CH4、CO、O3、O2、SF6等)，根据气体环境进行感知模块选择；也可搭载紫外电晕检测仪进行设备局放检测等相关功能拓展。
机器人采集的设备状态数据经过无线网络系统进行智能传输，服务平台通过精准算法，实现设备感知数据智能诊断分析，判断设备故障情况，服务平台还可针对设备缺陷情况进行缺陷情况报表分析输出，为决策层提供设备健康状态决策依据，服务平台还可与门禁、安防、通风等子系统模块，实现对变电站所全方位本地自动监控和远程监控。
服务平台可实现视频存储、分发、系统配置、实时监视、信息发布、视频查询、视频回放、实时控制、抓图、抓录等功能，真正实现智能巡检机器人视频监控智能化管理。服务平台针对采集的图像样本进行针对性训练，由于现场设备朝向、位置、光照等外界因素影响，处理系统进行灰度化、二值化、纠正等预处理，使得样本图像处于处理效果最佳状态，预处理完毕后，针对不同设备图像对象(指针类、液位行程类、分合类、文字类等)，采用不同算法进行分类特征提取，为保证训练特征稳定性，需要进行特征降维处理，降低特征维度，提升特征处理速度，面对较为困难的特征提取后，进行分类识别，识别数据结果，实时输出。
本实施例机器人系统的依托智能移动平台，实现智能控制机器人移动、通信、供电、多种传感器集成等；结合激光导航算法，实现智能机器人移动导航、精准定位，执行多种巡检任务。其中机器人的部分相关参数选择如下：
机器人：主控器：Intel Core i5-6500；Can总线：Can2.01Mbit/s；6自由度 惯性单元：MPU6500A；车载交换机：10/100BaseT(x)自动侦测，全/半双工，MDI/MDI-X自适应；空开：施耐德A9N22408(60V DC32A)；最大行驶速度：1m/s；爬坡能力：20°；涉水深度：100mm；越障能力：50mm；IP防护等级：IP65；重复导航定位误差为：±10mm；转弯半径：原地转弯。
温度诊断热像仪：探测器像素：640*512；光谱范围：7.5～13.5μm；热灵敏度：≤40mK；测温范围：-40℃～+550℃；低温模式：-40℃到160℃；高温模式：-40℃到550℃；测温精度：Max(±2℃,±2％)；防护等级：IP66；工作环境：-40℃～+70℃。
图像感知摄像机：图像传感器：240万像素、1/1.9"CMOS；分辨率：1920*1080；编码格式：H.264；光学变倍：30倍；视场角：H*V＝61.2°*34.2°(近焦)～2.32°*1.27°(远焦)；最低照度：彩色：0.001Lux@(F1.5，AGC ON)；防护等级：IP66；工作环境：-40℃～+70℃。
机器人还搭载有位姿控制系统，通过智能移动底盘进行设备状态智能感知，实现设备表计、红外、外观等状态数据精准感知。
机器人还设有声光报警器用于对应急事故进行报警，变电站中应急事故主要是指：设备发生大面积停电、起火、爆炸等严重威胁变电站设备财产安全与巡检人员人身安全的生产事故。通常情况下，当调控中心发现变电站内故障信号后，将快速通知运维人员前往进行查看，运维人员与抢险人员难以进入现场，及时反馈现场信息。使用机器人后，主站在收到事故信息通知后可以利用机器人作为前锋，第一时间深入现场，利用移动高清视频协助运维人员观察设备实际状况。若存在大量烟雾难以看清设备时，可以通过机器人红外热像仪透过烟雾快速发现故障点，及时反馈事故信息。利用机器人实时传送现场信息，有利于组建远程网络专家联合分析队伍，加强设备疑难缺陷分析的实时性，便于运维人员在远方快速了解现场情况、掌握现场动态。通过机器人排查其他设备，快速掌握设备故障情况及时确定处理方案，从而保障运维人员人身安全。
机器人还具有可见光摄像机等设备，并能将所采集的视频上传至监控后台。机器人自身具有可见光图像采集功能，工作人员可通过上位机软件操作监控系统移动到指定位置，控制云台自由转动，可实现近距离地观察拍摄目标物体，将监控范围覆盖到盲区。机器人可可对周围环境进行视频图像的采集，并根据工作人员的需要将图像信息经无线网络传回主控室，并自动记录拍摄地点和设备名称，以备工作人员日后能查询完整的信息。
机器人还搭载超声波传感器，具有障碍物检测功能，在行走过程中如遇到障碍物应及时停止并报警，指定时间内障碍物移除后能恢复行走，指定时间障碍物未移除，可采取合理的避让策略。
机器人还设有交互式实时对讲平台，以备操作人员实时与机器人进行信息交互反馈。
变电站巡检机器人系统还包括为机器人提供即时充电服务的充电桩，充电桩包括充电房门、充电桩控制器、充电器和WIFI或以太网，充电房门由充电桩控制器在接收WIFI或以太网相关信号后控制打开或关闭，服务平台WIFI天线与充电桩的WIFI或以太网信号连接；WIFI或以太网接收服务平台发送过来的开门信号，由充电桩控制器打开充电房门，机器人在充电房门打开后，导航至充电桩上，充电桩通过检测信号判断机器人对接完成后，打开充电器完成自动充电。机器人采用220V交流供电，机器人的电池完全充电时间为2h，充放电次数为1500次。在现场进行选择相关便于取点位置，布置充电桩，在执行任务过程中，机器人实时监测电量情况，如出现电量过低情况，自主触发返航充电任务，并电量过低告警；返航至充电位置，执行自主充电任务，并确认是否充电成功，如未充电成功，重复执行充电任务，充电完成继续执行巡检任务；如果巡检任务执行过程中，电量充足，继续执行巡检任务，直至巡检任务完成。
显然，上述实施例仅仅是为清楚地说明本发明的技术方案所作的举例，而并非是对本发明的实施方式的限定。对于所属领域的普通技术人员来说，在上述说明的基础上还可以做出其它不同形式的变化或变动。这里无需也无法对所有的实施方式予以穷举。凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明权利要求的保护范围之内。

Claims (9)

  1. 一种基于人工智能的变电站巡检机器人系统，其特征在于，包括机器人和服务平台，服务平台下发巡航指令给机器人同时接收机器人传送来的相关信号；机器人包括主控制器、激光导航器、运动控制器、信息采集模块和信息传输模块，服务平台包括信号连接的服务器、以太网、无线网桥和WIFI天线，WIFI天线与机器人主控制器信号连接，信息采集模块与信息传输模块电连接，信息传输模块用于将得到的信号反馈给主控制器，主控制器分别对激光导航器、运动控制器、信息采集模块、信息传输模块进行指令控制和接收信号反馈。
  2. 根据权利要求1所述的基于人工智能的变电站巡检机器人系统，其特征在于，激光导航器包括建图模块和电机驱动轮式模块，电机驱动轮式模块通过结合运动控制器提供的运动控制编码信号、IMU信号进行固定路径自动运行，建图模块用于机器人在陌生场所生成环境图片。
  3. 根据权利要求2所述的基于人工智能的变电站巡检机器人系统，其特征在于，信息采集模块包括图像感知摄像机、温度诊断热像仪、气体传感器、声音检测传感器、温湿度传感器中的一种或多种。
  4. 根据权利要求3所述的基于人工智能的变电站巡检机器人系统，其特征在于，机器人还搭载有位姿控制系统，通过智能移动底盘进行设备状态智能感知。
  5. 根据权利要求1所述的基于人工智能的变电站巡检机器人系统，其特征在于，机器人还设有声光报警器。
  6. 根据权利要求5所述的基于人工智能的变电站巡检机器人系统，其特征在于，机器人还设有交互式实时对讲平台。
  7. 根据权利要求5所述的基于人工智能的变电站巡检机器人系统，其特征在于，机器人还设有超声停障系统。
  8. 根据权利要求1所述的基于人工智能的变电站巡检机器人系统，其特征在于，机器人为轮式巡检机器人，驱动单元配备四轮驱动底盘。
  9. 根据权利要求1～6任意一项所述的基于人工智能的变电站巡检机器人系统，其特征在于，变电站巡检机器人系统还包括为机器人提供即时充电服务的充电桩，充电桩包括充电房门、充电桩控制器、充电器和WIFI或以太网，充电房门由充电桩控制器在接收WIFI或以太网相关信号后控制打开或关闭，服务平台WIFI天线与充电桩的WIFI或以太网信号连接。
PCT/CN2021/103956 2020-08-19 2021-07-01 基于人工智能的变电站巡检机器人系统 WO2022037278A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010836929.3 2020-08-19
CN202010836929 2020-08-19

Publications (1)

Publication Number Publication Date
WO2022037278A1 true WO2022037278A1 (zh) 2022-02-24

Family

ID=80322524

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/103956 WO2022037278A1 (zh) 2020-08-19 2021-07-01 基于人工智能的变电站巡检机器人系统

Country Status (1)

Country Link
WO (1) WO2022037278A1 (zh)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114560087A (zh) * 2022-03-09 2022-05-31 国网河北省电力有限公司石家庄供电分公司 一种基于无人机的城市电网电力智能运检设备
CN114603563A (zh) * 2022-04-25 2022-06-10 国网湖南省电力有限公司 一种连续体机械臂的多丝杆电机总线型嵌入式控制系统
CN114627570A (zh) * 2022-04-08 2022-06-14 南京悦能智能科技有限公司 一种四足仿生机器人在电力巡检的应用系统
CN114872574A (zh) * 2022-07-01 2022-08-09 惠州市丝鹭新能源科技有限公司 基于充电桩的采集与监控系统及方法
CN114915645A (zh) * 2022-05-13 2022-08-16 徐州克瑞艾特电子科技有限公司 一种充电桩安全智能监控系统
CN114943408A (zh) * 2022-04-06 2022-08-26 深圳市弗赛特科技股份有限公司 一种加氢站安全隐患智能监测方法、系统
CN115200570A (zh) * 2022-09-15 2022-10-18 国网山东省电力公司费县供电公司 一种电网巡视用导航设备及其导航方法
CN115205997A (zh) * 2022-07-14 2022-10-18 西安建筑科技大学 一种基于人工智能的暖通空调机房无人化巡检系统及方法
CN115604317A (zh) * 2022-10-21 2023-01-13 北京珞安科技有限责任公司（Cn） 一种工业日常巡查的提示系统及方法
CN116260840A (zh) * 2023-02-02 2023-06-13 北京英诺格林科技有限公司 一种水处理设备的智能化巡检预警维修系统
CN116388034A (zh) * 2023-03-30 2023-07-04 江苏泽宇智能电力股份有限公司 一种随检显像的变电站远程巡检系统
CN116988837A (zh) * 2023-09-25 2023-11-03 太原科技大学 一种煤矿井下自主巡检系统及方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102082466A (zh) * 2010-10-15 2011-06-01 重庆市电力公司超高压局 变电站设备智能巡检机器人系统
CN102280826A (zh) * 2011-07-30 2011-12-14 山东鲁能智能技术有限公司 变电站智能机器人巡检系统及巡检方法
CN104914865A (zh) * 2015-05-29 2015-09-16 国网山东省电力公司电力科学研究院 变电站巡检机器人定位导航系统及方法
US20170136631A1 (en) * 2014-01-17 2017-05-18 Knightscope, Inc. Autonomous data machines and systems
CN109599945A (zh) * 2018-11-30 2019-04-09 武汉大学 一种智慧电厂自主巡检机器人巡检系统及方法
CN110614639A (zh) * 2019-09-19 2019-12-27 国网山东省电力公司电力科学研究院 一种基于ros的变电站巡检机器人系统及其方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102082466A (zh) * 2010-10-15 2011-06-01 重庆市电力公司超高压局 变电站设备智能巡检机器人系统
CN102280826A (zh) * 2011-07-30 2011-12-14 山东鲁能智能技术有限公司 变电站智能机器人巡检系统及巡检方法
US20170136631A1 (en) * 2014-01-17 2017-05-18 Knightscope, Inc. Autonomous data machines and systems
CN104914865A (zh) * 2015-05-29 2015-09-16 国网山东省电力公司电力科学研究院 变电站巡检机器人定位导航系统及方法
CN109599945A (zh) * 2018-11-30 2019-04-09 武汉大学 一种智慧电厂自主巡检机器人巡检系统及方法
CN110614639A (zh) * 2019-09-19 2019-12-27 国网山东省电力公司电力科学研究院 一种基于ros的变电站巡检机器人系统及其方法

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114560087A (zh) * 2022-03-09 2022-05-31 国网河北省电力有限公司石家庄供电分公司 一种基于无人机的城市电网电力智能运检设备
CN114943408A (zh) * 2022-04-06 2022-08-26 深圳市弗赛特科技股份有限公司 一种加氢站安全隐患智能监测方法、系统
CN114627570A (zh) * 2022-04-08 2022-06-14 南京悦能智能科技有限公司 一种四足仿生机器人在电力巡检的应用系统
CN114603563A (zh) * 2022-04-25 2022-06-10 国网湖南省电力有限公司 一种连续体机械臂的多丝杆电机总线型嵌入式控制系统
CN114915645A (zh) * 2022-05-13 2022-08-16 徐州克瑞艾特电子科技有限公司 一种充电桩安全智能监控系统
CN114872574B (zh) * 2022-07-01 2022-09-23 惠州市丝鹭新能源科技有限公司 基于充电桩的采集与监控系统及方法
CN114872574A (zh) * 2022-07-01 2022-08-09 惠州市丝鹭新能源科技有限公司 基于充电桩的采集与监控系统及方法
CN115205997A (zh) * 2022-07-14 2022-10-18 西安建筑科技大学 一种基于人工智能的暖通空调机房无人化巡检系统及方法
CN115200570A (zh) * 2022-09-15 2022-10-18 国网山东省电力公司费县供电公司 一种电网巡视用导航设备及其导航方法
CN115604317A (zh) * 2022-10-21 2023-01-13 北京珞安科技有限责任公司（Cn） 一种工业日常巡查的提示系统及方法
CN116260840A (zh) * 2023-02-02 2023-06-13 北京英诺格林科技有限公司 一种水处理设备的智能化巡检预警维修系统
CN116388034A (zh) * 2023-03-30 2023-07-04 江苏泽宇智能电力股份有限公司 一种随检显像的变电站远程巡检系统
CN116388034B (zh) * 2023-03-30 2024-03-08 江苏泽宇智能电力股份有限公司 一种随检显像的变电站远程巡检系统
CN116988837A (zh) * 2023-09-25 2023-11-03 太原科技大学 一种煤矿井下自主巡检系统及方法
CN116988837B (zh) * 2023-09-25 2024-04-05 太原科技大学 一种煤矿井下自主巡检系统及方法

Similar Documents

Publication Publication Date Title
WO2022037278A1 (zh) 基于人工智能的变电站巡检机器人系统
CN108922188B (zh) 雷达跟踪定位的四维实景交通路况感知预警监控管理系统
CN105812733B (zh) 一种民航空中交通管制的场面监视引导系统
CN108491758A (zh) 一种轨道检测方法及机器人
CN206194076U (zh) 一种变电站设备检测系统
CN103279949B (zh) 基于自定位机器人的多相机参数自动标定系统运行方法
CN103235562B (zh) 变电站基于巡检机器人的综合参数检测系统及巡检方法
CN205029436U (zh) 一种集控式变电站内智能巡检机器人系统
CN206628891U (zh) 变电站设备监控系统
CN108957240A (zh) 电网故障远程定位方法及系统
CN106451201A (zh) 一种变电站巡检系统
CN102097860A (zh) 一种变电站安全检测的智能机器人巡检系统
CN102593739A (zh) 磁导航巡检机器人智能控制系统
CN109818416A (zh) 一种多功能变电站智能机器人巡检系统
CN105989682A (zh) 一种输电线路线下施工机械安全预警监控系统及监控方法
CN114115296B (zh) 一种重点区域智能巡检与预警系统及方法
CN103455036A (zh) 一种场景空中巡视方法和飞行器
CN111753780B (zh) 变电站违章检测系统及违章检测方法
WO2022242759A1 (zh) 应用于海上升压站的无人智能巡检系统及方法
CN113144470A (zh) 一种消防应急预警处理灭火一体化控制系统
CN114859972A (zh) 空中无人机与地面巡检机器人协同作业的巡检系统及方法
CN113075686B (zh) 一种基于多传感器融合的电缆沟智能巡检机器人建图方法
CN105382837A (zh) 变电站无盲区智能巡检机器人的监测方法及其监测机构
CN202564812U (zh) 磁导航巡检机器人智能控制系统
CN201975857U (zh) 一种变电站安全检测的智能机器人巡检系统

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: 21857374

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21857374

Country of ref document: EP

Kind code of ref document: A1